ML18033B712

instruments

per trip system.

The test

makes

Note I.B for

Table

3.2.B

applicable

and

the

system

or

component

must

be

inoperable.

This action contradicts

other

TS requirements

to have

CS

(3.5.A)

and

RHR Systems

(3.5.B) operable with fuel in the vessel

at

greater

than atmospheric

pressures.

The licensee's

position is that

since the required action (to be in cold shutdown) for the inoperable

CS and

RHR has

been met, the test

can

be performed under existing TS.

The inspector

did not agree with this practice.

The

TS should

be

clarified or testing

performed without fuel in the vessel.

The

NRC

did not preapprove this test

as mentioned in the

LRED.

This item is

identified as

URI 259,260,296/91-10-01,

TS Requirements

During Check

Valve Testing.

b.

Standby

Gas Treatment

Room Flooding

During

a routine tour of the plant

on March 28,

1991, the inspector

observed

water

running across

an entrance

roadway to the

radwaste

building.

The inspector

followed the water to the

SBGT building

entrance

and

observed

water

running

out

underneath

the

entrance

door.

The inspector

entered

the

SBGT building

and notified the

control

room of the problem.

Operations

and health physics

personnel

responded

to the

scene.

The area

was

found to be not contaminated.

The source of the water

was

determined

to

be water overflowing from

a

sump.

The

sump

has

a

keep fill system to maintain water in the

sump to minimize potential airborne contamination

problems.

The keep

fill valve

had stuck open.

The operator

banged

on

a controller that

caused

the fill valve to shut

and the

sump

pump to start.

The

SBGT building is toured

by plant personnel

once

on each eight

hour shift.

The inspector

obtained

a printout of the

room access

from security

and

determined

that the

room

had

been

entered

four

hours earlier.

Therefore,

the water spillage

had apparently occurred

during the previous four hours.

The licensee initiated

an incident

investigation for the controller fai lure.

The inspector

concluded

that the plant tours were being conducted

and that the plant response

to this was proper.

Failure to Implement

TS Required

Compensatory

Measures

On March 1,

1991,

TVA determined

that

gaseous

and liquid effluent

compensatory

measures

were not performed

as required

by TS 3.2.D and

3.2.K for inoperable

monitors.

TS Section 3.2.D

and Table 3.2.D,

which includes

RCW monitor

2-RM-90-132, require that grab samples

be

collected

and analyzed at least

once per 8 hours9.259259e-5 days 0.00222 hours 1.322751e-5 weeks 3.044e-6 months

when the

RCW monitor

is inoperable

and effluent releases

are

in progress.

TS Section 3.2.K

and

Table

3.2.K,

which

includes

monitors

1-RM-90-250,

2-RM-90-250,

3-RM-90-250,

and O-RM-90-252, require that

a flow rate

estimate

be

taken

at least

once

per

four

hours

whenever

the

instruments

are declared

inoperable

and effluent releases

are being

conducted

through

an affected

pathway.

On October

31,

1990, all four CAMs on the refuel floor (1-RN-90-250,

2-RM-90-250,

3-RM-90-250,

and

0-RM-90-252)

were declared

inoperable

and

removed

from service for replacement.

Compensatory

monitoring

was established

per O-SI-4.8.B.l.a.2,

Airborne Effluent Release

Rate

By Manual

Sampling

When

A Gaseous

Effluent Monitor Is Inoperable.

Compensatory

monitor ing

continued

until

the

modifications

were

completed

and the

CANs declared

on January

18,

1991.

On

December

30,

1990,

the

Chemistry

section

became

aware

of

discrepancies

between

compensatory

SI data entries

and refuel floor

security access

logs.

Specifically, the sample flow data required

by

TS to be obtained

every four hours

was signed off as being obtained

by

a

RLA who, according to the security refuel floor entry log, was

not on the refuel floor where the

CANs are physically located at the

time

the flow was

recorded.

Chemistry

conducted

a preliminary

investigation

which indicated

possible

falsification

and

gA was

requested

to

perform

a

more

in-depth

investigation.

The

gA

investigation

identified

18

discrepancies

between

Chemistry

compensatory

SI data

sheets

and security logs within a

7 day period.

Of the

18 discrepancies,

17

involved

gaseous

sample

flow rate

measurements

and/or samples,

and one involved

a liquid sample.

On January

30,

1991,

gA issued

a

CARR listing the

18 discrepancies.

The licensee

found supporting

evidence

to corroborate

performance of

16 of the discrepancies.

The remaining

two discrepancies

were the

gaseous

effluent flow rate

measurement

required

by TS 3.2.K for the

four

monitors

in

the

reactor/turbine

building

and

radwaste

ventilation systems

on December

5, 1990, at 4:00 a.m.,

and the liquid

sample for the

RCW system monitor required

by TS 3.2.D

on December

11,

1990, at 10:03 a.m.

Although the

RLA had

documented

that work

activities

were

completed,

security

keycard

data

could not support

that the

RLA was in the sampling areas

at the time of sampling.

The

licensee

concluded that the

RLA did not conduct the two required

SI

activities and, therefore,

the requirements

of TS Sections 3.2.D and

3.2.K were not met.

The licensee

reported this information to the

NRC via the

ENS

on March

1,

1991,

and

submitted

LER 259/91-03

on

April 1,

1991.

The licensee

conducted

an incident investigation (II-B-91-045) which

concluded that the root cause of this event

was poor work practices.

Specifically,

RLAs signed off steps

in SIs that were not personally

performed.

A contributing factor

was attributed

to

inadequate

supervision

by the

CSS for not ensuring that compensatory

SIs were

completed

when performed.

The licensee

immediately

removed

the

RLA

responsible

for the

reportable

occurrences

from safety

This is considered

a violation of TS 3.2.K,

and identified as

VIO

259,260,296/91-10-02,

Missed Compensatory

Samples.

This

VIO is not a

NCV because

of the falsification of records.

Unplanned

On April 12,

1991,

at

12:34

a.m.

the outboard

MSIVs closed

due to

actuation of the

Group

1

PCIS logic.

The automatic closure occurred

while

instrument

technicians

were

valving

out

the

high

side

connection of level transmitter

2-LT-3-56B.

This occurred at step

7.6 of 2-SI-4.7.D. l.d.1, Instrument Line Flow Check Valve Operability

Test.

Although step

7.6 was expected

to initiate the

B1 logic of the

Group

I

PCIS logic by removing the

AC electrical

power supply from

the

AC solenoids for the outboard

MSIVs an isolation was not expected

since

the

DC solenoids

should

have held the

MSIVs open.

An initial

investigation

by operations

has

determined

that the

DC power supply

was interrupted to relay

16AK52 associated

with

WO 91-29308-00.

The

licensee

is still

investigating

this

event

and

an

incident

investigation

report

and

LER will be issued

when that investigation

is complete.

The inspector will review the licensee's

investigation

report when it is issued.

e.

Licensed Operator Overtime

A NRC violation was

issued at the Sequoyah facility due to

a problem

with unapproved

excessive

overtime for operations

personnel.

The

licensee

recently

revised

TVA Standard

2.1.7,

Administration of

Overtime,

as

the

result

of that

violation

to clarify the

requirements.

The inspector

determined

that

Browns Ferry currently

has

41

SROs

(31 unrestricted)

and

37

ROs

(12 unrestricted)

which

should

be more than

adequate

to satisfy all operational

requirements

without the use of routine excessive

overtime.

An inspector

reviewed

the licensee's

program at

Browns

Ferry for

ensuring

that licensed

operators

do not work overtime in excess

of

NRC guidelines identified in Generic Letters 82-02, 82-12,

and 83-14.

These guidelines

are

implemented

by TVA in SDSP 19.3, Administration

of Overtime,

and

TVA Standard

2. 1.7.

These restrictions

apply to all

employees

performing safety related

work.

Deviations

from these

restrictions

are allowed if prior approval

by the proper

management

level is obtained.

The inspector

reviewed selected

portions of the

control

room logs

and the weekly operations shift schedule for March

1 - April 15,

1991,

to determine if any apparent

discrepancies

existed for operator

overtime.

The inspector did not identify any

examples of excessive

overtime during that period.

Additionally the inspector held discussions

with Site

gA Organization

personnel

to determine

the extent of that

groups

involvement for

oversite of activities in this area.

The inspector

was informed that

the onsite

gA organization

had responded

to the Sequoyah violation by

performing

a special

monitoring report during the last part of 1990.

The inspector

reviewed portions of this licensee

monitoring report

and

determined

that

one

example of an operator

exceeding

overtime

limits had

been identified.

This discrepancy

is also identified in

CA(R

BFg900397P

and

deals

with exceeding

the

72 hour8.333333e-4 days 0.02 hours 1.190476e-4 weeks 2.7396e-5 months in

7

days

limitation without proper authorization.

gA personnel

had reviewed

time sheets

for 75 operations

personnel

as part of this review.

The

inspector

determined that this discrepancy

was

an isolated

case.

Based

on the above reviews the inspector

determined that the licensee

has

an

adequate

program

to control

the

use of overtime

and

a

sufficient number of licensed

operators

exists

to allow restart of

Unit 2 without the use of excessive

overtime.

One violation was identified in the Operational

Safety Verification area.

4.

Post Modification Testing

(37700,

72701)

The

inspector

continued

to

observe

and

review

the

licensee's

PMT

activities.

In

a

the

inspector

noted that

the

licensee

not document

any TD's if a

PMT was

being performed

as

a testing

work plan.

The inspector

informed the licensee

that this

was considered

a significant weakness.

After additional

reviews the following was noted.

a ~

b.

c ~

d.

The inspector

noted that during the

ILRT the reactor building to

torus

vacuum breakers

opened

on positive pressure.

This indicated

that the logic for vacuum breaker

opening

was reversed.

The breakers

opened

when the pressure

on the torus

was approximate

1 psig greater

than the pressure

in Secondary

Containment.

No excessive

amount of

air was

blown into the reactor building because

the

check

valves

held.

Testing following design

change

P3051

was

inadequate.

This

item is considered

the first example of

a 'violation of 10 CFR 50

Appendix B, Criterion XI, Test Control,

and is identified as

VIO 2,'59,

260, 296/91-10-03,

Inadequate

Test Control.

The inspector

noted that during the performance of testing work plan

WP2274-90

on June

12,

1990, for System

73,

HPCI, valve 2-FCV-73-30

was inadequately

tested.

The testing work planned required that the

valve be tested

from the local control handswitch station but did not

specify

how the valve

was to be tested.

The item was signed off as

being satisfactorily tested.

Additional work activities involving

WO

91-24404-00

on January

12,

1991,

discovered

that valve 2-FCV-73-30

would not function from the local control switch station

due to a

follow on installation.

The. valve was tested

from the control

room.

The operator

held the switch continuously

because

when

he released

the switch the valve would stop in the open position.

The switch is

a "seal-in" switch which can

be released

once actuated.

The follow

on installation

was part of

a cable

replacement.

This

item was

discussed

with the

licensee

and

a

PMT scheduled

in February

1991

should

have corrected this problem.

