ML17310B147

eddy current testing of steam generator

2-1

and was continuing

to test

2-2 at the

end of the period

(see Section

7. 1).

The

licensee

had identified

a significant number of axial crack indications in

2-2.

1.3

Unit 3

Unit '3 operated

throughout the inspection period at essentially

85 percent

power.

On January

19,

1994,

the licensee

gagged

closed

safety valve which had developed

a small seat leak.

Gagging

one safety valve

reduced

the Technical Specification

maximum allowable reactor

power to 98.2

percent.

I

t

l

1

J

ONSITE RESPONSE

TO EVENTS (93702)

2. 1

Hain Steam Isolation Valve Fast Closure

Unit

1

On January

20,

1994, during

a surveillance test to partial stroke the Hain

Steam Isolation Valves (HSIV), HSIV-170 fast-stroked full closed

and

immediately fast-stroked full open.

The plant responded

as follows:

~

Reactor coolant temperature

increased

about

one degree.

~

Primary system pressure

did not change.

~

The affected

pressure

increased

about

20 psig.

The licensee

declared

the "A" train of the hydraulic system for HSIV-170

inoperable.

HSIV-170 remained

operable

since the "8" train of the hydraulic

system

was still available to operate

the valve.

The licensee

determined

the

cause of the event

was

a failure of the "C" solenoid valve in the "A" train.

This aligned hydraulic accumulator,

instead of the hydraulic pump, to the HSIV

operator

and fast-closed

the valve.

The solenoid valve was replaced

and

a

partial stroke test

was satisfactorily completed.

The inspector

reviewed the licensee's

troubleshooting of HSIV-170 and

concluded that the licensee's

response

to this event

was appropriate.

3

OPERATIONAL SAFETY VERIFICATION (71707)

The inspectors

performed this inspection to ensure that the licensee

operated

the facility safely

and in conformance with license

and regulatory

requirements

and that the licensee's

management

control

systems effectively

discharged

the licensee's

responsibilities for safe operation.

The methods

used to perform this inspection

included direct observation of

activities

and equipment,

observation of control

room operations,

tours of the

facility, interviews

and discussions

with licensee

personnel,

independent

verification of safety

system status

and Technical Specifications limiting

conditions for operation, verification of corrective 'actions,

and review of

facility records.

3. 1

Worker Removed Dosimetr

and Securit

Bad

e in Radiolo ical Controls

Area Protected

Area

Unit 2

On January

25,

1994, the inspector noted that

an individual was working inside

the protected

area

and within the radiological controlled area

(RCA) and was

not wearing his automated

controlled access

device

(ACAD) or dosimetry.

The

individual was

a contract worker involved in steam generator

chemical

cleaning

operations

(see Section 7.1).

The inspector

observed that the

ACAD and

dosimetry

was

on top of a nearby transformer,

approximately five feet from the

worker.

When the worker noticed the inspector looking at the

ACAD and

dosimetry

on the transformer,

he properly attached it to his body.

The worker

was working in the

pump trailer used for steam generator

chemical

cleaning.

The radiation levels in the trailer were subject to change

due to various

chemical

solutions

being

pumped through the trailer.

The inspector notified

l

t

l

the worker's supervisor of this observation.

The supervisor

subsequently

informed radiation protection

(RP)

and security.

Procedure

75AC-9RP01,

Revision 2, "Radiation Exposure

and Access Control,"

Step 3.2.3, states,

in part, that personnel will be issued dosimetry which

shall

be worn at all times within the

RCA.

Step 3.2.4.1 states,

in part, that

dosimetry shall normally be worn on the front of the body between the thigh

and head,

unless directed otherwise

by RP.

Procedure

20AC-OSK04, Revision 9,

"Protected/Vital

Area Personnel

Access Control," Step 3.2. 1, states,

in part,

that

ACADs shall

be displayed

by all .individuals while inside Protected/Vital

areas

and shall

be positioned

on the front of the outermost

garment,

between

the neck

and the waist,

photograph

side out.

The failure of the employee to

follow procedures

is

a violation of Technical Specification 6.8.1 (Violation

50-529/94-02-01).

The licensee initiated Condition Report/Disposition

Request

(CRDR) 2-4-0041 to

evaluate this problem.

The licensee

also performed

an exposure

evaluation of

the worker.

The worker's

TLD was read

and the area

he had worked in was

surveyed.

Based

on the survey,

a maximum unmonitored

exposure of 0.003

mrem

could have

been received

by worker.

The

RP Department

denied the employee

further access

to the

RCA.

In addition, the Security Department

performed

a

check to verify that unauthorized

use of the worker's

ACAD had not occurred

and that the

ACAD had not left the worker's sight.

The Security Department

also

removed the employee's

unescorted

access

to the protected

area.

Further,

the contractor for chemical

cleaning held training for all of its employees

to

re-emphasize

the licensee's

expectations

in security, radiation protection,

and job performance.

The inspector

noted that the licensee's

response

to this event

was thorough.

In particular,

the inspector noted that the licensee

took prompt action to

assess

the incident

and took action when it was brought to the attention of

first line supervision

by the inspector.

3.2

Dosimetr

Securit

Bad

e Placement

On February

7,

1994, the inspector

observed

a licensee

maintenance

individual

entering the Unit 3 radiologically controlled area

(RCA) wearing the

dosimetry/security

badge (i.e.,

ACAD) hanging off the right front pant's

pocket.

The dosimetry at Palo Verde is attached

to the security badge.

The

individual moved the

badge to the torso after the inspector questioned

a

radiation protection technician whether this was the correct location for the

dosimetry.

Later, the inspector

noted that there were other personnel

wearing

their dosimetry in similar locations inside the

RCA.

As noted in Section 3. 1 of this report,

the licensee

procedures

required that

dosimetry shall normally be worn on the front of the body between

the thigh

and

head

unless

otherwise specified

by Radiation Protection

(RP).

The

licensee's

procedure for personnel

access

control required that the security

badge

be displayed

on the front of the outermost

garment,

between

the neck and

'the waist.

Additionally, the procedure

required that all personnel

report any

unbadged

personnel

in the protected/vital

area.

Although the badge location

probably did not affect the measured

whole body dose reading in these

cases,

s

I

.

)

the inspector

noted that the improper location of the badge

could hamper

proper identification by security.

The licensee

acknowledged

the problem

and committed to have all managers

and

supervisors

review with their employees

the proper location for the dosimetry

and security badge.

Additionally, the licensee

noted that they would review

the differences

in the requirements for security

badge

placement

and dosimetry

placement to determine if a change

was needed.

Finally, the licensee

committed to review employee training/retraining to determine if the location

was properly defined.

The inspector .considered

these corrective actions to be

adequate.

3.3

Plant Material Condition

3.3.1 Unit 2 Hi

h Pressure

Safet

In 'ection

Pum

HPSI

During a routine plant tour, the inspector

observed that large boric acid

formations

had formed at both ends of the

HPSI

2A pump.

The acid formation

was apparently

due to pump seal

leakage.

The licensee

responded

by issuing

a

work order

and

had the

pump cleaned.

The licensee

stated that the seals

were

scheduled for replacement

in the next refueling outage

and every subsequent

third refueling outage.

The inspector reviewed the Updated Final Safety Analysis Report

(UFSAR)

and

determined that Section 6.3. 1.3.N.l.a states,

in part, that the maximum HPSI

pump seal

leakage

allowed is 100 cc/hr.

Although unable to determine

the

extent of the

pump seal

leakage

since the

HPSI

pump was in standby

mode,

the

inspector

noted that the licensee

had performed the Technical Specification

surveillance

inspection of emergency

core cooling system

(ECCS) leakage

(TS 4.4.5.2. 1) during the refueling outage in August 1993.

The purpose of this

test

was to verify that there

was less

than

1 gallon per minute leakage of the

ECCS equipment outside containment providing long-term,

post loss-of-coolant-

accident recirculation.

The seals

had

no identified leakage during the

performance of this test.