In

a

report,

the

inspector

expressed

a

concern

involving the

licensee's

modification to the

decay

heat

removal

mode of operation for the

SBGT system.

This concern dealt

with a single failure of the

SBGT Train

B with the other two trains

operating.

During this reporting period,

the licensee

performed

a

special test with the

SBGT Train

B stopped,

the inlet damper

open

and

both trains

A and

C running.

The recorded flow up through the stack

was

16,519

cfm and

back flow though train

B was

254 cfm.

The 16,519

cfm flow up the stack

was greater

than the

16,200

cfm discussed

in

the

FSAR Section 5.3.3.7,

Standby

Gas Treatment

System.

The inspector

requested

and

received

from the licensee

a list of

testing

work plans

which were

performed

on the following systems:

System

23,

RHR

Service

Water;

System

31,

Control

Bay

HVAC;

System 57-2,

120/208 Volt AC Distribution; System 57-5,

4160 Volt AC

Distribution;

System

64,

Primary Containment;

System

67,

Emergency

Equipment

Cooling Water;

and

System

74,

Residual

Heat Removal.

The

items reviewed consisted of 31 testing

WPs which were

used to test

26

ECN/DCNs.

The system

reviews were

as follows:

System

23,

RHR Service

Water.

Testing

WP 0403-90

was written to

verify vibration readings

for

pump

A after the completion of

ECN/DCN H4330.

Testing

WP 2554-90

was originally written to perform

electrical testing of equipment associated

with System

23 affected

by

the

implementation

of

ECN/DCN

11605.

A review of the

completed

testing

WP indicated that all

PMT for System

23

had

been deleted.

The inspector

was unable to determine

why this

PMT had

been deleted.

System

31, Control

Bay

HVAC.

Testing

WP 2554-90

bump tested

the

1B

control

bay chilled

water

pump

motor for proper

rotation

in

accordance

with procedure

ECI-O-OOO-MOT00-1, for

ECN/DCN

11605;

Testing

WP 2332-90

bump tested

the

ACUs

2A and

2B drive motors for

proper rotation for ECN/DCN 10421;

and Testing

WP 2637-90 tested

the

emergency air conditioner motor for electrical

board

rooms

1A and

1B

for ECN/DCN 2242.

System

67,

Emergency

Equipment

Cooling Water.

Testing

WP2356-90

tested

the strainer

motors for A and

D and also stroke tested

valves

2-67-13

and

67-49 for

ECN/DCN

5547;

and Testing

WP 3042-90

was

written to

perform

dynamic

testing

on

pressure

control

valve

3-PCY-67-78

per

procedure

EMI-60,

Inspection

and

Preventive

Maintenance of Control

Bay Ch'illers.

In

testing

WP

3042-90,

a

reference

was

made

to

procedure

PMTP-BF-67-024.

The test

form did not indicate that

PMTP-BF-67.024

had

been

performed

or when it had

been

performed.

The inspector

noted that

some of the testing

WPs involved more than just rolling a

few wires

as initially indicated

by the licensee.

The inspector

concluded

from the

observations,

reviews,

and

followup that the

documentation

of testing

WPs activities

was

tenuous

at best.

The

inspector

was

informed that

the licensee's

gA organization

was

reviewing testing

WPs

and

had

made similar observations.

The

gA

personnel

stated

that

a

PRO

was

being

issued

to formally identify

their observations.

The licensee's

gA/Engineering Department

issued

monitoring report

gBF-R-91-1177,

Work Plan

Review for

WP Testing

Requirements,

dated April 4-5,

1991.

This report documented similar

observations

made

by the inspector.

This report also

contained

a

copy of

PRD

BFP

91

0118P which documented

the specific findings of

the

review.

The results

included

seven

improvements

to

program

items:

Instructions for handling

documentation

such

as

Form SDSP-417s

and work plans

should clear ly delineate

the final destination of

the

record

copy

and

where

the official

gA record

can

be

obtained.

Preliminary

( Information Only)

Form

SDSP-417s

in closed

work

plans

in the vault

can

cause

confusion.

Some consideration

should

be

given to either issuing/revising

Form

SDSP-417s

or

ensuring final 417s catch

up with the work plan prior to putting

in record storage.

Also,

417 forms other than official design

change notice copy should

be marked "for Information Only."

More attention to detail is needed

and

perhaps

some additional

training.

Clarification in program documents

(SDSP 17.2) that TD's are not

required for true installation tests.

Plant instructions

need to provide

space for component

unique

identification at beginning of instruction.

Instructions

need to address

under what conditions

an installat-

ion test

can

be performed,

during

PMT.

SDSP

17.2

needs

to provide instructions

to specify specific

applicable

portions of the test procedure

to be run instead of

all of the test procedure.

Based

on

the

reviews,

observations

and

followup the

inspector

concluded

that the licensee

has

resolved

the issue of testing

work

plans.

One violation was identified in the Post Modification Testing area.

5.

Restart Test Program

(71711)

During

the

reporting

per iod,

the

inspectors

reviewed

the

licensee's

restart activities

to test

and verify that

equipment will perform in

accordance

with required

sp6cifications.

In addition,

procedures

were

reviewed

and

walked-down for adequacy

determinations.

Results

of the

areas

reviewed are

as follows:

1.

Major Equipment Star t-Up and Testing

Reviewed

a 0

Reactor

Core Isolation Cooling (RCIC), System

71.

The system restart test

was performed per 2-TI-188, Reactor

Core

Isolation

Cooling,

which

included

the

performance

of

SI-4.5.F.l.e,

RCIC System

Rated

Flow At 150 PSIG, to verify the

operability of the

RCIC system

in conformance with requirements

specified

in

TS 4.5.F.l.e.

The scope

was to verify that the

RCIC turbine,

pump,

and auxiliaries could

be operated

from the

control

room and deliver

a rated flow of 600

gpm at

a discharge

pressure

of 80 psig

above

the operating

steam

pressure.

In

addition,

RTP-071,

required that the flow rate

and pressure

be

obtained within 30 seconds

from a cold start.

The inspectors

walked

down the TI prior to the test

and reviewed

this testing in progress

on March 22,

1991.

The

pump was unable

to deliver the flow at

a discharge

pressure

of 480 psig within

10

b.

the required

30 second

time period.

The discharge

pressure

was

adjusted

to slightly above

the required

330 psig

and the test

was successfully

re-performed

on March 28,

1991.

High Pressure

Coolant Injection (HPCI), System

73

An overspeed

turbine trip was

performed

on March 29,

1991 per

MMI-23.

This activity had

been attempted

on March 28,

1991, but

a malfunction in the mechanical

trip device

did not perform

correctly

and

had to

be replaced.

After the replacement,

the

unit was restarted

and the trip device

was adjusted

to trip the

unit between

4920

rpms

and

5080

rpms.

The trip occurred at

approximately

5010 rpms which met the acceptance

criteria.

The inspector

viewed the test in progress.

No deficiencies

were

noted during the

performance

of the test or with the results

obtained.

co

Reactor Recirculation

Pumps,

System

68.

The

A recirculation

pump

and

associated

equipment,

including

the

MG set,

was

run on. March 26,

1991,

as part of the plant

restart effort in accordance

with 2-0I-68, Reactor Recirculation

System.

Pump

speed

is limited to

28% when feedwater flow was

less

than or equal to 205.

The test run was reviewed

by the inspector while in progress

and

was determined

to be successful

in that the equipment

performed

well except for two problems.

They were:

(1) the cooling water

controller to

the

hydraulic oil heat

exchanger

on

the

MG

hydraulic coupling did not control

in automatic

mode

and

was

manually operated,

and (2) The normal electrical

feeder breaker

for the

MG motor would not operate

which required the alternate

breaker to be in service for the operation.

Work requests

were

issued to correct the deficiencies.

d.

The

B recirculation

pump

was

run

on

March

28,

1991,

and

witnessed

by the inspector.

The

pump speed

was restricted

to

that of the

A

pump

and

the test

was

completed

without any

identified problems.

Reactor

Systems

Integrated

Cold Functional

This test,

2-BFN-RTP-ICF,

CN07, Integrated

Cold Functional,

was

performed

on April ll, 1991.

The test successfully

demonstrated

that the plant could

be cooled-down

and maintained

from outside

the main control

room if required during an accident situation.

This testing

was reviewed while in progress

by the inspector.

No deficiencies

were noted during the inspectors

review of the testing

activities.

~ ~

11

6.

Safety Parameter

Display System

(SPDS)

Following the event at TMI, operating reactor

licensees

were required to

install

a

to display to operating

personnel

a

minimum set of

parameters

defining the safety

status

of the plant.

Supplement

1 of

NUREG-0737 clarified eight

SPDS requirements.

The

NRR staff conducted

an audit at

BFN in November,

1990, to assess

the

status

of the

ISPDS with regard to the eight requirements

of NUREG-0737,

Supplement

1.

The staff concluded

that the

ISPDS

implemented for

BFN

Unit 2 satisfied six of the eight

SPDS requirements.

The two requirements

which were

not met were:

(1) provide rapid

and reliable aide

and

(2)

incorporate

accepted

human factors principles.

A third requirement

which

had not been

implemented at the time of the audit

was to have procedures

in place

and operators

trained with and without the

SPDS.

During this reporting period,

an inspector discussed

the open

issues

with

cognizant

licensee

personnel

and reviewed the

ISPDS in the Unit 2 control

room.

The

requirements,

NRR staff issues,

licensee

corrective

actions,

and the inspector's

findings were

as follows:

a ~

Requirement

- The

should rapidly and reliably aid the Control

Room operators

in determining the safety status of the plant.

Issue

1:

The software configuration

management

system

needs

to

be

formalized.

Action:

Issue

2:

Software

changes

on the

and plant process

computer are

controlled by section 3.15 of SSP-2.12,

Control of Computer

Application Software.

Software

changes

are initiated with

a "Software Services

Request"

which requires

review and

approval

from Nuclear

Engineering,

Information Services

Operations,

Technical

Support,

and other organizations

that

may be affected

by the change.

During the change

process

a

10 CFR 50.59 review and independent quality review are also

performed.

Some

touch screen

displays

and

keyboard function keys were

observed

to be unreliable

and must

be corrected.

Action:

Issue

3:

The function key problem

was

a result of a missing file on

the

terminal

which

was

repaired.

In addition,

SAIC

developed

and

provided to

TYA a

touchscreen

calibration

program,

which

improves

the calibration

process.

Few

problems

have occurred since the

NRR audit.

Procedures

need

to

be

developed

for which

terminal

locations

and

personnel

can

make

ISPDS

software

and

database

changes.

12

Action:

The only terminal

that

can

make

software

and

data

base

changes

is located

in the

BFN computer

room.

Terminal

access

is controlled

by software

codes

and all software

changes

are

implemented

in accordance

with

TVA Nuclear

Standard

2. 12 and .BFN SSP-2. 12 (see

Issue I).

Issue 4:

Procedures

need

to

be

developed

to specify

how sensor

inputs with different scanning

rates

would be handled.