Based

on the observation of a number of other boric acid leaks in

ECCS

pumps

and valves

and the fact that the licensee

did not attempt to quantify leaks

as

they developed,

the inspector questioned

the basis of the

UFSAR pump seal

leakage

requirement

and whether the

TS surveillance

was adequate

to verify

compliance with the requirement.

The inspector will review the licensee's

response

to these

questions

during

a future inspection

(Followup item 50-

529/94-02-02).

3.3.2 Unit

1 Malkdown

During

a routine tour of Unit 1, the inspector identified

a small

steam leak

from a drain valve in the supply line to the steam-driven,

auxiliary feedwater

pump

and packing leaks from both condensate

transfer

pumps.

The inspector

reported

these conditions to the shift supervisor

who was not aware of the

deficiencies.

Unit

1 mechanical

maintenance

personnel

evaluated

the

conditions

and determined that the steam leak was unisolable

and that

a leak

repair

had

been previously attempted.

A new work order was written to attempt

1

another leak repair.

The inspector noted that

a work order

had

been

previously written to adjust the packing

on one of the condensate

transfer

pumps.

However, the other packing leak had not been identified and

a work

order was written to repack the

pump.

At the

end of the inspection period,

the packing

was replaced

and the steam leak and packing adjustment

on the

other

pump were included in the 12-week work schedule.

3.3.3 Conclusions

In addition to the conditions discussed

in the above sections,

the inspectors

noted several

other minor material condition problems during routine

walkdowns,

such

as boric acid leaks in pump

and valve packings

and lube oil

leaks

on pump bearings.

Upon followup, the inspectors

found that several

of

these conditions

had not been previously identified by the licensee.

The

inspectors

noted that the licensee

does not require that maintenance

other examples of cracking problems.

0

10

l

~

1

I

~

The licensee

developed

an action plan to (1) remove

SSAUV-203 and

send it to a

laboratory for fracture

mode analysis;

(2) remove

SSBUV-200

and conduct on-

site ultrasonic testing

(UT); and,

(3) cut SSBUV-200 into quarters

and perform

visual

and microscopic examinations

following the ultrasonic testing.

The

licensee initially planned to

UT the

15 valves that had not been visually

i'nspected.

However,

when they performed the

UT of SSBUV-200, they were not

able to characterize

the extent of the cracking.

As

a result, they determined

that

UT of the remaining

15 valves would not provide useful information.

By February

14,

SSAUV-203 had not been delivered for inspection to the

independent

laboratory.

The licensee

had decided not to perform destructive

examination of SSBUV-200 until the other valve had

been delivered.

As

a

result,

the cause of the cracking

had not been determined.

The inspector

noted at the exit meeting that while the initial action plan was well

developed,

the progress of this investigation

had

been slow.

Licensee

management

concurred with this assessment

and noted that the valve had

subsequently

been delivered

and that they expected

a more timely resolution of

the issue.

5

SURVEILLANCE OBSERVATION (61726)

The inspectors

reviewed this area to ascertain

that the licensee

conducts

surveillance of safety-significant

systems

and components

in accordance

with

Technical Specifications

and approved

procedures.

5.1

Reactor Protection

S stem

RPS

Res

onse

Time Testin

Unit 2.

On February 3,

1994, the inspector

observed

portions of surveillance test

(ST)

36ST-9SB44,

"RPS Matrix Relays to Reactor Trip Response

Time Testing," in

Unit 2.

For most of the plant protection

system

(PPS)

instruments,

Technical Specification (TS) 3.3. 1 requires that the total channel

response

time to be

less

than

1. 15 seconds.

TS 4.3.1.3 also requires that

one channel

of each

function be tested

at least

once every

18 months.

The total response

time is measured

in the following three steps:

~

Procedure

36ST-9SB41

measured

the process

equipment

response

time [i.e.,

the time from the process transmitter (e.g. pressurizer

pressure)

to the

PPS cabinet].

~

Procedure

36ST-9SB42

measured

the

PPS cabinet

response

time (i.e., the

time from the

PPS cabinet to the matrix relays).

~

Procedure

36ST-9SB44

measured

the response

time from the matrix relays

to the reactor trip breaker opening.

The inspector

concluded that the these surveillance

procedures

adequately

demonstrated

that the overall response

time requirements

of TS 4.3. 1.3 were

satisfied.

Additionally, the inspector

rioted that procedure

36ST-9SB44

was

well written and that the technicians

used

good communications

during the

performance of 36ST-9SB44.

11

I

The inspector reviewed the data collected during the performance of 36ST-9SB41

completed

on Harch 26,

1993;

36ST-9SB42

completed

on August 16,

1993;

and

during the inspector's

observation of 36ST-9SB44

on February 3,

1994.

In

addition, the inspector reviewed calculation

13-JC-SB-02 to determine the

basis for the acceptance

criteria provided in these surveillance tests.

The

purpose of this review was to verify that the overall time response

of the "B"

channel

high and low pressurizer'pressure

trips were within the

TS requirement

of 1. 15 seconds.

Based

on these

reviews the inspector

concluded that, the

TS requirement to

verify the "B" channel

pressurizer

pressure

high and low trip response

time

was satisfied;

the "B" channel

was tested within the 18-month period specified

in TS 4.3. 1.3; the overall process

response

time was satisfied

by meeting the

acceptance

criteria for each portion of the test;

and the calculation for the

acceptance

criteria contained

a large margin of safety

and the actual

response

times were well within the

TS requirement.

5.2

Plant Protection

S stem Functional

Test

Unit

1

On February

9,

1994, during the performance of surveillance test

36ST-9SB04,

"Plant Protection

System

(PPS)

Functional Test-Reactor

Protection

System/Engineered

Safety Features

Actuation System

(ESFAS) Logic," the

technicians identified

a problem which they believed to be in the test

circuit.

The inspector

observed that the technicians

immediately notified the

shift supervisor

and documented

the problem in the test log.

The shift

supervisor

and shift technical

advisor reviewed the test logic diagrams

and

agreed that the most likely cause of the problem was with the test circuit.

However, the shift supervisor

continued to question the technicians

and

contacted

the

I&C supervisor

and system engineer to ensure

a proper evaluation

was performed.

The system engineer

also thought the problem was in the test

logic portion of the circuit.

The

same step

was repeated

5 times

and the

proper response

was obtained.

The remaining portions of 36ST-9SB04

were

satisfactorily completed.

The licensee initiated

CRDR 1-4-0060 to trend the

spurious test circuit anomaly.

The inspector

concluded that the problem was appropriately

documented

and

that'he

shift supervisor aggressively

evaluated

the problem to ensure that the

performance of the

RPS was not affected.

6

DIESEL GENERATOR CONDITION AND TESTING (61726,

62703,

71707)

During the inspection period, the licensee

experienced

unanticipated trips of

an emergency diesel

generator

(EDG) in all three units.

Each of the trips

resulted

from a non-safety related

problem.

In each

case,

the inspectors

assessed

the licensee's

review and plans for corrective actions.

In addition,

this section discusses

a degraded

condition in the Unit

1 "B" EDG jacket water

system that was not thoroughly evaluated

by the licensee.

6. 1

Unit

1

Diesel

Generator

Reverse

Power Tri

Durin

Shutdown

Se

uence

I'n

January

25,

1994,

a reverse

power trip of the Unit

1 "B" EDG occurred while

operators

were securing the

EDG per procedure

410P-1DG02,

"Emergency Diesel

12

!

I

Generator

B."

The

EDG was running to perform the weekly Technical

Specification surveillance test.

The inspector

concluded that the operator

correctly followed the procedure for securing

the

Despite

weaknesses

in

the procedure

and

a lack of sensitivity by the operator concerning

the

potential for a reverse

power trip contributed to the event,

the inspector

c'oncluded that the licensee's

corrective actions

were appropriate.

Procedure

410P-1DG02 directed the operator to lower the generator

output to

less than 0. 1 megawatts

(HW), and then

open the generator

output breaker.

The

operator

lowered power to approximately 0. 1

MW and turned to discuss

the

evolution with a trainee.

When

he turned to trip the output breaker,

the

wattmeter still indicated 0. 1

HW and the

EDG tripped automatically

on reverse

power.

The licensee

inspected

the "B" EDG output breaker

and found no damage to the

breaker or the generator.

An evaluation of the event

was conducted

and

documented

in Condition Report/Disposition

Request

(CRDR) 1-4-0023.

The

licensee

concluded that the following factors

caused

the reserve

power

condition:

~

The operating

procedure directs the

EDG to be unloaded to 0. 1

HW, but

the megawatt meter is calibrated to only +/- 0.2

HW.

Operators

were not aware,

and

had not been trained, that the megawatt

meter in the control

room would show

a positive indication in a reserve

power condition.

The operator delayed tripping the output breaker;

however, there

are

no

precautions

in the procedure to ensure that the output breaker is

immediately opened after the load is reduced to 0.1

HW

Unit

1 operations

management

issued

a night order emphasizing

the need to

immediately trip the output breaker

when the

EDG is unloaded.

The night order

also discussed

the expected

response

of the megawatt meter during

a reverse

power condition.

In addition,

an Instruction

Change

Request

was initiated to

change

the

EDG operating

procedures

in all three units to lower the generator

output to 0.3

HW before opening the output breaker.

The inspector

concluded

that these

actions

appeared

to be appropriate.

6.2

Unit 2

Diesel

Generator Tri s

Due to Control Air S stem Problems

On January

30,

1994, the Unit 2 "B" EDG tripped after

10 seconds

of the

cooldown cycle of a surveillance test.

The licensee

reviewed the trip and

determined that it resulted

from the failure of a check valve in the

pneumatic,

non-safety related,

control air system.

The control air system

provides protective trips to the

EDG governor during maintenance

runs;

however,

these trips are

bypassed

during

a safety start.

The licensee

subsequently

repaired the check valve.

On January

31,

1994, during the post-maintenance

start of the "B" EDG, the

fuel racks closed before the

EDG reached

rated

speed,

and the diesel

13

,

I

subsequently

tripped

on under-frequency.

Although the problem appeared

to be

caused

by the control air system,

the licensee

could not identify the

component

which caused

the failure. It appeared

that

a control air system

solenoid allowed the air to pass

through to the fuel racks and'close

them.

During an emergency

run of the diesel,

two emergency

solenoids

in series

would

have prevented

the trip by preventing the fuel racks'from closing.

The licensee restarted

the

EDG on February

1,

1994, with instrumentation

installed for troubleshooting.

However, the cause of the earlier trip was not

determined.

The licensee

determined 'that

a diesel start failure,. as defined

in the Technical Specifications,

had not occurred since the diesel

would have

started if required during an emergency.

The inspector

agreed with the

licensee

and concluded that

a start failure had not occurred during the

January

31,

1994, start of the

6.3

Unit 3

Diesel

Generator

Overs

eed Tri

Durin

Shutdown

Se

uence

On January

26,

1994, the Unit 3

EDG "A" tripped

on overspeed

during

a post

maintenance

retest.

The diesel

was being retested

following planned

maintenance

and tripped immediately after the operator

depressed

the manual

shutdown button.

Depressing

the manual

shutdown button disengages

the

electrical

governor

and engine

speed

control is taken over by the mechanical

governor.

Engineering

and maintenance

troubleshooting

determined that the

mechanical

governor had'not controlled properly and that the problem was

- probably caused

by air in the governor's hydraulic control lines.

The oil in

'he

governor

had

been

changed previously as part of the planned

maintenance.

The licensee restarted

the

EDG, cycled the speed setting

on the manual

governor several

times,

and then performed

a normal

shutdown.

The inspector

compared

the procedure for changing the governor oil with the

vendor technical

manual.

The inspector concluded that the technical

manual

did not specify

how to change

the governor oil to ensure that the hydraulic

lines were not air bound.

The licensee

informed the inspector that they

intended to incorporate

the cycling of the mechanical

governor

speed control

into the oil change

procedure to prevent future problems with the governor.

The inspector

concluded that this action was appropriate.

6.4

Unit

1

Diesel

Generator Jacket

Water

Ex ansion

Tank Automatic Hake-u

Ca abilit

Disabled Without A

ro riate Review

During a system walkdown of the Unit

1 "B" EDG, the inspector noted that the

jacket water make-up

combined-header

stop valve,

1PDGBV013,

was shut.

There

was

a caution tag

on the valve 'indicating that the valve was closed

due to

leakage

past the solenoid-operated,

auto make-up valve.

With the jacket water

make-up valve closed,

the jacket water expansion

tank level could not be

controlled automatically.

The ability to automatically control level

was

described

in the Updated Final Safety Analysis Report

(UFSAR).

The inspector

reviewed the licensee's

work control processes

to determine

whether the licensee

had conducted

an operability evaluation of this

condition.

The inspector

found that the licensee

had not considered

the

automatic function described

in the

UFSAR.

In addition, the licensee's

14

I

l

I

C

screening

process for conducting operability evaluations of degraded

conditions

was narrowly focused

and did not include conditions

where automatic

functions described

in the current licensing basis

were removed.

The inspector

was concerned that the failure to include

an evaluation of the

r'emoval of automatic functions in the operability screen

process

was

a

significant weakness,

in that safety significant design

changes

to the plant

could be inadvertently performed without an appropriate operability

evaluation.

As discussed

in Generic Letter 91-18,

which distributed

NRC

Inspection

Manual

9900, "Operability", it is important to evaluate

the

physical differences

between

the automatic

and manual

actions to ensure

the

change

does not alter the licensing basis for the plant.

Nevertheless,

the

inspector considered

that the safety, significance of this particular condition

was low because

the

EDG automatic jacket water make-up function was designed

as

an operator

convenience

and

was not needed to ensure

the proper operation

of the

6.4. 1

De raded

Nonconformin

Condition

The inspector determined that the leaking jacket water auto make-up valve was

a degraded

condition

and reviewed the licensee's

program requirements

for

degraded

conditions described

in procedure

02PR-Ogg01,

"Control of Degraded

and Nonconforming Material."

This document stated that

a work request

(WR) or

Condition Report/Disposition

Request

(CRDR) should

be used to report

a

degraded

condition.

The inspector

noted that

WR 862611

was written on

December

10,

1993,

which was the

same date that the auto make-up valve was

found leaking

and caution tagged.

The

WR was subsequently

canceled

on

December

15,

1993,

by the work planner.

The work planner

had initiated

CRDR

1-3-0208 in Narch

1993

based

on repeated

problems with the make-up valve

leaking.

The planner decided to defer the maintenance

and leave the jacket

water auto make-up valve isolated until engineering

completed this evaluation

(the action was

due in April 1994).

The inspector

concluded that the degraded

condition of the auto make-up valve was appropriately reported

and documented.

6.4.2

0 erabilit

Evaluation

The inspector

noted that Appendix

D to procedure

30DP-9WP01

"Work Initiation,"

contained

a potential

impact screening

step

used to determine if a

MR required

additional

review by the shift supervisor.

Although the impact screening for

the jacket water auto make-up valve

WR was conducted

by an operations

evaluator

and

was not forwarded to the shift supervisor for an operability

determination,

the inspector

concluded that the operability decision

was

consistent

with the screening criteria.

The inspector further concluded that the guidelines for screening

WRs for

operability concerns

were narrow.

The guidelines'addressed

only those

situations

where Technical Specifications

(TS)

and associated

Limiting

Conditions for Operations

(LCOs) were obviously impacted

by

a degraded

condition.

In the case of the auto make-up valve, the evaluator

decided that

the loss of the automatic

make-up capability for the jacket water expansion

tank did not impact the operability of the

This decision

was apparently

15

I

1

I

i

based

on isolating the jacket water auto make-up valve in the past

due to

similar problems with the valve leaking.

The inspector discussed

the design basis for the auto make-up capability with

the system engineers.

In July 1993,'during the evaluation of CRDR 1-3-0208,

the .system engineer

documented that the function of the automatic jacket water

make-up

system

was for operator

convenience

and not for design safety or to

minimize the impact of any postulated jacket water system failures.

When the

auto make-up valve was shut in December

1993, the operations

evaluator

was not

aware of the function of the valve described

in the

CRDR evaluation.