Action:

All sensors

on the currently installed

ISPDS have the

same

one

second

update

rate.

For the final

SPDS,

TVA will

ensure

that

a point is readily identified if it is scanned

at

a rate other than

a normal

update rate

by indicating in

the point description or point ID number.

b.

Requirement

- The

SPDS shall

be

designed

to incorporate

accepted

human factors principles.

Issue I:

The workstation

display for the

reactor

operator

had

excessive

glare from the control

room overhead lights.

Action:

The

CRT displays

were replaced with non-glare

CRTs.

Issue

2:

The

hood

on the

above workstation

obscures

part of the

ISPDS display (e.g.,

the

ISPDS parameter

summary box) when

an operator is in a standing position.

Action:

The hood was permanently

removed.

c.

Requirement - Procedures

should

be developed

and operators

trained

with and without the

SPDS available.

Issue:

Action:

At the time of the audit,

BFN Unit 2 operators

had received

only

a two-hour briefing on the operation of the

ISPDS.

The licensee

stated that operators

would be trained

on the

use of the

ISPDS during both the normal

EOI (i.e. without

the

SPDS)

and requalification training programs.

All operators

have

been

trained with and without

ISPDS.

Operator

training

without

ISPDS

is

accomplished

by

simulating failure of data

inputs

to the

ISPDS

during

training

on

the simulator.

In addition,

a feature

to

simulate failure of the computer which supplies

ISPDS data

has

been

installed

and

was

used

on the most recent

NRC

initial license

exams at

BFN.

Both of these

features

are

included in the training lesson

plans.

The inspector

concluded that all of the issues

identified by the

NRR audit

had

been satisfactorily

addressed

by the licensee.

No deficiencies

or

further concerns

were identified.

When

TVA declares

the full BFN Unit 2

SPDS operational

(during operating

cycle 7), the

NRR staff wi 11

issue

a

supplemental

safety evaluation.

~ ~

13

7.

System Pre-Operability Checklist (71707)

The inspectors

continued

to monitor the licensee's

activities to evaluate

and

upgrade

plant equipment

and documentation

as necessary

to insure that

plant systems

are in compliance with applicable

standards

and commitments

to support their required functions.

Those

systems

reviewed during this

reporting period are listed as follows:

a ~

b.

C ~

Service Air (System

33)

A walkdown of the major system

equipment

and completed

package

was

reviewed

by the inspector.

The

system

had required

extensive

maintenance

to meet to acceptable

standards.

Much of this

maintenance

was reviewed

by the inspector while in progress.

Only one deferral

remained

open which addressed

the quality level for

procurement

of consumables

utilized in the

system.

However, this

item has

been justified as not being

an operability issue.

All primary/critical drawings for the system

have

been

reviewed

and.

updated

as

appropriate

and

have

been

added

to the computer-aided

drafting program.

In addition, all drawing discrepancies

and design

change

notices

were reviewed

and

a determination

was

made there

were

none which would impact operability of the system.

No outstanding

deficiencies

were noted during the review that would

affect system operability.

Secondary

Containment,

System

64C

This system

was discussed

previously in IR 90-37.

The

SPOC for this

system

was

completed

on

December

10,

1990.

The inspector

reviewed

the

package

with the cognizant

system

engineer

and

noted that

three deferrals

and three exceptions

were issued.

The inspector also

noted that all the exceptions

and deferrals

had

been

closed.

No

deficiencies

were identified during the review of the

SPOC package.

The inspector also discussed

current

SMPL activities with the system

engineer.

The

inspector

noted that

there

were

several

actions

remaining

open for this

system;

however,

none

were

required

to

support the restart of Unit 2.

Off-gas

(System

66)

The inspector

reviewed the licensee's

completed

package of the

Off-Gas

System.

The

package

had identified

3 deferrals

taken for

items that could not be completed at the time of the

SPOC completion.

Two of the deferrals

involved testing that required the system to be

in operation to complete.

The third deferral

required replacement

of

4

RTDs that

was to be replaced

to meet design requirements.

Each of

the

3 deferrals

were evaluated

as not to affect system operability.

Oi

14

The inspector

had previously walked-down

the

system

and determined

that the equipment

appeared

to have

been well maintained.

However,

various

checkouts

were

performed

on the

system

which resulted

in

equipment,

instrumentation,

and electrical

components

being replaced

and/or modified.

This included,

but not limited to the replacement

of one of the mechanical

vacuum

pumps, electrical

cable for ampacity

problems,

pipe

and

tubing supports,

and

various

instruments

and

electrical

components.

In addition, the charcoal

absorbers

underwent

a dry-out/regeneration

activity to reduce

the moisture loading.

Based

on the review of the

system

and

Package,

there

were

no

remaining

open items determined

to prevent operation of the system.

Reactor Recirculation,

System

68

The

SPOC for this

system

was

completed

on

March

15,

1991.

An

inspector

accompanied

licensee

personnel

on the final walkdown for

this

system

on

January

24,

1991.

No major deficiencies

were

identified.

The

inspector

reviewed

the

package

with the

cognizant

system

engineer

and

noted

that

one

exception

and six

deferrals

were

issued.

The

exception

was

issued

to follow

replacement

of the recirculation

pump seals.

Replacement of the

pump

seals

has

been

completed

and

tested

during

the

reactor

vessel

hydrostatic

test.

Four

of

the

deferrals

were

issued

for

modifications for which the field work was complete

but testing is

required for closure.

One deferral

was

issued for modifications

which were complete for system

68 but remained

open for work on other

systems.

The final deferral

was issued to follow blockage of the

811

jet

pump instrument line.

The instrument line was

unplugged

during

performance

of the

reactor

hydrostatic

pressure

test

and

this

deferral

can

be closed.

No deficiencies

were identified during the

review of the

SPOC package.

Reactor

Core Isolation Cooling, System

71

The

SPOC for this

system

was

completed

on April 7, 1991.

The

inspector

reviewed

the checklist with the

system engineer

on April

11,

1991.

The

system

was

accepted

by the

plant staff with

5

deferrals

and

no exceptions.

The inspector

reviewed the list of

approved

deferrals

for

RCIC

and did not note

any item that

was

significant or would affect system operability.

The

inspector

followed the

licensee's

activities

associated

with

testing of the

RCIC turbine

on several

occasions

during this

and the

preceding

reporting periods.

The licensee

has

conducted

a series of

special

operational

tests of the

RCIC turbine

on low pressure

steam

supplied

by the auxiliary steam

system.

This test

has

demonstrated

that the system is ready to support plant restart.

Reactor Protection

System,

System

99

The

SPOC for this

system

was

completed

on

December

26,

1990.

An

inspector

accompanied

licensee

personnel

on the final walkdown

on

0

15

g,

h.

December

14,

1990.

No major deficiencies

were identified.

The

inspector

reviewed

the

completed

package

with the cognizant

system engineer

and noted that four deferrals

were issued

against

the

SPOC.

Two of the deferrals

remain

open

pending

the completion of

testing.

One deferral

is

open for modification

on another

system

which is related

to system

99.

The final deferral

remains

open

pending

the closure of paperwork.

All deferrals

are scheduled for

completion prior to Unit 2 restart.

No deficiencies

were identified

during the review of the

SPOC package.

Buildings and Structures,

System

303

This

system

includes

those

various

safety

associated

with the Reactor Building,

Control

Bay, Intake Structure,

Offgas Building, and

D/G Buildings.

The design

and construction of

these

structures

must

provide for mitigation of the effects of

earthquake,

tornado,

and flooding.

Although interior flooding is

considered

under

the

SPAE for this

system

most effects

associated,

with flooding from exterior sources

were covered

as part of the

for System 327, Flood Protection.

The system checklist

was

completed

on March 27,

1991.

The inspector

reviewed the

package with the system engineer

on April 15,

1991.

The

package

included

two deferrals

associated

with licensee

CAgRs which concerned

several

protective coating failures

and

use of

unqualified coatings

inside

the Unit

2 Drywell.

All field work

associated

with these

two

CAgRs

is

complete

and

the

remaining

activities not affecting system operability.

Carbon Dioxide and Fire Protection/Generator

Purge,

System

39

Containment Ventilation, System

64B

Radwaste

System,

System

77

The inspector

reviewed

and observed

the licensee's

activities in the

process.

These

observations

and

reviews

included

walkdowns,

review of punchlists,

and

PM.

Diesel Generator,

System

82

Additional reviews

were

performed

on

System

82.

This system

was

previously

SPOCed

and reviewed.

A TS change

was expected

and later

approved.

This change required that each

DG be tested at 2800

kw for

at least

24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months .

The licensee

commenced

these series of run tests

on March 18,

1991.

0

8.

Emergency Operating

Procedures

The

use

of the

BFN revision

3

versus

revision

4

has

been

reviewed

on several

occasions

by the

NRC staff.

During this reporting

period,

an inspector

reviewed

documentation

for EOIs

and

AOIs to verify

that the licensee

had completed

commitments

required for Unit 2 restart.

0

~ ~

16

a

~

In a letter dated January

18,

1990, the

NRC staff concluded that Unit

2 could

be safely restarted

using

the

BFN revision

3 EOIs.

This

letter contained

NRC staff

comments

on

6 items of concern.

TVA

reviewed

each of the comments

and submitted corrective actions to the

NRC by letter

on July 6,

1990.

The inspector

reviewed

the letters

and verified that the licensee

had implemented

the commitments.

The

issues,

responses,

and actions

taken were as follows:

Comment:

EOI

Guideline

PC/P,

Primary.

Containment

Pressure

Control, action PC/P-6. 1 is intended to prevent actuation of the

wetwell sprays if the nozzles

are

submerged.

Spray actuation

would result in water

from the nozzles

entering directly into

the

PSC.

The

have

taken

this action

when

the level

instrumentation

is at the maximum value.

For BFN, this limit is

20 feet while the nozzles

are at 26.4 feet.

This results

in a

space

of 6.4 feet where the sprays

could possibly

be effective,

but are not actuated

due to limitations of instrumentation.

(2)

(3)

TVA should

review alternate

methods

to determine level, rather

than limiting the actions

due to instrumentation limits.

Action:

The licensee

r'evised

PC/P to direct actuation of

PSC

sprays

based

on pressure

parameters

only.

Limitations based

on

level

were

removed

from the

procedure

at

steps

PC/P-2

and

PC/P-6.

Base

documents

were

also

revised

to support this

change.

Comment:

In general,

the

reason

for the action

given in the

EPGs in parentheses

is the basis for the action.

However, the

have also triggered actions

due to instrument limitations

as indicated in (1) above.

If the action is based

on instrument

limits, it should

be so stated.

Action:

The

EOIs were reviewed to ensure

that actions

taken

based

on instrument limits reflect that the instrument

range

limitations is the basis for the action.

Guidelines

PC/P-2

and

PC/P-6 were revised

as described

in (I) above.

Comment:

The draft that

the staff

reviewed

had

several

inconsistent referrals.

For example,

PSC water level

was given

in negative

inches

from normal

as well as

a measurement

in feet

from an absolute reference.

Also, the wetwell is referred to as

the primary containment

as well

as the wetwell.