The

inspector

concluded that removing the automatic jacket water make-up

system

did not adversely affect the operation of the

6.4.3 Licensee Actions

0

The inspector discussed

the weaknesses

in the operability guidelines with

licensee

management..

The licensee

had formed

a review group to develop

more

comprehensive

guidelines for operability determinations

based

on previous

NRC

concerns

(see

NRC Inspection

Report 50-528/93-12,

Paragraphs

13

and 17.b.).

At the exit meeting,

licensee

management

recognized that this issue

demonstrated

a vulnerability in their degraded

condition review process

and

indicated that the lessons

learned

would be factored into their operability

determination guideline development.

The inspector will continue to follow

the licensee's

progress

in this area during routine inspection.

6.5

Conclusions

At the exit meeting,

the inspector noted that while the problems resulting in

unanticipated

EDG trips and the degraded

condition of the jacket water system

in Unit

1 did not appear to be safety significant, they may be precursors

of

degrading

EDG condition.

In addition, the inspector

noted that

each of the

EDGs appeared

to have

numerous

minor lube oil, fuel oil, and air system leaks.

The inspector recognized

a licensee

management initiative to assess

the

EDG

trips to determine if there were

common problems which could affect

EDG

reliability.

Licensee

management

noted that in recent years there

had

been

increased

emphasis

in reducing

EDG out-of-service times.

This may have raised

the threshold for correcting minor system problems.

They stated that

an

assessment

of EDG maintenance

practices

and their affect of EDG reliability

would be performed.

7

STEAN GENERATOR INSPECTION,

CLEANING AND PLUGGING

7. 1

Steam Generator

Edd

Current Testin

73753

7. 1. 1 Back round

and Pur ose

During this inspection,

the licensee

conducted

extensive

eddy current

inspections of the Unit 2 steam generators.

The licensee's

inspections

were

being performed to comply with commitments

made in their letter to the

NRC,

dated July 18,

1993.

The purpose of this inspection

was to determine if the

licensee

and licensee

contractors

had

been performing inspections,

data

0

16

analysis,

and inspection

scope

changes

in accordance

with licensee

procedures

and commitments.

7.1.2 Procedures

The inspector reviewed licensee

procedure

73TI-9RCOl,

"Steam Generator

Eddy

Current Examinations,"

Revision

10, dated January

12,

1994.

The procedure

was

reviewed to determine if requirements

for bobbin coil and motorized-rotating

pancake coil

(HRPC) eddy current data analysis

and evaluation

had

been

defined.

The inspector

reviewed the .procedure

to assess

the licensee's

criteria for equipment calibration

and bobbin coil and

HRPC data discrepancy

resolution.

The procedure

was also reviewed to determine if the flaw

indications,

which were expected

by the licensee

in certain

areas of the steam

generator,

had

been identified.

The inspector

found that specific requirements

for bobbin coil and motorized-

rotating pancake coil eddy current data analysis

.and discrepancy resolution

had been

adequately

defined in the procedure.

The inspector also found that

equipment calibration

and particular types of flaw indications for each

section of each

area

had

been

adequately

defined in the

procedure.

The inspector

concluded that licensee

procedure

73TI-9RC01 included

requirements for bobbin coil

and motorized-rotating

pancake coil

(HRPC) eddy

current data analysis

and evaluation.

The inspector concluded that the

procedure

also included equipment calibration criteria,

bobbin coil and

HRPC

data discrepancy resolution criteria,

and descriptions of particular flaw

indications expected to be found in certain

areas of the steam generator

tubes.

7.1.4 Observations

The inspector

observed

licensee activities at four HRPC and two bobbin coil

data acquisition stations to determine if the licensee

had

been performing

and

recording data in accordance

with licensee

procedures.

The inspector also

reviewed

eddy current test

equipment calibration records.

The inspector

found

that the licensee

had

been performing the bobbin coil and

HRPC eddy current

tube examinations

in accordance

with licensee

procedure.

The

inspector also found that the licensee

had

been recording data

on approved

data sheets

and that eddy current inspection

equipment calibrations

had

been

performed

and were being checked

in accordance

with procedure.

The inspector also observed activities at four data analysis stations.

The

inspector

found that each of the four stations

had

a current technique

sheet.

The inspector noted that the eddy current operator

was utilizing frequencies

and mixes specified

on the technique

sheet for eddy current testing analysis

.in accordance

with procedure

73TI-9RC01.

Therefore,

the inspector

concluded

that licensee

personnel

had

been performing steam generator

tube eddy current

inspections,

data recording,

and data analysis

in accordance

with licensee

procedures.

17

I

Cal ibrati on

On January

31,

1994, during the licensee's

gA review of eddy current testing

work, the licensee identified that

4 of 36 eddy current calibration groups

exceeded

the four-hour limit for calibration verification.

This is

a

requirement of Procedure

73TI-9RCO1,

Paragraph

8.3.3.

The licensee's

root-

cause

analysis identified several

reasons

that the calibrations

were not done:

~

In some cases,

the operator

was not examining

any tubes

when the four-

hour calibration requirement

was required.

In some cases,

the operator

was performing activities without problems

and simply lost track of the time requirement.

There

was

no mechanism

established

to remind the operator

when the 4-hour limit was

approaching.

In some cases,

as the four-hour limit approached,

the operator

had

problems getting the calibration completed.

The inspector noted that

calibration can often

be difficult to obtain.

The licensee's

corrective actions

included conducting

a stand-down

meeting of

eddy current personnel,

the use of a preset timing device to alert the

operator

when the four-hour time period is going to expire,

and having the

primary and secondary

data analysts

make

a report noting the beginning

calibration time and the ending calibration time.

This finding was licensee-

identified and appropriate corrective actions

were taken.

Stuck

HRPC Probe

During this inspection period,

a

HRPC probe

became

lodged in

tube.

This probe

was removed

by running another

probe in from the cold leg

and dislodging the struck'probe.

The root cause of this incident was that the

operator

was given

an incorrect tube number.

The programmer incorrectly

entered

the tube number, resulting in the acquisition operator probing the

. wrong tube.

The operator

was moving the

HRPC probe quickly to the desired

elevation;

however,

the probe

was inserted into the tube's

U-bend well before

that desired level (for the correct tube)

because

he was provided with the

wrong tube number.

As

a result,

the probe

became

lodged in the U-bend.

The licensee's

root-cause

evaluation identified human error in two areas.

First, the programmer entered

the wrong tube number,

and second

the person

responsible for checking that tube list against

the original list, failed to

note that

an incorrect tube

number

had

been entered.

Corrective actions

taken

included writing a procedure for programming

special

interest tubes,

such

as

this one.

Personnel

in data

management

had

a stand

down meeting to address

attention to details

and review the procedures for programming tubes.

7. 1.3

Sam lin

Hethodolo

and

Ex ansion Criteria

The inspector reviewed the licensee's

Unit 2 steam generator

sampling

methodology inspection plan.

The inspection plan was reviewed to determine if

18

e

i

'i

r

I

)

the licensee

included inspection

expansion criteria, historical

steam

generator

data from licensee

inspections

and other utilities, and loose part

monitoring.

The inspector

found that the licensee's

Unit 2 steam generator

(SG) sampling

plan included criteria for expanding the inspection

scope.

The inspector

noted that the licensee's

inspection

plan included:

~

a full tube length bobbin coil inspection of 4200 of the approximately

12,000 tubes

in the two Unit 2 steam generators,

an

MRPC inspection at the top of the tubesheet for 2283 (in

SG 2-1)

and

2203 (in

SG 2-2) of the hot leg tubes,

and

approximately

1800 tubes in the upper bundle area (in the arc of

interest that analysis

indicates to be most susceptible

to free span

cracking)

from the first vertical support to the

08H support.

The inspector also found that the licensee's

inspection

plan included previous

tubing indications

found in Palo Verde steam generators

and problems

identified by other utilities, vendors,

and the

NRC.

The inspector

concluded that the sampling methodology

used to develop the

inspection

plan included the areas

necessary

to provide information needed to

ascertain

the condition of the steam generators

tubes.