Special

note

should

be taken to eliminate these

types of inconsistencies.

Action:

TVA agreed

that

inconsistent

referrals

should

be

eliminated.

In the first example

given, the inconsistency

is

necessary

due

to the

need

to

use

narrow-range

rather

than

wide-range instrumentation for determining

PSC water level.

The

narrow-range

instrument

has

a different zero reference

point,

with zero

being

normal level,

and

reads

in inches rather

than

feet.

I

0

Cl

0

17

The EOIs were reviewed to ensure

consistency

in the use of terms

concerning

primary

containment,

drywell,

and

PSC.

No

inconsistencies

were found in the procedure.

The program manual

was revised to include definitions for all terms

used to signify

components of primary containment.

Comment:

The actions

indicate

there

is

no manual

action to

initiate either the

RHR or

CS room coolers.

This limitation is

not normal for a Mark I design

and should

be verified before the

option is discarded.

Action:

The

RHR and

CS equipment

area coolers

are not provided

with manual start capability.

The cooler fans

auto start

on

a

corresponding

pump start or increasing

temperature

(100

F) in

the

immediate

area.

Control

Room indicators

are provided for

monitoring

pump

room temperatures

and annunciators

are provided

to alert operators

of cooler failure at

150 F.

The option to place

jumpers at the

room cooler circuit breakers

in the

RMOV boards

was reviewed.

This option was considered

to

be unfavorable

when weighing the capacity of the coolers

and the

adequacy

of existing cohtrols

and instrumentation

against

the

necessity

to dedicate

personnel

to this task

under

accident

conditions.

Comment:

Several

actions

in EOI-3, Secondary

Containment

and

Radioactive

Release

Control,

are

based

on the operator

being

able determine

whether the pipe rupture is within the primary or

secondary

system.

However, there is no guidance

provided to the

operator

in determining

which system

is affected.

This is

a

deviation

from the

symptom

based

EOIs since it requires

an

evaluation

by the operator.

The

should

provide

some

guidance

on how an operator should determine

the break location.

Action:

EOI-3 was revised to provide guidance

to the operator

for determining

break location.

Comment:

EOI Guideline

RC/P,

Pressure

Control, contingency

C2,

Emergency

Depressurization,

is entered if reactor

water level

cannot

be determined.

However, in the

BWROG revision

3 EOIs,

the appropriate

action is flooding the

RPV when reactor water

level

cannot

be

determined.

TVA committed

to revise this

procedure to reflect

BWROG guidance.

Action:

Previously,

in the

BFN EOIs,

C2 was entered if reactor

water

level could not

be determined.

After depressurization,

C4,

RPV Flooding,

was

then entered.

TVA revised EOI-1, Reactor

Control,

to require

entry into

C4

when

RPV level

cannot

be

determined,

rather than enter

C2.

0

18

b.

In

a letter dated

October

16,

1989,

the

TVA made

two commitments to

improve

and

AOIs for Unit

2 restart.

The inspector

reviewed

documentation

and verified that the actions

had

been

completed.

No

deficiencies

were identified.

The commitments

and actions

taken were

as follows:

(1)

Commitment:

Develop

and

include

additional

information for

hydrogen control in an AOI which references

EOI-2.

Action:

Procedure

2-A01-64-8, Primary. Containment

High Hydrogen

and/or

Oxygen,

was issued

on June

1,

1990, to provide

symptoms

and operator actions f'r control of combustible

gases

(Hydrogen

or Oxygen) in the primary containment.

(2)

Commitment:

BFN will have

a method

in place

to provide

an

alternate

means of injecting boron following a failure to scram.

Action:

Procedure

1-SOI-26,

Use of U-1

SLC Storage

Tank

As An

Alternate Source of SLC Injection For U-2, was issued

on January

10,

1990,

to maintain the Unit

1

SLC system

tank available for

injection into the Unit 2 reactor

vessel

by using

the

1B

CRD

system

pump.

This

method

is

included

in

EOI

Appendix

2,

Alternate

SLC Injection.

c.

As part of the procedure

upgrade

program committed to in Volume 3 of

the

NPP, all AOIs were to be upgraded

and verified prior to Unit 2

restart.

The verification process

for AOIs includes

an in plant

walkdown to verify that the instruction will work and obtain operator

comments.

The backup control

room provides

a means for operators

to safely shut

down the reactor

from outside of the main control

room in the event

of

a main control

room evacuation.

Procedure

2-AOI-100-2, Control

Room Abandonment,

provides

instructions for Unit 2.

The inspector

reviewed the AOI to verify that comments

from the plant walkdown were

incorporated

into the instruction.

The

walkdown

was

conducted

by

Operations

personnel

in October,

1990.

The inspector

noted that most

of the

comments

involved equipment

labeling

and procedure clarity.

The

inspector

reviewed

the

walkdown

comments

against

the current

revision of the AOI (rev. 7) and noted that appropriate

comments

were

incorporated.

The inspector identified

a few typographical

errors to

Operations

personnel

who

issued

procedure

changes.

No further

deficiencies

or inspector

concerns

were identified.

9.

Nuclear Instrumentation Reliability

Browns Ferry has

had

a history of poor reliability associated

with System

92,

Nuclear

Instrumentation.

As part of the

Information Notice No. 90-78,

process

the

system

checklist for System

92 was

completed

on August 27,

1990.

This activity

was monitored

by the inspectors

as

documented

in inspection reports

90-25

and 90-27.

The system

was accepted

by plant staff with a relatively high

number of exceptions

and deferrals

compared

to other system

SPOCs.

This

19

was mostly due to

a large

amount of unfinished work activities that were

needed

to support

system operability.

These

in'eluded

replacement

of ll

defective

LPRM strings

and repair or replacement

of various

undervessel

cables.

Various work on the

NIs

was

ongoing until just before fuel load.

The

licensee

initiated core reloading activities

on February

21,

1991, while

experiencing

problems

with

SRM Channel

B.

The

sensor

input to this

channel

was

from

a

FLC which is

more sensitive

than

the

normal

SRM

detectors.

A later licensee

investigation

determined

that the excessive

noise resulted

from damaged

connectors

was

a contributing factor to this

problem.

The

IRM gain

adjustments

were set to

maximum

by 2-TI-233 the

weekend

before fuel load.

Just prior to fuel load the range

switches

were set

down to position

1 (the lowest position).

Noise

was evident

on the

IRM

channels

with channel

readings

remaining

above

downscale

throughout fuel

loading.

There

was

no noticeable

correlation

between

channel

indication

and fuel bundle addition.

Since fuel

load

was

completed

the licensee

has

experienced

many nuclear

instrumentation

spikes

and

spurioUs

RPS trips.

Some of the spikes

were

severe

to cause

a channel

Hi-Hi alarm but not of sufficient duration to

pick

up

a

RPS

relay

and

cause

a halfscram.

Most of this erratic

indication could readily

be attributed to noise

and faulty

IRM signals

that could result

from various

causes

such

as

grounding

problems,

bad

connectors,

cable crosstalk or inductive cable coupling.

GE

was

contacted

and

arrangements

made

for

a visit by

a

vendor

representative

from GE's office in

San

Jose

to travel to the site to

assist

in the troubleshooting effort.

Based

on their review

GE made the

following test methodology recommendations:

Measurements

of shield to ground resistance,

signal to noise ratio,

and capacitance

to check the transmission

path from the detectors.

Verification of proper

impedance

values for channel

penetration,

cable,

connector

and detectors.

Connection

of

a

recorder

with adequate

"response

to selected

IRM

channels

to allow recording of spike amplitude.

This data to be used

for accurate

baseline

data.

Licensee

should diagnose

SRM,

IRM,

LPRM channels

with known spiking

problems to determine

location of problems.

Concentrate

on channels

which shown inoperative

by analysis of cable data.

After adequate

date is obtained,

the licensee

should readjust

the

gain

on the

IRMs to an

optimum value (reduction in gain could mask

the noise

and

make troubleshooting

more difficult).

t

20

The

inspectors

followed the

progress

of the licensee's

troubleshooting

activities

in this

area.

A description

of activities

observed

is

described

in paragraph

2.a.

During his visit the vendor also

showed the

licensee

better

ways to acquire

low current measurements

for data taking

and

newer connector

cleaning techniques.

The troubleshooting activities

on the NIs had to be halted

on April 22,

1991,

due to scheduled

work under

the Unit

2 reactor

vessel

to replace

the leaking

CRD Mechanism 0-rings

that

had

been identified during the

RPV hydro.

NI troubleshooting

is

scheduled

to

resume

when the 0-ring replacement

work is completed.

GE

personnel

in San Jose

are presently working on procedures

to assist

TVA in

further testing

and

setup of the nuclear instrumentation.

They plan to

complete this effort in time to allow returning to the site for the second

stage

of troubleshooting.

The inspectors will follow the licensee's

actions in this area during the next reporting period.

10.

TMI Action Items

a.

(CLOSED)

260/TMI Action Item II.E.4.2. 1-4,

Containment

Isolation

Dependability - Implement.

BU-80-06,

Engineered

Safety Feature

Reset

Controls,

which was closed ia

IR 90-40.

The

NRC issued

TS Amendment

No.

193 on March 22,

1991.

This was

a review of 10 CFR 50 Appendix J

and TMI Item II.E.4.2.1-4.

The

SE concluded that TVA identified each

containment

system for Unit 2 as essential

or non-essential,

assured

that all essential

systems

were

remote-manually

operated

and

the

non-essential

systems

met

the

intent of isolation

requirements

specified in General

Design Criteria of 10 CFR Part 50, Appendix A.

Additionally,

a

successful

integrated

containment

leak

rate

inspection

was

completed

on

March

18,

1991.

These

reviews

and

inspection close this item.

b.

(CLOSED)

260/TMI Action Item II.F.2, Instrumentation

for Detection

of Inadequate

Core Cooling.

This item evolved from the Three Mile Island Incident and

addressed

the

need for additional

instrumentation

or controls

to supplement

existing instrumentation

to provide

an unambiguous,

easy-to-interpret

indication of inadequate

core cooling.

Two categories

of permanent

physical

improvements

were identified in Generic Letter

No. 84-23

which provided

BWRs with clarification of

NRC requirements

in this

area.

The two required areas of improvement are

as follows:

Improvements

to plant design that will reduce level indication

errors

caused

by high drywell temperature.

Review of plant experience

about

mechanical

level indication

equipment.

Evaluation

of

any existing

mechanical

level

indication for replacement with analog level transmitters.

21

The

inspector

reviewed

documentation

provided

by the

licensee

to

support

closure of this

TMI Action Item.

During this review the

inspector

determined

that the licensee

had adequately

addressed

this

issue

to support closure of the item.

Specifically the

inspector

determined

the following:

ECN

P7131

rerouted

the Unit

2 reactor

water level

reference

legs,

to minimize the routing of the reference

legs

inside the

drywell.

The resident staff followed the

progress

of this

modification

including

performing

a

walkdown of the

work

activities.

This

inspection activity was

documented

in

IR

88-32.

This

ECN is

now field complete with the cold functional

portion of the post modification testing

completed.

The only

remaining testing that is not complete is the hot functional

testing of water level instrumentation

which must

be performed

after restart

during

power

ascension

testing.

An inspector

reviewed

the completed modifications work package

and observed

portions of 'the requalification training for licensed

operators

on this subject.

The licensee

replaced substantially all mechanical

reactor water

level indicators for Unit 2 during the performance of ECN P0126.

This modification is field complete.

The inspector

determined

there

are still

two existing

Yaryay

type

water

level

instruments,

2-LI-3-46A and

2-LI-3-46B, which provide

water

level

indication at local

instrument

racks

and at the

remote

shutdown

panel,

2-25-32.

Although these

two instruments

are

mechanical

type level

instrumentation

all Unit 2 water level

instruments

that initiate Reactor Protection

and

ECCS functions

and

provide

level

indication in the control

room

have

been

replaced

with analog

type

instrumentation.

An inspector

observed

portions of the work activities associated

with this

modification

and

attended

a portion of the requalification

training for licensed operators

on this subject.

The staff

had

reviewed

the

above

proposed

modifications

and

determined that both modifications were acceptable

to address

the two

improvement categories.

That review is

documented

in NRC's letter

dated

November

18,

1986.

During the

above

reviews

the inspector verified that the completed

modifications activities were accomplished

by the licensee's

approved

design

change

process.

The licensee

has

adequately

implemented all

improvements identified in the licensee's

responses

to

GL 84-23 dated

April 8,

1985, July 15,

1985,

October

15,

1985,

and March 12,

1986.

Additionally this

satisfies

the

commitment

to

complete

all

NUREG-0737,

Item II.F.2 related modifications prior to restart of the

unit identified in TVA letters

dated

March 1,

1988,

and October

18,

1988.

r

0

22

The inspectors will continue

to monitor licensee activities in the

area

of reactor

vessel

water

level

including followup of any

abnormalities

such

as

level

mismatch.

This will also

include

followup of licensee activities to address

IFI 259, 260, 296/89-35-01

which although not

a restart

item should

be resolved during the power

ascension

testing

program prior to full power operation.

(CLOSED) 260/TMI Item II.K.3.18.C,

ADS Logic.

reports

documented

reviews of this item.

The

outstanding

work activity

required

was

the

performance

of

PMTP-BF-.01.014.

The

post

modification test

was

successfully

completed

on March 22,

1991.

In addition to

a review of the

PMTP the inspector

also reviewed the

following procedure:

2-SI-4.2.8-1,

Core

and

Containment

Cooling

Systems

Reactor

Water

Level

Instrument

Channel

Calibration;

2-SI-4.2.B-ATU,

Core

and

Containment

Cooling Systems

Analog Trip

Unit Functional

Test;

2-ARP-3C, Alarm Response;

and 2-EOI-1, Reactor

Control,

Section

RC/C, Monitor and

Control

RPV Water Level.

The

inspector

also

discussed

this

item with licensee

representatives.

Based

on

the observations,

reviews,

and discussions

the inspector

concluded that this item was implemented.

(CLOSED)

260/TMI Action

Item II.K.3.27,

Common

Vessel

Level

Reference.

The licensee notified the

NRC in a letter dated

March 14,

1991, that

TMI Action Item II.K.3.27, and

Human Engineering

Discrepancy

HED 283

were

completed.

The

design

had

been

finalized,

modifications

completed,

procedures

issued

and the

necessary

training completed.

The inspector

reviewed

the licensee

closure

package for this item.

TS Amendment

157 was issued

November 28,

1988 and changed

the reactor

water level zero reference

point from the top of active fuel to the

bottom of the reactor vessel.

Documented

in IR 90-40

was

the results

of a

NRC audit

team which

concluded that all restart

HEDs including

HED 283 were satisfactorily

implemented.

Plant

DCN

W10396A replaced

the existing

scales

on

2-LT-3-52, 620, 2-LR-3-62 and 2-LIS-3-52,

62A in the control

room and

auxiliary instrument

room to reference

instrument zero

equal

to 528

inches reactor vessel

inside height.

The inspector

concluded

the TMI

Action Item had been resolved.

(CLOSED) 260/TMI Item II.K.3.28, ADS Accumulator gualification.

In

the

inspector

documented

reviews

and

observations

involving this

issue.

The

inspector

reviewed

the

following procedures:

2-0I-84,

Containment

Atmosphere

Dilution

System

Operating Instructions,

2-0I-32A, Drywell Control Air System

23

Operating Instructions,

and 2-AOI-32A-1, Loss of Drywell Control Air.

In

addition,

the

inspector

held

discussions

with

licensee

representatives

involving the modification to Systems

84,

CAD and 32,

DCA.

One item remained

open

and involved containment isolation valves in

Systems

32

and 84.

TVA decided

to replace

the Unit 2 outboard

CAD

primary containment

check

valve

in

each

train with

a qualified

normally closed

solenoid

valve

and

a normally closed

manual

valve,

which bypasses

the solenoid

valve, prior to restart

from the next

refueling outage,

Unit 2, cycle 6.

Based

on the reviews,

observations,

and discussions

the inspector

concluded

that this

TMI Action Item was adequately

implemented for

Unit 2, cycle 5.

11.

Information Notice

Previously Unidentified Release

Path

From

Boiling Water Reactor

Control

Rod Hydraulic Units,

was

issued

to alert

licensees

of

a potential

accident

that could

lead

to

a design

basis

accident with radiation

doses significantly exceeding

the values specified

in the

FSAR.

The accident

scenario

involves

a release

path in the

CRD

system

resulting

from

a break in the non-seismic

portion of the water

supply line to the suction of the

CRD pumps at

a time one of the

CRD pumps

is not in operation.

The portion of the referenced

piping is that located

outside the secondary

containment

boundary where

a line break could result

in the path of radiation release.

The licensee

reviewed

the

IN and determined

that stop

check valves

are

installed

in the

BFN system at the discharge

of the

pumps

which would

prevent

a backflow through the system to those portions of piping system.

Although these

stop check valves

have

been leak-rate

tested

and found to

have very minor leakage rates,

they are not part of the

ASME, Section

XI

Testing

Program.

However,

the

licensee

has

internally committed to

include

these

check

valves

in the

program prior to the Unit 2, refuel

cycle 6.

12.

Reportable

Occurrences

(92700)

The

LERs listed

below

were

reviewed

to determine if the information

provided

met

NRC

requirements.

The

determinations

included

the

verification of compliance

with

TS

and

regulatory

requirements,

and

addressed

the

adequacy

of the event description,

the corrective actions

taken,

the

existence

of potential

generic

problems,

compliance

with

reporting

requirements,

and

the relative

safety

significance

of each

event.

Additional in-plant reviews

and discussions

with plant personnel,

as appropriate,

were conducted.

a.

(CLOSED - Unit

2

ONLY)

LER 259/89-03,

Design of Suppression

Pool

Vacuum Relief System

Does

Not Provide

Single Failure

Double

Isolation of Primary Containment.

~

0

24

This was

a voluntary

LER which addressed

the possible failure of the

suppression

pool

vacuum relief system in the open position

on loss of

its unqualified air supply.

This did not meet

the current single

failure double

isolation

design criteria for primary containment

integrity, identified in Generic Letter 88-14.

DCN W14096A

was

implemented

to provide

CAD system nitrogen at the required

pressure

to the

vacuum breaker butterfly valves if the normal air supply from

the control air system is unavailable.

The inspector

reviewed the

closure

package

for this

item

and

the applicable

DCN.

This

DCN

addressed

the concern

and

has

been

completed for Unit 2.

'CLOSED)

LER 259/90-16,

Residual

Heat Service

Water

(RHRSW)

Pump

Auto-Start

Problem Following an

Unplanned

Engineered

Safety

Feature

Actuation.

Between

the

dates

of August

28-30,

1990

plant modifications

electricians

replaced

time delay relay

TD2A and installed

a

new test

block in the control circuit of RHRSW

pump

A3 under

DCN W4515A.

At

the

completion

of this work,

no

PMT

was

immediately

performed.

Although one of the "General

Requirements"

contained

in step

6.1 of

SDSP-14.9

requires

that caution orders

be established

for components

awaiting

PNT and that

a caution order

tag

be

hung at each control

location from which that component

can

be operated,

the inspector's

followup of this event disclosed

that

no such caution order tag was

issued for RHRSW pump A3.

'n October

4,

1990

pump

A3 was lined

up to supply

an

EECW

header

and

"assigned"

per

step

7.8 of 3-SI-4.9.A.1.a(3D)

to

automatically start following the start of the

3D diesel

generator.

Pump

A3 failed to start.

During the inspector's

followup of this

event, it was noticed that the

same

A3

RHRSW pump had also failed to

autostart

a few days earlier during the monthly operability test of

the

3C diesel

generator

performed via 3-SI-4.9.A.1.a(3C)

on September

27,

1990.

This September

27,

1990, failure then resulted

in the

issuance of TD-1 and

WR O'C-039315

(and associated

WO 890-18430-00).

Since the cause of its September

27,

1990, failure to start

was still

unknown

and the associated

corrective action

documents (i.e.

TD and

WO) remained

open at the time, the assignment of the A3

RHRSW pump to

autostart for the

performance

of the October 4,

1990,

SI test

was

incorrect.

This test

control

inadequacy

resulted

in incomplete

compliance

with all the requirements

of plant procedure

PNI-17.1,

Conduct of Testing.

Trouble shooting activities

on the

A3

pump completed after its

October 4,

1990, autostart failure, resulted

in the discovery that

a

wire not terminated

on terminal

8 of auxiliary relay

NVA-Al was

responsible

for the failure.

This lifted lead provides

the coil

current to time delay relay TD2A.

Failure to terminate this wire was

attributed

by

the

licensee

to

personnel

error

during

the

implementation of DCN W4515A.

25

The Dl

pump which was not aligned (i.e. its discharge

valve to

the

EECW system

was closed

even

though the licensed unit operator

on

shift had circled the "assigned"

option for RHRSW pump Dl in the

same

step 7.8 referenced

above

) to respond

to the diesel

generator start

signal, automatically started resulting in an unplanned

actuation of

an

ESF.

During

subsequent

trouble

shooting

of the

Dl

pump start

interlock circuit, the licensee

was unable to identify or duplicate

the cause of the failure that resulted

in the incorrect

pump start.

An improperly fitting relay cover suspected

to be associated

with the

unexpected

pump start

was

removed

from auxiliary relay RI located in

junction box 4860 under work order 90-20209-00

on October 25,

1990.

The absence

of a caution order tag

on the control switch for the A3

pump in addition to the inadequate

review of open corrective

action

documents

are test control

inadequacies

which permitted the

incorrect assignment

of the

A3

RHRSW pump to autostart

on October 4,

1990.

This is

the

second

example

of

VIO 259,260,296/91-10-03,

Inadequate

Test Control.

Any further inspector

followup of the circumstances

associated

with

this

event will be

performed

during followup of the licensee's

corrective action for the violation.