The inspector

found

the sample

expansion criteria used to be satisfactory.

7.1.5 Ins ector

and Anal st's

uglification

The inspector

reviewed licensee qualification and certification requirements

for steam generator

eddy current inspectors

and data analysts.

Licensee

requirements

for steam generator

eddy current inspectors

were included in

licensee

procedure

73TI-9RC01.

The inspector reviewed approximately thirty eddy current

and analysis

personnel

qualifications.

The inspector found that licensee

and contractor

personnel

qualification and certification records

were up-to-date

and that the

inspectors

or analysts

were qualified in accordance

with licensee

procedure

73TI-9RCOl.

The inspector also found that the licensee

and contractor

inspectors

were certified level I, II, or III in accordance

with American

Society for Nondestructive

Testing Standard

SNT-TC-IA.

7.2

Steam Generator

Chemical

Cleanin

- Unit 2

37703

42700

62700

The inspector

reviewed the licensee's

preparations

for chemical

cleaning the

secondary

sides of the Unit 2 steam generators.

The primary purpose of the

chemical

cleaning

was to remove

tube deposits that could

become initiation sites for intergranular stress

corrosion cracking.

During

the licensee

observed

axial cracks in steam generator

tubes at locations where deposits

19

~

,l

'

)

l

existed.

The licensee

contracted

with

BSW Nuclear Technologies

(BWNT) to

perform the chemical

cleaning operation.

The chemical

cleaning

process

used different solvents to remove magnetite

and

copper deposits

from the steam generator

internal surfaces.

The generic

process

solvents

were developed

by the Electric Power Research

Institute

and

the Steam Generator

Owners

Group

(EPRI/SGOG),

and the process

had

been

previously used at other nuclear facilities.

7.2. 1

Process

uglification

The inspector reviewed portions of ABB Combustion

Engineering

and

BWNT reports

which evaluated

steam generator materials,

determined corrosion allowances for

materials in contact with solvents,

and described

laboratory tests to

determine

the effectiveness

of the solutions for dissolving deposits.

Samples

of tubes previously removed

from Unit 2 Steam Generator

2-2 were

used in

portions of the laboratory tests.

As a result of the laboratory tests,

BWNT

modified the generic

EPRI/SGOG process to apply to Palo Verde Unit 2

conditions.

The inspector discussed

the qualification of the process

with

licensee

and

BWNT representatives.

The reports

provided

a detailed

background

of the process qualification and the individuals appeared

to be very

knowledgeable of corrosion allowances,

susceptible

materials,

and process

qualification testing.

The inspector

also reviewed

two 10 CFR 50.59 evaluations

prepared

by the

licensee for.the cleaning evolution.

One of the evaluations

indicated that

there

was

no plausible accident

which would result in the simultaneous failure

of more than

one chemical

container

and that

a spill of 55 gallons of ammonium

hydroxide

(one container)

would not affect control

room habitability.

During

discussions

with BWNT personnel,

the inspector learned that

up to four

containers

of ammonium hydroxide could

be moved at one time using

a forklift.

The inspector

commented to the licensee that moving four containers

at one

time could result in dropping

and failure of the containers,

which was not

reviewed in the

IO CFR 50.59 evaluation.

The licensee

subsequently

determined

that failure of four ammonium hydroxide containers

would not affect control

room habitability.

7.2.2

Process

Im lementation

BWNT supplied the equipment for performing the process,

and the inspector

walked

down the majority of the equipment which was located outside of

containment

in the radioactive

waste yard.

The inspector

observed that the

equipment

was in good condition,

components

were labeled,

and there

were

no

indications of leaks

from components that would carry chemical

solvents.

The

inspector

found that

BWNT equipment operators

and process

engineers

were

familiar with the operation of the equipment

and were familiar with the

overall process.

The inspector

observed

portions of pre-cleaning

system operational

testing

which was performed to verify equipment operation

and to ensure that

procedures

were appropriate.

The inspector

considered that the pre-cleaning

operational

testing

was beneficial in ensur'ing

successful

performance of the

20

J

I

evolution.

The inspector reviewed portions of various procedures

used for the

evolution

and found the procedures

to be appropriately detailed.

The

procedures

had

been

approved

by the licensee,

and procedure

changes

were

made

in accordance

licensee

requirements.

The inspector discussed

the implementation of the process with various

licensee

organizations

including Operations,

guality Control, Radiation

Protection,

Chemistry, Fire Protection,

and guality Assurance.

The inspector

found that the organizations

were adequately

involved in preparing for the

process;

however, it appeared

that the licensee's

project manager

was not

receiving all of the comments or concerns

generated

by the organizations.

The

licensee

intended to conduct pre-evolution briefings for all involved

organizations

and

based

on comments

from the inspector,

the licensee

intended

to make procedure

changes

to ensure that licensee

organizations

were involved

in concurring to start the process.

7.2.

The inspector

observed that chemical spill kits, showers,

and

eyewash

stations

were staged

at various locations

and that

BWNT personnel

appeared

to be

knowledgeable of chemical

hazards.

The inspector discussed

contingencies

with

personnel

from the Palo Verde Fire Department

and found that personnel

had

been trained to respond to spills or a fire.

The inspector

found that

procedures

contained

contingency actions if chemistry results or corrosion

rates

were not in expected

parameters.

Licensee representatives

were assigned

to work with the

BWNT personnel

during

the evolution to provide coordination with licensee

organizations for both

normal or abnormal

operations,

and the inspector

found these

personnel

to be

knowledgeable of the process

and their assigned

duties.

The inspector

discussed

contingencies

for handling contaminated

solutions with

licensee

Radiation Protection

personnel

and with BWNT chemists.

The inspector

found that while licensee

personnel

expected

the solvents to become

contaminated

as

a result of the previous Unit 2 steam generator

tube rupture,

the

BWNT personnel

did not expect contaminated

solvents

based

on past

experience

at other facilities.

The inspector

commented that the expectations

of the licensee

and the contractor

appeared

to differ and that additional

coordination

may be needed to ensure that the contractor would be prepared

to

implement contingencies

imposed

by Radiation Protection.

As

a result,

the

licensee

revised the Radiation

Exposure

Permit for the process,

specified

Radiation Protection expectations,

assigned

additional technicians

to the job,

and worked with contractor personnel

to ensure that licensee

expectations

were

understood.

7.2.4 Chemical

Cleanin

Performance

The inspector

observed

portions of the chemical

cleaning operations

and

closely followed the licensee's

progress.

Overall, the inspector

observed

good operations

practices

and noted

good communications

between

BWNT and the

licensee.

However, the inspector

noted that

some activities

had not been

adequately controlled.

For example,

21

I

l

During the transfer of equipment inside containment

from Steam Generator

2-1 to the Steam Generator

2-2,

one person

was sprayed with chemical

solution

on his clothing and another

person

was sprayed

in the face.

The leaking connection

was immediately reconnected

and the leak stopped.

The person

sprayed in the face

used the eye wash station to rinse his

face.

Both individuals exited containment,

and showered.

Neither

individual experienced

side effects from the solution.

The licensee re-

briefed all chemical

cleaning personnel

on the requirement to wear face

shields

and proper protective clothing when working with chemical

cleaning

systems.

Subsequently,

BWNT used

a large catch basis to

collect possible

leakage

and provided coverage

over connection

being

opened to reduce the possibility of spray.

~

On February ll, a spill of water and corrosion products occurred outside

the Unit 2 containment

hatch

when operators

aligned

a sludge lancing

tank to the chemical

cleaning

system.

The spill had resulted

from the

improper manipulation of one of the sludge lancing system valves.

The

BWNT operator did not have specific instructions to manipulate

the valve

and

was not cognizant of the system configuration.

The licensee

initiated procedure

changes

to ensure that sludge lancing valves are not

re-positioned

unless directed

by a supervisor.

~

The licensee fire captain,

the leader of the hazardous

material

response

team,

was called to respond to the February ll spill.

He had difficulty

determining

who was the

BWNT contact at the scene

and it took several

minutes for the captain to find the

BWNT contact.

As corrective action,

BWNT established

that the person in charge of the chemical

cleaning

operation

would be required to identify himself to the fire captain

upon

arrival of the hazardous

material

response

team.

(Note:

The spill that

occurred

on February

11 was subsequently

determined

not to be

a

hazardous

material.)

In addition, the inspector

noted that the licensee

responded

appropriately to

level indication problems

observed

during the chemical

cleaning.

The steam

generator

2-2 level

gage

was blown down with nitrogen several

times in

attempts to achieve

an accurate

level indication.

BWNT operators

monitored

the

amount of solution

pumped into the steam generator

and compared this with

the indicated level

and used the lower of the two indications

as the level.

This conservative

action ensured that the unanalyzed

metal portions of the

would not be exposed to cleaning chemicals.

Subsequently,

operators

determined that the top portion of the tube bundle

had not been

cleaned.

As

a result, the cleaning process

was increased

by 20 hours2.314815e-4 days 0.00556 hours 3.306878e-5 weeks 7.61e-6 months

so that

the top of the tube bundle could be cleaned.

To resolve the level inaccuracy

problems,

the licensee

planned to use the six-inch hand hole

as

a tap for an

attachment for the level instrument in the upcoming Unit 3 steam generator

chemical

cleaning.

The inspector

concluded that the licensee

response

to the events that occurred

during the chemical

cleaning of the steam generators

was adequate.

, 22

l

7.2.5 Results

Chemical

cleaning

removed approximately

5600 pounds of material

from Steam

Generator

(SG) 2-1

and approximately

5000 pounds of material

from SG 2-2.

The

bulk of the material

removed

was iron.

Other material

removed included

nickel, manganese,

and small

amounts of copper, .zinc,

and chromium.

Sludge

lancing subsequently

removed approximately

400 pounds of material

from SG 2-1

and approximately

500 pounds of material

from SG 2-2.

The licensee

found that the. levels

of. chromium in the corrosion products

removed resulted in the classification of the material

as mixed waste

(both

hazardous

and radioactive waste).

At the end of the inspection period,

the

licensee

was evaluating the waste to determine

the actions

necessary

for its

disposal.

The photographs

of the

SG tubes

were taken before

and after

cleaning.

The licensee

determined that chemical

cleaning successfully

removed

deposits

on the steam generator

tube surfaces:

The axial crack indications from eddy current testing

(ECT) in SG 2-2 became

more prevalent after the chemical

cleaning.

Steam Generator

2-1

had

no

appreciable

difference in the before

and after results of the

ECT.

At the

end

of the inspection period,

the licensee

and the

NRC's Office of Nuclear Reactor

Regulation

(NRR) were reviewing these results.

7.3.

Oconee

Tube Plu

in

Problems

The inspector

reviewed the possibility of a plug installation error occurring

at Palo Verde similar to the

one that caused

14 plugs to become dislodged

from

an steam generator

at Oconee Unit 3 in South Carolina.

The dislodged

plugs

were discovered

during

a refueling outage that

commenced

on January

2,

1994.

These plugs

had

been improperly installed during the August

1992 outage

by the

same contractor that performed the tube plugging at Palo Verde.

The contractor explained to the inspector that the plug-rolling tool

used at

Oconee

was

an older model than the one used at Palo Verde.

This tool used

a

flow verification assembly to monitor the air flow and air pressure

delivered

to the air motor which was

used to roll the plugs.

When the proper

amount of

air flow and air pressure

were reached,

the assembly

would stop the air flow.

However, this method did not measure

the actual

torque applied to the plug,

nor could it detect

problems in the rolling process

such

as

a misaligned tool.

The contractor

uses

the Delta roll expansion tool at Palo Verde.

The Delta

roll tool provides

an on-line feedback of the installation torque.

The

contractor developed

a torque trace

method using the Delta tool called

a "roll

energy" verification.

The roll energy verification integrated

the area

under

the torque verses

time curve which was proportional to the work applied to the

plug.

Because

the work applied to the plug was proportional to plug wall

thinning, it could

be used to determine

how well the plug was secured to the

tube.

Additionally, this method could reveal

improper lubrication, tool

alignment

and other problems

by changes

in the shape of the curve

and the

amount of time required to roll the plug.

The inspector

concluded that this

23

J

i

method provided appropriate

assurances

that the plugs were adequately

secured

to the tubes.

8

REVIEM OF

EMPLOYEE CONCERNS

PROGRAM (92720)

As of January

26,

1994, the following statistics

were noted with regard to the

ECP files initiated by the licensee

during the last four years:

1991

1992

1993

1994

Number of ECP concerns

submitted:

Number of submitters:

Number of anonymous .submitters:

Number of concerns

assigned

outside

ECP:

Average length of time to close file (weeks):

Longest period file was

open

(weeks):

Number of concerns

substantiated:

Number of files remaining

open:

129

251

199

94

144

133

12

17

14

92

136

113

5

15

12

19

80

42

22

42

23

0

20

41

The inspector

reviewed the licensee's

ECP procedures,

a random selection of

ECP files closed within the last several

months,

and interviewed several

licensee

employees

who had submitted

ECP concerns.

The inspector

noted that

the licensee

has recently initiated

a comprehensive effort to reduce the

backlog of open

ECP files.

The inspector

noted the following concerns

in that

regard:

The licensee's

ECP procedure

(60AC-OQQ22) requires that

an

ECP file be

initiated for every concern received

by the

ECP group.

Historically,

this has resulted in a large

number of files being opened

and

independently

investigated

by the

ECP group which were not related to

nuclear safety,

which would more appropriately

be handled

by another

group.

The licensee

has

implemented

a program of screening

concerns

received

by the

ECP group,

and making

a determination

as to whether to

initiate an

ECP file or to refer the concern to another licensee

organization for resolution.

However, this

new screening

program

and

associated

new "initial contact form" has not been incorporated into the

ECP procedure.

As

a result the criteria for initiating an

ECP file are

not clearly or consistently defined,

nor are the bases for such

determinations

clearly or consistently

documented

or retrievable.

In one

instance,

the inspector noted that

a concern

received

on October

1,

1993,

involving inadequate

pre-maintenance

tailboard meetings

and poor ALARA,

appeared

to warrant opening of an

ECP file.

Hany

ECP files have

been closed without

a final closure letter being sent

to the concerned

employee,

as required

by the licensee's

ECP procedure.

ECP file 93-130-01,

which involved potential discrimination,

was closed

on the basis that the concerned

employee

requested

that the concern

be

withdrawn.

Based

on

a review of the employee's

concerns

and discussion

with the employee,

the inspector

concluded that the licensee failed to

properly resolve this concern.

As

a result,

the licensee

missed

a

l

significant opportunity to reenforce its stated

commitment to a timely

and discrimination-free resolution of employee

concerns.

In particular,

the involved employee stated that

he would not use the licensee's

ECP

program in the future because

he was convinced that licensee

managers

had

conducted

a "witch hunt" to identi.fy him as the alleger.

The employee

had only withdrawn his employee

concern

on the basis that

he had already

gone to his manager

and identified himself as the alleger.

The concerned

employee told the inspector that none of the managers

in the employee's

reporting chain or from the

ECP group had ever sat

down with him to

resolve his discrimination concern.

This failure has the potential to

created

a "chilling effect" that is likely to reach further than the

involved individual, since

he continues to interface with other licensee

employees

with the belief that

he was not properly treated.

~

The licensee

has

a long history of a high volume of both technical

and

discrimination concerns

coming into its

ECP program,

as well as having

the highest incidence in Region

V of substantiated,

employee

discrimination concerns

submitted to the

NRC and the Department of Labor.

~

ECP files involving substantiated

concerns

did not consistently

document

or reference

documents for tracking corrective action followup (e.g. file

93-010-01,

and 93-109-05).,

In the instance of file 93-109-05, it was not

clearly documented that the licensee

had thoroughly evaluated

the impact

of failed snubbers.

t

The licensee

acknowledged

the inspectors

comments

and stated that actions

would be taken to correct the problems

noted

by the inspector.