(CLOSED)

LER 259/90-19,

Failure to Establish

Hourly Firewatch

in

the

Required

Timeframe

Necessary

to Meet Technical

Specification

Requirements

for Inoperable

Fire Detection Panels.

On December

12,

1990,

TS 3. ll.A associated

with compensatory

measures

for inoperable fire detection

equipment

was

not met in that the

required

hourly firewatches

were not established

within one

hour.

The failure occurred

when the

1A ILC Bus tripped resulting in a loss

of electrical

power to various fire protection

system fire detection

panels.

Licensee

personnel

attempted

to establish

compensatory fire

watches

within the required

time frame.

However

a lack of clear

guidance for watch areas

in plant procedures

resulted in that process

requiring more that one hour to complete.

The inspector

reviewed

the licensee's

submittal for this event

and

determined

that it met current

NRC requirements

for reporting.

Additionally the

inspector

noted

that

two other similar events

occurred

during

the

same

period

(December,

1990).

These

two

additional

events

are

documented

in

LERs

259/90-17

and

259/90-21.

However this event

had

a different root cause

in that it resulted

from lack of procedural

guidance

which required different corrective

actions.

As corrective action to this event the licensee

developed

a

list of affected

areas for each of the three units.

This list was

issued

as

Operations

Standing

Order

OS-0023

and

was

made effective

March 8,

1991.

Based

on the

above review the inspector

determined

26

that

the

licensee's

actions

should

be

adequate

to

preclude

recurrence.

(CLOSED)

LER 259/90-20,

Rev. 1,

Actuation

During

Relay

Testing

Caused

by Procedure

Inadequacy

and Personnel

Error.

This event occurred

December

14,

1990, during the performance of a

time delay relay setpoint

check for the fuel

pool cooling

pump

1B

motor.

The

power source for the synchronous

timer used

to test the

relay was obtained

from 480V shutdown

board. 1B breaker control

power.

The

leads

of the timer were attached

to the terminals

across

the

shutdown

board's

breaker control transfer switch.

The normal feeder

breaker for the

480V shutdown

board

1B tripped

when the timer was

energized.

The

deenergization

of this

shutdown

board

in turn

deenergized

RPS

bus

1B and the

PCIS logic relays

powered

by the bus,

resulting

in the isolation of group

2 valves

(drywell floor and

equipment drains discharge

valves).

The inspector

reviewed the licensee's

closure

package for this

LRED.

The licensee

attributed

the root cause

of the event to procedural

inadequacy

and

personnel

error.

Personnel

involved received

a

written warning.

The

EMI wi.ll be revised to require

a portable

DC

power supply for time delay relay setpoint checks.

A portable

power

supply will be procured

by May 31,

1991.

(CLOSED for Unit

1

and

2 only)

LER 259/91-01,

Failure of

Two

Trains of the Standby

Power System to Load Sequence

Thereby Creating

the Potential for a Loss of Critical Safety Functions

On

December

27,

1990,

during

the

performance

of

DG

"A" load

acceptance

test,

the

shutdown

board

"A" DGVA relays failed to stay

energized.

On

December

31,

1990, during the performance

of

DG "D"

load acceptance

test,

the

shutdown

board

"D" DGVA relays failed to

energize.

The

licensee

initiated

CA(R

BFP-910004

on

January

2,

1991

and

subsequently

issued

report II-B-91-004 to formally address

the cause

of these failures.

Based

on their

investigation

of the

circumstances,

the

licensee

concluded

that malfunctions in the operation of certain contacts

on

the stationary auxiliary switches

located

inside

each of the diesel

generator

breaker

cubicles

were

responsible

for the

observed

anomalies.

Closure of these

contacts

is required for energizing the

DGVA relays

which is

a prerequisite for the proper sequencing

of the

ECCS loads

and for the initiation of the load shedding logic when

an

accident

signal is present.

Consequently,

any inadvertent

premature

reopening

or failure to close of these stationary auxiliary switch

contacts

is undesirable

from

an

ECCS availability standpoint.

A

simultaneous

combination of the events

that occurred

December

27,

1990

and

December

31,

1990 could have resulted

in a failure of both

27

core spray

pumps

to start, resulting in the potential

loss of both

loops of a critical safety function.

To ensure

the

DGVA relays energize correctly,

a redundant

and diverse

contact,

located

on the breaker

mounted auxiliary switches

of the

diesel

generator

output breakers,

has

been installed in parallel with

the contacts

on the existing stationary auxiliary switches for the

Unit

1

and

Unit

2

shutdown

boards.

A similar design

has

been

implemented

for

the

NVA relays

which could also initiate

load

sequencing.

A similar modification for the

Unit

3

shutdown

boards will be

implemented prior to the Unit 3 startup.

In addition,

on

March

28,

1991,

the

licensee

initiated

a

change

request

to procedure

EPI-O-OOO-BKR002,

"Maintenance

of

GE

(Magne-

Blast)

Switchgear

and Circuit Breakers"

which will require

a visual

inspection

of the stationary auxiliary switch contacts

during the

once per outage

breaker inspections.

Any further inspector followup of these

ESF testing deficiencies will

be conducted

during the followup of unresolved

item

URI 259,

260,

296/90-40-01.

This

LER is closed for Units

1 and 2.

(CLOSED)

LER 296/91-01,

Deenergization

of

Reactor

Protection

System

Bus

by Normal

Supply Circuit Protector

Operations

Caused

by

Failure of Overvoltage

Voltage Monitor.

A February

1, 1991, trip of the

RPS

3A2 circuit protector resulted

in

the deenergization

of the

120 volt 3A

RPS distribution bus.

The

circuit protector trip occurred

because

of a faulty internal voltage

monitor used to sense

supply overvoltage.

The deenergizing of the

3A

RPS distribution bus resulted in an unplanned

actuation of engineered

safety features.

The

systems

affected

by this actuation

were the

Unit

3

Primary

Containment

Isolation

system,

the

Standby

Gas

Treatment

system,

the Control

Room Emergency Ventilation system

and

the Unit 3 Reactor

and

unit-common

Refuel

zones

normal ventilation

systems.

The licensee

issued

IIR No. B-91-030

on February

11,

1991.

Random

component failure of the

3A2 circuit protector

was identified as the

root cause of the event.

A new overvoltage relay was installed under

work order

891-25966-00

and its associated

post-maintenance

testing

per procedure

3-SI-4.1.B-16,

"RPS Circuit Protector Calibration/FT"

was completed

on February 2,

1991.

The defective voltage monitor is being returned to the vendor for an

evaluation

of the failure.

The licensee

is also evaluating

the

availability of substitute

class

1E voltage monitoring hardware for

possible

replacement

of the existing

monitors.

The

inspector

determined that the licensee's

evaluation of this event

was adequate.

~y

28

13.

Action on Previous

Inspection

Findings

(92701,

92702)

a

~

(CLOSED)

IFI

260/87-09-05,

Final

Resolution

of

Unverified

Portions of CCD Drawings.

This issue

was previously'reviewed

in IR 88-33

and

IR 91-02.

The

results

of those

reviews

were that this item was resolved for fuel

load, but

NRC review of audit reports,

CAgRs,

and schedules

for CCD

implementation

were necessary

to close this item for Unit 2 restart.

During this reporting period,

an inspector

reviewed licensee

audits

which involved

reviews of plant design

bases

and

design

control

issues,

including CCDs.

The reports

reviewed were

EA Audit BFT89901

"Design

Change Control",

N(AEE Audit BFA89003 "Technical

Evaluation

of the

RHR System",

and

NgASE Audit BFA90022

"BFN Configuration

Control

and

SPOC/SPAE

Process."

The inspector

noted that the audits

contained

several

examples

of drawing inconsistencies;

however,

no

problems

were identified involving unverified portions of CCDs.

The

problems

identified in the audits

reviewed

were dispositioned

by

approved

plant

procedures.

Drawing

inconsistencies

are

being

followed by other

NRC open items.

From

discussions

held

with cognizant

licensee

personnel,

the

inspector

found that there

were

358

CCDs

completed

and

260

CCDs

remaining to be completed

as of March 19,

1991.

The remaining

CCDs

must

be completed prior to Unit 2 restart.

The inspector

concluded

that

the

licensee

was

implementing

a

program

to

document

the

evaluated

plant configuration of the portions of systems

within the

safe

shutdown

boundaries.

This is in accordance

with the licensee

commitment

for the

Phase

I

DBVP.

No further

concerns

were

identified.

b.

(CLOSED)

IFI

50-260/89-16-09,

Need

to

Consider

Flow

Degradation

Due to Corrosion of Carbon Steel

Piping.

This issue

was identified during the

VSR inspection of the

CS system

and

documented

in IR 50-260/89-16.

The team identified that for all

calculations

involving the determination of a system

pressure

drop,

the

method

used

did not account for an

increase

in the relative

roughness

of the pipe inside surface

as the pipe

ages (i.e.,

due to

corrosion

or

other

causes.)

Specifically,

calculations

MD-f2075-87215,

Revision 0,

"Pipe Sizing of Core Spray System,"

and

MD-(2075-87258,

Revision

1,

"Core Spray

Pump

NPSH

and

Performance

Calculation" did not account for the aging effect

on carbon

steel

piping.

After the initial inspection,

the

licensee

provided

additional

information regarding

flow degradation

due to corrosion of carbon

steel

piping.

This additional

information indicated that other

architect-engineering

f'irms and utilities contacted

by TVA also

do

not assume

flow loss from carbon steel

piping corrosion

on condensate

or feed

system.

The licensee

agreed

to reevaluate it's position

on

the

need

to

assume

flow loss

because

of recent

information

from

J

~

0

29

industry that indicated that corrosion

may have contributed to system

flow loss at other utilities.

During this inspection

TVA provided

the following information

and

position:

TVA has

re-evaluated

the information submitted

to the

NRC

as

follow-up to the

VSR question

on flow degradation

of carbon

steel

piping.

TVA reaffirms its original position

on flow

losses

from degradation

of carbon

steel

pipe

when performing

pressure

drop calculations.

Degradation of carbon steel

pipe in

chemically treated

water

systems

is not

a factor which will

adversely affect pressure

loss.

Any corrosion which may occur

would be

a smooth black iron oxide film, passive

in nature,

and

would retard further corrosion.

This type of corrosion

is

consistent

with

the

roughness

factors

utilized

in

the

calculation.

The

main

component

which controls

the rate of

corrosion

in condensate/feedwater

systems

is dissolved

oxygen

content.

BFNP conforms

to the

GE guidelines for this element

and

has experienced

no degradation of systems

in which the water

chemistry conforms to these

GE guidelines.

TVA has

evaluated

information received

on the

H.

B.

Robinson

auxiliary feedwater suction line problem and determined that the

corrosive condition described

is not relevant to the

CS system

carbon

steel

piping at

BFNP.

The

H.

B.

Robinson auxiliary

feedwater

suction line was susceptible

to introduction of raw

water

which could greatly accelerate

the corrosion

process.

That type of condition cannot

be developed at

BFNP since the

CS

system

piping is normally charged with condensate

grade water.