9

TRAINING AND QUALIFICATION EFFECTIVENESS

(41500)

In late

1993, the

NRC was

made

aware of concerns that the operators

at Palo

Verde could be reluctant to take actions in the simulator because

adverse

actions

would be taken against

them if they misdiagnosed

a problem.

In order

to address

this concern,

.an

NRC inspector interviewed five reactor operators

and five senior reactor operators

on January

26,

1994.

Four individuals were

from Unit 1, two from Unit 2,

and four from Unit 3.

Eight individuals clearly stated that they would always take appropriate

actions in the simulator or in the plant and that they were not influenced

by

a fear of adverse

actions if they misdiagnosed

a problem.

They all stated

they would take the correct actions.

One of the eight stated that

he

had

been

challenged

by personnel

from another facility concerning

"malicious

compliance" during the Palo Verde steam generator

tube rupture event,

but

clearly stated to that individual and to the ins'pector that

no "malicious

compliance"

was intended or occurred.

One individual stated

he would not hesitate

to take the correct actions,

both

in the simulator

and in the plant, but that requirements

for strict adherence

to procedures

in the past

had caused

him some fear.

He went on to state that

the manner in which the steam generator

tube rupture event

was dealt with by

procedures

was slow.

The operators

knew they had

tube

25

l

J

rupture

and what they had to do, but were frustrated

by the plant procedures.

Although there

were

no safety implications, the delays

complicated

recovery

and cost the utility a lot of money.

The individual indicated that the

facility now recognizes

these

problems

and it appears

they are

headed

in the

correct direction (i.e., to give operators

reasonable flexibility). Starting

last year senior

management

has repeatedly

asked for operators'nputs

and

these

issues

have

been identified.

Regarding the delays complicating the recovery from the steam generator

tube

rupture,

the inspector

observed that, since the steam generator

tube rupture

event the facility has provided additional

guidance

and training to ensure

operators

can complete appropriate

emergency

steps early.

In October

1993,

the facility also committed to the

NRC to completely review and revise the

emergency

procedures

to make them more usable for the operators.

The.

inspector

concluded that these

two corrective actions

adequately

resolved the

operator's

concern.

One other operator stated that

he would make the best decision

he could and

follow through regardless

of what others

would do to him.

He did not have

any

specific problems in mind, but was concerned

about the constant

emphasis

on

accidents

and training far beyond the design-basis

events described

in the

Final Safety Analysis Report.

He felt that operators

may cause

problems

because

they were looking too hard for problems during

an uncomplicated trip

when there

were no problems.

The inspector discussed this issue with the operator,

but did not identify any

significant safety concerns.

The inspector

agreed that the scenarios

used for

training or testing

are often beyond the plant design basis,

but this provides

a safety margin for operators,

training,

and plant procedures.

The operator

did identify two procedure

steps

in emergency operating

procedures

which, to

be useful,

almost always

had to be completed early.

These

step directed the

sampling of the steam generators,

and restoration of power to non-vital

480

Volt AC busses.

The inspector

found that the planned revision of the

licensee's

emergency

operating

procedures

should help this problem,

and that

operators

had

been trained to complete steps early when necessary.

The

inspector

concluded that there

was

no need to correct existing training.

The inspector determined that all the individuals interviewed understood

the

facility policy for following procedures

and were confident they were

implementing these

procedures.

The operators

did not identify any safety

problems

and were aware of the emergency

p'rovisions of 10 CFR 50.54(x).

At the conclusion of the inspection the inspector discussed

the findings with

the Manager of Operations

Training

and the Vice President,

Nuclear Generation

to ensure

they were aware of the issues

raised

by the operators.

26

'

10

FOLLOWUP

ON CORRECTIVE ACTIONS FOR VIOLATIONS (92702)

10. 1

Closed

Violation 50-529 93-04-02: Fire Protection Test Procedure

not Followed

Unit 2

This violation was issued after

a fire protection technician failed to perform

a surveillance test in accordance

with the procedure.

The technician

performed procedure

steps out-of-order, did not establish

and maintain

communications,

and signed off restoration

steps prior to completing the

procedure.

The licensee

reviewed the incident

and determined that the violation was

caused

by personnel

error.

The technician believed that the requirement to

perform steps sequentially

was general

guidance

and that steps

could

be

performed in a different sequence if the objectives of the test

were still

being met.

The licensee

counseled

the technician regarding procedural

compliance.

Additionally, the licensee

briefed the entire fire department

on

the incident

and the need for procedural

compliance.

The inspector

concluded

that these

steps

were appropriate for this violation.

10.2

Closed

Violation 50-529 93-35-02: Failure of Offsite Safet

Review

Grou

to Review Abnormal Indication

Unit 2

This violation was issued

because

the Offsite Safety

Review Group

(OSRC)

failed to review abnormal

indications obtained during the third Unit 2

refueling outage in 1991.

The licensee

had discovered

one axial mid-span

crack and six axial cracks at the first. tube support.

Technical Specification 6.5.3.4f required the

OSRC (formerly the Nuclear Safety Group) to review

abnormalities

or deviations

from normal.

The inspector

reviewed the

licensee's

response

to the violation and determined that they were

appropriate.

The licensee

acknowledged that the abnormal

indications should

have

been

reviewed

by the

OSRC.

Following the tube rupture in Unit 2, the licensee

formed

task force, consisting of senior managers

and

engineers,

to evaluate

the failure.

The licensee

intended to develop the task

force into a permanent

group whose mission

was to identify and predict failure

modes

and determine

and implement strategies

to minimize steam generator

problems.

The licensee

stated that steam generator

group would make

a

presentation

to the

OSRC if steam generator testing revealed

a significant

deviation or abnormality, or the necessity for a reduced operating cycle.

The inspector

concluded that the licensee's

level of management

involvement in

the site's

steam generators

and the formation of a permanent

group should ensure that future steam generator

problems

are addressed

by the

appropriate

level of management.

10.3

Closed

Violation 50-529 93-40-04: Surveillance Test

ST

Administrative Procedure

Not Followed

Unit 2

This item involved operations

personnel

in Unit 2 not following the

administrative controls for surveillance testing documentation.

Specifically,

27

'

t

contrary to the administrative

procedure

governing surveillance testing,

operators

did not mark

a step unsatisfactory

during the performance of a

charging

pump surveillance or make

a test log entry documenting the problem.

The item also noted that the Unit

1 Operations

personnel

did not understand

when the administrative requirements for surveillance tests

were applicable.

This had resulted in a previous violation (Violation 50-528/93-26-01).

The inspector

reviewed the licensee's

response

to the previous violation in

NRC Inspection

Report 93-43

and noted that the licensee's

corrective actions

included forming a focus group to improve the overall surveillance testing

administrative procedure.

The inspector

noted that

an extensive revision to

the surveillance testing administrative

procedure

was issued in December

1993.

The inspector

reviewed the revision

and noted that Step 3.6.1 stated that

"...any failed step or out of tolerance

data shall

be identified by circling

the initialled space

or data entry."

The step also

has specific instructions

for notifying the operation's shift supervisor,

making

a test log entry,

and

initiating a work order or CRDR.

The inspector concluded that the revision to

the procedure clarified management's

expectations

for documenting

problems

during surveillance tests.

The inspector also noted

examples

in the field

(see Section 5.2 of this inspection report) where problems

noted during

surveillance tests

were properly documented.

11

FOLLOWUP (92701)

Closed

Ins ection Followu

Item 50-530 93-11-05:

Ma ne-Blast

Breaker

Ino erabilit

Close Latch

S rin

Interference

This item involved the inoperability of a safety-related

General Electric

(GE)

Magne-Blast breaker

due to interference of the close-latch

spring with the

close-latch monitoring switch.

This item was left open to review the

licensee's

actions regarding

several

recommendations

listed in Condition

Report/Disposition

Request

(CRDR) 3-3-0152.

The licensee's first recommendation

was to issue

a plant change

request to

annunciate

any failure of the breaker closing springs to charge in the control

room.

Plant engineering initiated Plant

Change

Request

PCR 93-13-PB-001 in

July 1993.

The

PCR was approved

by the plant modification committee

on

January

24,

1994.

The modification will be prioritized and scheduled for

installation.

The licensee's

second

recommendation

was to issue

an engineering

evaluation

request

(EER) to permit the use of torsion type close latch springs

on

Magne-Blast Breakers.

EER 93-PB-004

was written to allow using the torsion

springs

as

a design equivalent

change

and directed the replacement

of the

older style springs during scheduled

breaker overhauls.

The licensee's

third recommendation

was for the licensee to obtain service

advice letters

(SALs)

and other information regarding the spring changes

from

the vendor,

General Electric (GE).

The licensee

determined that

GE had not

issued

any SALs concerning

the new-style, close-latch

springs or any problems

with the springs that would prevent the charging of the closing springs.

28

I

The licensee's

last recommendation

was to revise all the applicable

maintenance

and surveillance

procedures

to ensure there

was sufficient

clearance

between

the spring

and the closing mechanism.

The inspector reviewed-

the overhaul

procedures

for the various types of GE magne-blast

breakers

(32HT-9ZZ37 through 32HT-9ZZ39)

and noted that steps

were included to ensure

a

minimum of 1/8-inch clearance

existed

between the spring

and the closing

mechanism.

On January

4,

1994,

NRC Information Notice 94-02 was issued describing this

event

and the potential

impact

on breaker operation.

The inspector

concluded

that the licensee's

recommended

corrective actions

were appropriately

implemented.

11.2

Closed

Ins ection Followu

Item 50-529 93-55-01:

Reactor

Coolant

S stem Flow Anomal

This followup item involved

a review of the licensee's

analysis of reactor

coolant system

(RCS) flow anomalies.

On the evening of January

6,

1994, with

Unit 2 at 85 percent

power, operators

noted that

RCS flow had dropped

approximately

2 percent

over the previous

12 hours1.388889e-4 days 0.00333 hours 1.984127e-5 weeks 4.566e-6 months ,

as indicated

by the flow

instruments for all four reactor

coolant

pumps

(RCPs).

The operators

also

noted increases

in the differential pressure

across

the

RCPs

(approximately

2

percent)

and across

the core (approximately

1 percent),

and increases

in

RCP

amperes

(approximately

5 percent) for all four RCPs.

These

RCS parameters

appeared

to have gradually changed

over

a 12-hour period

and then stabilized.

Reactor

power

and the

RCS hot leg and cold leg temperatures

did not change

appreciably.

The licensee

concluded that change in

RCS flow resulted

from the build-up of a

thin, rough layer of corrosion products

on the fuel rods.

The deposit of

corrosion products

appeared

to result from the planned reduction of RCS

lithium concentrations.

This reduced

RCS

pH and apparently

caused

corrosion

products in colder parts of the system to go into solution

and plate out on

the relatively hotter fuel rods.

The licensee

calculated that

a layer of

corrosion products

approximately 0.0003

inches thick could result in the flow

decrease

observed.

Prior to the flow anomaly,

operators

had

been reducing

RCS lithium

concentration

in preparation for the mid-cycle outage.

This process

is used

to initiate a "crud burst" during

RCS cooldown that allows operators

to remove

RCS corrosion products through the chemical. and volume control system.

This

reduces

the radiation levels in the

RCS, particularly the steam generator

bowls.

The licensee

concluded that they had maintained

RCS chemistry within

industry established

guidelines.

However, the

RCS lithium concentration,

which is proportional to boron concentration,

was high since the Unit was

early in its operating cycle.

Therefore,

the net change in lithium

concentration

and,

subsequently

RCS pH,

was. significantly greater

than

has

been typically experienced

during

a plant shutdown at the end of an operating

cycle.

The licensee

concluded that the change in

RCS flow did not impact plant

safety.

Throughout the anomaly,

core flow remained

above design flow as well

29

~ g

'

0

l

I

f

1

!

l

as Technical Specifications

minimum flow requirements

(95 percent of design

flow).

Additionally, the licensee

determined that core physics

parameters

had

not been affected.

The licensee

expects that normal

RCS chemistry controls

following plant restart will remove the layer of corrosion products

from the

fuel rods

and that

RCS flows should return to normal.

The licensee

discussed their analysis with the resident inspectors

and the

staffs of Region

V, and the

NRC's Offices of Nuclear Reactor Regulation,

and

Analysis

E Evaluation of Operational

Data during

a conference call

on

January

31,

1994.

Their analysis

was determined to be acceptable.

12

ONSITE REVIEW OF LICENSEE EVENT REPORTS

(92700)

12. 1

Closed

Licensee

Event

Re ort 50-530 93-03

Revision 0:

Emer enc

Diesel

Generator

EDG

Unable to Start

and

Run in the Manual Test

Mode

This

LER reported

an event where the "B" EDG in Unit 3 was not capable of

being manually started

from the control

room from July 3-10,

1993.

As

a

result,

the licensee

determined that the

EDG was inoperable

since the

Technical Specifications

(TS) requirement to start the

EDB in the manual test

mode could not have

been

performed during this period.

The inspector

conducted

a review of the event to determine the safety significance of not

being able to manually start the

EDG from the control

room.

The manual test

method of starting the

EDG is primarily used to verify the

engine is functional following maintenance

on the

When the

EDG is

started in this mode, additional protective

shutdowns

are provided in case the

maintenance

introduced

a condition adverse to the safe operation of the

EDB.

The inspector determined that the

EDB could have

been started

manually using

the simulated loss of offsite power

(LOOP)

and simulated

emergency

safeguards

features

(ESF)

manual start buttons at the local

EDG control panel.

These

methods of starting are

used to test the design features

described

in the

safety analysis for the

EDB to start

on

a

LOOP or ESF actuation.

Additionally, the

EDB was able to automatically start in the event of an

actual

LOOP or

ESF actuation.

The inspector

concluded that the safety significance of this event

was low

since the

EDG would have automatically started

as designed

and also could have

been manually started

from the local control panel.

13

IN OFFICE REVIEM OF LICENSE EVENT REPORTS

(90712)

LER 50-528/93-04,

Revision

1,

"ASHE Section

XI Testing of Charging

Pumps not

in Compliance with Code Requirements"

was closed

based on'n-office review.

30

'

l

ATTACHMENT

1

PERSONS

CONTACTED,

Arizona Public Service

Com an

APS

R. Adney,

K. Akers,

W. Chapin,

R. Cherba,

R. Flood,

R. Fountain,

B. Grabo,

W. Ide,

J. Levine,

D. Hauldin,

J. Minnicks,

J.

Ong,

G. Overbeck,

F. Riedel,

P. Rail,

K. Roberson,

D. Robertson,

C. Russo,

J. Scott,

C. Seaman,

H. Searcy,

H. Shea,

R. Stevens,

R. Stroud,

P. Wiley,

Others

Plant Manager,

Unit 3

guality Assurance

Manager,

Refueling

and Maintenance

Services

Manager, guality Audits

Plant Manager,

Unit 2

Supervisor,

guality Audits and Monitoring

Supervisor,

Nuclear Regulatory Affairs

Plant Manager,

Unit

1

Vice President,

Nuclear Production

Director, Site Maintenance

and Modifications

Manager,

ECP Program

ECP Investigator

Director, Site Technical

Support

Manager,

Operations,

Unit

1

ECP Investigator

Senior Engineer,

Nuclear Regulatory Affairs

ECP Investigator

Manager, guality'ontrol

Assistant

Plant Manager,

Unit 3

Director, guality Assurance

and Control

ECP Investigator

Manager',

Radiation Protection

Director, Nuclear Regulatory

8 Industry Affairs

ECP Investigator

'anager,

Operations,

Unit 2

R. Henry,

P. Luther,

Site Representative,

Salt River Project

Site Representative,

Public Services

New Mexico

Denotes

personnel

in attendance

at the Exit meeting held with the

NRC

resident

inspectors

on February

16,

1994.

2

EXIT MEETING

An exit meeting

was conducted

on February

16,

1994.

During this meeting,

the

inspectors

summarized

the scope

and findings of the report.

The licensee

acknowledged

the inspection findings documented

in this report.

The

inspectors

acknowledged that the licensee

had provided various proprietary