During outages

the

system

may

be drained for maintenance,

but

there

has

not

been

any evidence

of accelerated

corrosion or

roughness.

The Core Spray System cannot

be directly aligned to

a

raw water

system

to allow the introduction of water

not

chemically and/or mechanically treated.

Thus,

TVA's original position

and methodology in the pressure

drop calculations

are appropriate.

The inspector

reviewed the information provided

by the licensee

and

reviewed the effects of roughness

on head loss.

The crane technical

paper

410; flow of fluid through valves, fittings and pipe

was used

for this

review.

The effects

of pipe wall

roughness

in the

calculation

is relatively small

when

compared

to. other

head

loss

factors.

The overall effect of roughness

is only 2 - 3 percent

and

if a slightly higher relative roughness

ration

E/D is used it would

have little effect.

Additionally,

GE indicated that the

20 percent

head

loss

value

used

on the

process

drawing

was for pump sizing

calculations

and

should

not

be interpreted

as

being

a

surface

degradation

requirement

on this system.

30

(CLOSED)

IFI

259,

260,

296/89-49-01,

Maintenance

Activities

Involving Rework, Repair, Use-As-Is.

This item was identified during

a significant review of maintenance

activities.

The activity was originally associated

with insulation

required for freeze

protection

and

was

expanded

to include other

activities.

The concern

involved maintenance

personnel

restoring

a

system

impacted

by significant maintenance

to a configuration not in

keeping with design.

The inspector

observed activities,

reviewed

records,

and

spoke

with several

licensee

personnel.

From these

observations

and

reviews,

the

inspector

determined

that

when

maintenance

personnel

are restoring

a system following significant

maintenance

no

unauthorized

parts,

circuit modifications

or

insulation

can

be installed without a deliberate

act of not following

procedures.

In the case of spare parts,

no shop spares

are utilized

and this

includes

fasteners.

All parts

are

maintained

in the

warehouse

and

have

PEG involvement when

a replacement

is required.

(CLOSED) IFI 260/90-29-02,

Leaking Drywell Penetrations.

During

a final walkdown of the

RWCU system

two drywell penetration

was

observed

to

have

been

leaking.

The inspector

reviewed

the

licensee's

closure

package for this item.

Work order 902266900

and

902266800

were written to inspect,

clean,

and repaint penetration

2

X

15

and

2

X 28.

Both of the penetrations

were

checked for leaks

during the ILRT.

No leaks were identified.

(CLOSED) IFI 260/90-36-01,

Inadequate

Vendor Source

Inspection.

This item was identified by the inspector of record

when

a review of

source

surveillance

records

disclosed

that

TVA contract

no.

91NJG-82408C,

the materials

and

procurement

quality surveillance

report

W43901107-219,

dated

October

25,

1990,

stated

that

the

manufacturer's

conformance

to process

and materials of construction

could not be verified because

the items

under contract

were

made at

another

manufacturing plant.

The items involved were identified as

non gA, pressure controller parts (flapper and nozzle)

procured

under

purchase

order number 25-13360 from Controls Southern

Inc.

The parts

manufacturer

was identified

as

Fisher Service

Company of Columbia,

South

Carolina,

and

the

place

of manufacture

was

one of this

company's

plants

located in Marshall

Town,

Iowa.

The problems

were

created

when

TVA's vendor

inspector

failed to

communicated

his

inability to verify the surveillance

plan attributes with management

and engineering.

This communications

breakdown allowed the material

to

be approved for shipment to the site.

Following the disclosure

and the subsequent

issuance

of CARR no.

BFP900375,

dated

December

12,

1990,

the material

in question

was

surplused

by the

licensee.

Corrective actions

taken to prevent recurrence

included instructions

31

with emphasis

on coordinating

any exceptions

to specific surveillance

activities with managers

and engineering.

Additionally, instructions

were issued to document exceptions

on both the shipping release

forms

and surveillance

reports

along with identifying who concurred with

the exceptions.

Finally, to reinforce the above effort, the manager

of materials

and procurement,

by memorandum

dated

December

21,

1990,

reminded all employees

in this group of the

need to implement these

corrective measures.

(CLOSED)

IFI

259,

260,

296/91-02-03,

Possible

Single

Failure

Criteria Identified with SBGT.

The inspector

documented

the results

of

a test

performed

by the

licensee

in this report.

The test

indicated that with all three

trains of

SBGT initiated and

one of the train fans failed to start

the system will still meet the minimum flow requirements

as stated

in

the

FSAR.

(CLOSED)

URI 259, 260/89-56-02,

Use of Closed

Manual

Valves in

EECW

Line Control

Bay Chi llers.

This

item

was identified in. the

course

of observing

check

valve

maintenance

activities.

During an intrusive inspection,

TVA found

that

two check

valves,67-652

and

67-653

had

been

stuck

open.

Following discussions

with the cognizant

TVA system engineer

and

by

reviewing

FSAR Section

10. 10.3

and the applicable flow diagram

no.

1-47E859-1,

the

NRC inspection,

noted

an apparent

discrepancy

between

the

FSAR's

description

of the

EECW

system

and its operational

alignment.

In addition,

the

NRC

team questioned

whether the

two

check

valves

were

necessary

by design

conditions

since

they were

installed

in tandem

and

located

immediately

downstream

of manual

isolation valve 67-651 which is normally closed.

In response

to the stated

concerns,

TVA agreed that the

FSAR does not

fully discuss if the

EECW isolation valves which affect the supply of

cooling water to the control

bay chillers

have automatic actuation

and indicated that Amendment

8 to the

FSAR will include

a revised

and

correct description of the

EECW system.

In reference

to the team's

second

concern,

TVA indicated that the

check

valve

configuration

is

correct.

In their

present

configuration,

the

subject

valves

by design

maintain

associated

header

separation

when the manual

isolation valves

are

opened (i.e.

supplying

EECW to the control

bay chillers).

TVA indicated that

these

valves

prevent

backflow when

one of the associated

headers

experience

a

loss of flow condition.

With regards

to the

code

requirement

that these

valves

be exercised

once every three

months,

TVA requested

and

was

granted relief from this requirement for six

check valves

including the

two subject valves.

As an alternate

to

the

Code requirement,

TVA will disassemble

a minimum of two check

valves at each refueling outage

on

a rotational basis,

to verify its

32

backseating

capability.

Should

the

disassembled

valves fail to

function properly, all

other

valves

in this

category will be

disassembled

and examined.

These

valves will be full stroked in the

open position quarterly

as required.

(CLOSED)

URI 259, 260, 296/90-29-04,

Deletion of ECNs/DCNs.

This

item

was that

ECNs/DCNs

were

being

deleted

from the Unit 2

restart list without proper review by the

CCB or middle management.

Eight of the

commitments

were listed

as

NRC

commitments.

The

inspector

reviewed the licensee's

closure

package for this item.

The

licensee's

evaluation

determined

that in none of the

cases

was

an

actual

NRC commitment deleted.

Five were determined

to not involve

NRC commitments,

three

items were rescheduled.

The categorization

as

a

NRC commitment

by the licensee is done

by the

DCN initiator.

This

initial categorization

is superseded

by the

CCB Subcommittee/Restart

Review

Subcommittee.

The

inspector

reviewed

each

of the eight

DCNs/ECNs.

One

example for ECN P0422,

was that the portion of'he

modification affecting Unit

2

was

complete.

The deferral

was for

Units I and

3 modification which continue

to

be tracked

as

NRC

commitments.

Based

on the review of the eight items

the inspector

concluded

there

was

a logical basis for disposition of each

item and

the program

was adequately

implemented.

(CLOSED)

259,

260,

296/90-33-04,

Hardware Activities Delayed

but not Approved by Senior management.

This item was originally identified during reviews of the licensee's

SPOC process.

The inspector

observed

and reviewed various activities

involving system

components

which were

being

delayed

and did not

appear

to have senior

TVA management

approval.

The licensee's

final

determination

as to system operability by TS required that

a system

plus its

attending

components

must

be

to satisfy

TS.

Regardless

of what sequence

is

used

to correct hardware

items, all

must

be completed prior to declaring

a system operable

(CLOSED)

259,

260/90-40-01,

Deficiencies

Identified

During

Integrated

ESF Testing.

This item

was identified

when

the Units 1/2

DGs

A and

D did not

accept the

ECCS loads

as required

by the applicable SI.

The licensee

traced

the

problem to

a cell switch that did not make

up properly.

This switch failure was determined

to be an adjustment

problem.

The

licensee

installed

a modification in the Units I/2

DGs

as

discussed

in the closure of LER 259/91-01,

in this report.

The licensee

did

not install

the modification in the Unit 3

DGs.

Consequently

296/90-40-01 will remain

open

pending restart of Unit 3.

(CLOSED)

260/90-40-06,

Reliability

and

Human

Factors

Concerns.

33

This

item

was

opened

to follow-up on the licensee's

actions

to

resolve

issues

identified by the

NRR staff audit of the

BFN Unit 2

conducted

in November,

1990.

These

issues

are

addressed

in

detail

and closed

in this report.

This

open

item also included

two additional control

room indication

issues

identified in

IR 90-40.

During this reporting period,

an

inspector

reviewed

the

licensee

actions

taken

to close

the

two

issues.

The issues

and actions

taken were

as follows:

1)

Issue:

Suppression

chamber

water

level

instruments

2-

LI-64-54A and 2-LI-64-66, with

a

range of negative

25 to

positive

25 inches,

lacked indications of negative values.

Action:

The scales for these

instruments

were changed

out with

appropriate

scales

showing positive and negative values.

2)

Issue:

B

channel

recorder

XR-64-159,

Suppression

Chamber

Water Level

and Drywell Pressure,

had

no units designation

label.

Action:

Unit designation

labels for feet

and psig were

added

as operator aid number 2-91-11.

No deficiencies

were identified during the inspector's

review.

The

inspector

concluded

that

no violation of

NRC requirements

had

occurred

and the licensee

had resolved

the issues

involved.

(CLOSED)

VIO

260/90-18-04,

Failure

to

Control

Modifications

Activities.

On

May 29,

1990,

a Dresser

Coupling failed

on an

18 inch diameter

section of piping in the 3B/3D

RHRSW Tunnel.

The failed coupling was

on the

North Supply

Header prior to the Unit 3 Reactor

Building

penetration.

The failure occurred

during modification work on

a

penetration

support

and

was

the direct result of failure

by

modifications

personnel

to

obtain

permission

from

the

Shift

Operations

Supervisor to work on this section of piping while it was

in service.

SDSP-7.9,

Integrated

Schedule

and

Work Control, Section

6.4.1,

required that prior to commencing

work activities that

have

the

potential

for affecting

equipment

operation,

that

a

Plant

Operation

Impact Evaluation

Sheet

be filled out.

In this case

the

Plant Operation

Impact Evaluation Sheet

had

been

approved for work on

the

south

EECW Header

only.

Upon removing bolting from the nearby

support

the piping separated

at the coupling.

When the coupling

failed large

amounts of water discharged

into the piping tunnel with

personnel

in the tunnel.

The inspector

reviewed the licensee's

response

to the violation dated

August

13,

1990.

In that

response

the

licensee

attributed

the

violation to personnel

error, i.e. failure of modification personnel

to recognize

that authorization

to implement the workplan

had

been

denied.

As corrective action the licensee

took disciplinary action

against

the craft

foreman.

Additionally, based

on

reviews of

calculations

of piping and support stress

levels experienced

during

the event,

TYA determined

that support

loads did not exceed

design

allowables.

Pipe stresses

were within code allowables

except for a

weldlet attachment

of

a three

inch diameter

pipe to the

18 inch

header.

The inspector

reviewed Final

Event Report I-I-B-90-061 which provided

the licensee's

investigation of the event.

The inspector

noted that

the licensee

had adequately

determined

the root cause of the failure

and

took appropriate

corrective

action

as

the result

of this

investigation.

The damaged

three inch weldlet was replaced

under

DCN

W7630B

as

implemented

under

WP 0458-90 which was completed

on August

25,

1990.

Additionally, during the original inspection the inspector identified

concerns

about the operability of the retention pins

and the adequacy

of the sequencing

criteria for axial

support

removal.

During the

licensee's

evaluation

of the

event it was

determined

that

the

retention

pins

had

been

operable prior to the event

and

had

been

sheared

when the coupling failed.

Since

the pins in question

are

intended

to maintain

the coupling position

on the pipe and are not

designed

to hold the piping together

under the thrust loads

as were

generated

from this event

the inspector

agrees

with the licensee's

determination.

The licensee

revised

the sequencing

instructions

on

Drawing 47B435-22 to clarify requirements

for modification sequencing

of axial support

removal.

This adequately

addresses

the inspectors

two concern.

Based

on the above review the inspector

determined that the licensee

had

performed

adequate

corrective

actions

which should

prevent

recurrence.

(CLOSED)

VIO 259,

260,

296/90-29-01,

Failure

to Control

Work

Activities.

During

a review of licensee

controls for work activities

on plant

equipment

the

inspector

identified

two examples

of failures

to

adequately

control

work within the

requirements

specified

in the

existing clearance

program.

Subsequent

to issuance

of this violation

the

inspector

identified

two

new events

which were

included

as

additional

examples

to this violation.

These

new examples

were

discussed

in IR 90-33.

The

inspector

reviewed

the licensee's

responses

to the violation

dated

December

28,

1990,

and January

31,

1991.

In those

responses

the licensee attributed the violation to personnel

errors rather than

the result of programmatic

deficiencies

in the hold order process.

This determination

was confirmed by an independent

review conducted

by the onsite

gA organization.

As corrective action the licensee

I

~

35

committed

to conduct sensitivity training of associated

personnel

(system

engineers,

operations,

maintenance,

and

modifications

personnel)

and

strengthen

the training

program for personnel

that

routinely hold clearances.

Additionally the licensee

committed to

completion of

a

HPES evaluation

of these

events.

Any corrective

actions identified by this evaluation

would also

be completed.

The inspector

agreed

with the

licensee's

determination

that

the

failures

were

not

due to programmatic

deficiencies,

i.e. that the

hold order

process

was basically

sound

and that the failures

were

failures to properly implement existing requirements.

The inspector

reviewed

records

to verify the completion of the required training.

Additionally the inspector

reviewed

the

completed

HPES evaluations

and associated

recommendations.

The inspector

determined

that the

completed

corrective

actions

should

be

adequate

to

preclude

recurrence.

n.

(CLOSED)

VIO 259,

260,

296/90-33-01,

Failure to Nake

10 CFR 50.72

and 50.73 Reports.

This VIO was issued for two examples of the failure to notify the

NRC

of unplanned

ESF actuations

and

one example of the failure to submit

a

LER on

a condition prohibited

by TS.

Example

A involved the

failure to notify the

NRC within

4

hours

of

an

unplanned

isolation of the

RWCU system

which occurred

on October

20,

1990.

Example

B involved the failure to notify the

NRC within 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months of an

unplanned

ESF isolation of the Refuel

Zone ventilation system

which

occurred

on

November

4,

1990.

Example

C involved the failure to

submit

a

LER to the

NRC within 30 days for the failure to maintain

a

TS required fire watch.

The licensee

responded

to this YIO by letter on January

16,

1991.

In

the response,

the licensee

denied

example

8 because

the isolation was

given

from

a pressure

instrument

which

was

not part of the

initiation signal.

After review of the circumstances,

the

NRC staff

accepted

the licensee's

denial of example

B by letter on February 28,

1991.

The staff also concluded that the responses

to examples

A and

C were acceptable.

An inspector

reviewed the licensee's

response

and closure

package for

this VIO.

Following issuance

of the violation, the licensee

made the

required

10 CFR 50.72 notification for example

A of the violation and

issued

LERs for

examples

A

and

C

(296/90-04

and

259/90-21,

respectively).

LER 296/90-04

was

closed

in

IR 90-40

and

LER 259/90-21

was

closed

in

IR 91-06.

No further

concerns

were

identified during the review of the licensee's

actions.

t

14.

Condition Adverse to gquality Reports

The inspector

reviewed

3 CARR's concerning welding activities to determine

if any hardware

problems existed in the plant.

36

a.

CARR 900317, Failure to Perform Fitup Verification, Unit 2

This item pertains

to inspection

and control of welding activities

associated

with modification of Unit

2 drywell

Non-ASME structural

steel.

The item was identified during the conduct of a

gA audit of

welding activities

in September

1990.

At that

time

a

concerned

individual informed the

TVA auditors that

some iron worker foreman,

were not entering

the drywell area to physically inspect weld fit-up.

The findings of

a followup investigation

disclosed,

that

between

December

11,

1989,

and January

19,

1990,

a modification foreman

had

signed fitups verification for twenty safety related structural

steel

welds all located

inside the drywell,

on dates

when entry records

~ showed that

he

had not entered

the Unit 2 drywell.

Procedure

SDSP

13.8, Welding Surveillance,

Section 6. 1, directs the craft foreman to

perform weld fit-up inspections

for instances

where certain

TVA

specifications

and

other

nuclear

construction

standards

were

referenced

by design output documents.

A fit-up Inspection

includes

verification of:

Weld joint dimensions (fit-up)

Proper material

Welder qualification

Proper welding procedure

Surface cleanliness

Alignment

Preheat

temperature

Tack welds

Following discussions

with cognizant

personnel

and

a comprehensive

records

review the inspector ascertained

the following:

Because

the

concern

originally implicated three

iron worker

foremen,

TVA modification

management

determined

that

a full

investigation

of all

iron worker

foremen

was

necessary

to

determine

the generic

aspects

of the issue.

The method

TVA used

to

pursue

this

task

included;

( 1) identification of all

work-plans inside

the drywell implemented

by iron workers;

(2)

development of an accurate list of iron worker foremen

who were

actively involved in these work-plans;

(3) for every individual

identified on the list, radiation control

was asked to generate

a printout

showing all entries

to the

drywell

dur ing the

time-frame of interest.

Welds

made

in the modifications'ab

shops

and welders

who fabricated

them were excluded

from the

investigation.

Corporate

NDE personnel

provided

assistance

in the investigation.

The records

showed that out of the

20 field welds in question,

four

had

been

selected

for inspection

of fit-up by

gC

as part of the

random

inspection

program for these class/type

welds..

In addition,

the records

show that following welding, all

20 of the subject welds

received

and

passed

a final visual inspection that

was

performed

by

I'

37

gC under

programmatic

requirements

and applicable

procedures.

To

demonstrate

that

even

though

the

iron worker

foreman failed to

perform the administrative

task of conducting

a fit-up inspection

prior to welding

TVA took the position that all of the attributes

associated

with this inspection

were verified at

some

time during

fabricating

of

these

welds.

In

support

of this

position,

documentation

in the

CARR package

stated

the following:

For each of the subject

20 welds, three of the preweld fit-up

activities/attributes

referenced

above,

were

checked

and

documented

by a foreman

on

a weld data

card.

These activities

were

performed

as

the

foreman

assigned

(a) certified welders

with current qualifications,

(b) proper detail

weld procedure

and (c) proper filler metal.

These

three activities

were also

verified

as

acceptable

during final visual

examination

by

quality control.

Three additional

preweld activities/attributes

which required

foreman

verification were:

proper material

alignment

weld 'joint dimensions (fit-up including root gap)

These

activities

were

performed

by

the

designated

welder

and

documented

on the appropriate

form.

One of the attributes

that

could

not

be verified during final

inspection

by quality control

was

preheat.

Having the correct

preheat,

as required

by the applicable

weld procedure,

is incumbent

upon

the designated

qualified welder.

The foreman administratively

checks

that

the correct

preheat

was

achieved

prior to welding.

However,

by specification

and

on the basis of material

size

and type,

none of the structural

steel

welds under discussion

required preheat.

Therefore,

the failure by the foreman to check the preheat prior to

welding would have

no impact

on the integrity of the subject welds.

Another of the attributes that could not be verified visually during

final inspection

by gC was the fit-up gap.

This is another of the

attributes

whose verification is incumbent

on the designated

welder

as

part of his responsibility

for good workmanship of his welds.

Also, it is

one of the attributes

that the

foreman

must

check to

assure

that

the correct

gap

was

achieved

prior to welding.

By

procedure,

the welder

measures

and

documents

on the weld history

record the root gap dimension.

This was

done for each of the subject

welds,

knowing the correct

gaps size allows

gC to verify correct weld

size,

since

the leg of

a fillet weld is dependent

on the root gap

size.

As stated

ear lier,

the

in question

were usually

inspected

and

found acceptable

therefore, it would appear

that the

fai lure of the iron workers'oreman

to perform the required

weld

fit-up inspection

would not have

an adverse

impact on weld integrity.

r

~

I

0,

38

In the broader

scope of this investigation,

gC performed

a review of

weld

records

generated

by the modification of drywell structural

steel.

The

review of the

records

showed

that

these

had

received

a final visual

examination

by gC and were found acceptable.

The records

also

showed that,

as part of the

on going structural

steel

welding program,

gC performed preweld checks

on

a random sample

of

1265

welds,

selected

from

a total

population

of 3328

fabricated

from November

1989 to April 1990 regardless

of welder

foremen.

Out of this

sample,

approximately

38Ã of the total

population,

TYA's review identified

39 unacceptable

hardware

and

software

conditions

for

a

3Ã rejection

rate.

The

unacceptable

conditions identified have

been corrected.

To explore

the

generic

aspects

of this matter,

records

of

gC

inspections

performed

between

August

1989

and

August

1990

were

reviewed

by TVA with emphasis

placed

on all the foremen

who worked

on

modifications of structural

steel

welds inside Unit 2 drywell.

The

records

review included

inspection

performed

on structural

steel

welds in other areas

of the plant where

foremen

under investigation

were involved.

Out of

a total population of 3236 welds,

792 weld

records

were

reviewed

by

gC

who discovered

that

none of the welds

were

found unsatisfactory.

. The

records

showed

that

26