ML17310B478
| ML17310B478 | |
| Person / Time | |
|---|---|
| Site: | Palo Verde  |
| Issue date: | 07/12/1994 |
| From: | Wong H NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML17310B475 | List: |
| References | |
| 50-528-94-20, 50-529-94-20, 50-530-94-20, NUDOCS 9408020062 | |
| Download: ML17310B478 (52) | |
See also: IR 05000528/1994020
Text
APPENDIX B
U.S.
NUCLEAR REGULATORY COMMISSION
REGION IV
Inspection
Report:
50-528/94-20
50-529/94-20
50-530/94-20
Licenses:
NPF-51
Licensee:
Arizona Public Service
Company
P.O.
Box 53999
Phoenix,
Facility Name:
Palo Verde Nuclear Generating-Station,
Units 1, 2,
and
3
Inspection At:
Haricopa County, Arizona
Inspection
Conducted:
Hay 8 through June ll, 1994
Inspectors:
K. Johnston,
Senior Resident
Inspector
H. Freeman,
Resident
Inspector
.
J.
Kramer, Resident
Inspector
A. HacDougall,
Resident
Inspector
Accompanying Personnel:
B. Holian,
NRR Project
Manager
J. Ganiere,
NRR Intern
Approved:
ong,
se
,
actor
roJe
s
rane
7
g +/gl
ate
Ins ection
Summar
Areas
Ins ected:
Routine,
announced
inspection of plant status,
onsite
response
to events,
operational
safety verification, maintenance
and
surveillance observations,
and refueling activities.
Results
Units
1
2
and
3
211
Control
room communication
weaknesses
were determined
to be
a significant
contributing factor in a Unit 2 reactor trip (Section 2.3)
and
a Unit 3
turbine-driven auxiliary feedwater
pump overspeed trip (Section
5. 1).
In the
case of the overspeed trip, these
communications
weaknesses
resulted
in a
failure to follow procedures
which was determined to be
a violation.
94080200b2
940729
- DOCK 050D0528
6
POR
~ W
3
0"
t
l
'I
I
)
I
Plant operators
failed to adequately 'consider the impact
on the operability of
a Unit 2 emergency
diesel
generator
(EDG) after racking
a breaker out for an
automatically actuated
exhaust
fan for the diesel
generator
room
(Section 6.2).
The licensee's
response
to
a violation concerning the use of an automatic
volume control tank makeup valve was not thorough in that appropriate
changes
to an ambiguous
procedure
were not made (Section
9. 1).
~
maintenance
The licensee
performed
an expedient
and detailed review of reactor trip
breaker failure to close problems.
They enlisted extensive
vendor support
and
initiated prompt corrective actions
(Section 4. 1).
Continuing problems. with feedwater
control
system
and steam
bypass
control
valves
caused
operators
to take certain
manual
actions
in response
to .a Unit 2
turbine trip (Section 2.2)
and
a Unit 2 reactor trip (Section 2.3).
The licensee
was slow to initiate a problem evaluation for deficiencies
identified by the inspector concerning control
room essential
air handling
unit bolting (Section 3. 1).
In addition,
a problem evaluation
had not been
initiat'ed in response
to weaknesses
in the maintenance
planning process
which
led to significant rework on the Unit
1 pressurizer
spray valves early .in the
year (Section
10. 1).
In the this case, it appeared
that personnel
involved
assumed
that the current reengineering effort would address
these
weaknesses
without performing
a thorough review.
The licensee
had not established
clear expectations
concerning the practice of
entering the electrical
cabinets for safety-related
equipment.
This
contributed to a situation where both diesel
generators
were inoper able in
Unit 3 when the, plant was shut
down (Section
6. 1).
~
En ineerin
The Plant ReviewtBoard
(PRB) performed
a detailed
and proactive evaluation of
a complex issue involving steam generator
tube plug leakage
(Section 3.2).
'The Independent
Safety Evaluation
Group performed
a thorough
assessment
of the
10 CFR 50.59 review program
and identified several
concerns
(Section 8).
Independently,
the
NRR Project
Nanager identified similar program weaknesses
and
a violation concerning the failure to follow the licensee's
procedure.
Several
good practices
were also noted,
such
as the broad
use of
reference
documents for 10 CFR 50.59 reviews..
The initial review of the auxiliary feedwater
pump trip described
in
Section
5. 1 did not identify the cause of the trip until the inspector
asked
an obvious question in light of the pump's
performance history.
1
<l
I
r
f
The licensee
concluded
reviews of reactor coolant
sample valve cracking
and
steam
generator
blowdown water
hammer events.
Both evaluations
were thorough
arid involved considerable
engineering effort. It was noted that the problem
evaluation for the water
hammer event did not initially address
the need to
inform all, operations
personnel
of the efforts needed
to avoid these
events.
~
Hang ement Overview
0
Two areas
of general
concern
were noted.
As discussed
in the plant operations
section
above,
there were two instances
of poor communications
which
contributed to plant events.
The issue of weak communications
practices
as
a
contributor to human performance errors
has
been previously identified.
It
was noted that, in response
to the Unit 2 reactor trip, the licensee
was
establishing
a site wide communication standard.
The report also highlights three instances
in which prnblem evaluations
were
either not initiated for problems or did not cover si g.>ificant attributes
of
the issue.
These
instances
were associated
with:
(1) an evaluation of loose
bolts
on
a control
room essential
ventilation fan (Section 3. 1);
(2) management's
expectations
for entering safety-related
equipment
(Section
6. 1);
and (3)
an evaluation of pressurizer
spray valve maintenance
problems
(Section
10. 1).
Summar
of Ins ection Findin s:
Violations were identified in Sections
5.1
and
8 (Violations 530/9420-01
and 528/9420-05).
~
A followup item was
opened in Section
6. 1 concerning
work on diesel
generator
relays
(Inspection
Followup Item 530/9420-02).
A followup item was identified in Section 6.3 concerning the application
of Regulatory Guide
(RG)
1. 108 regarding diesel
generator reliability
testing (Inspection
Followup Item 530/9420-03).
~
A followup item was identified in Section
8 concerning the updates
to'he
final safety analysis report for design
changes
(Inspection
Followup
Item 528/9420-04)
~
A response
to
a violation was reviewed
and
was left open (Violation
529/9348-02).
~
A violation response
was reviewed
and closed (Violation 528/9348-05).
~
A followup item was reviewed
an left open (Inspection
Followup
Item 528/9355-02).
l
e
A followup item was reviewed
and closed
(Inspection
Followup
Item 528/9409-02).
I
l
l
l
~
"l
A followup item was reviewed
and closed
(Inspection
Followup
Item 528/9409-02).
Attachment:
Persons
Contacted
and Exit meeting
I
)
0'
l
-5-
DETAILS
1
PLANT STATUS
1.1
Unit
1
Unit
1 operated
the entire inspection period at essentially
86 percent
power.
The licensee
continued to monitor
a very small primary-to-secondary
leak in
12 which remained
less
than
1 gallon per day.
S
1.2
Unit 2
Unit 2 began the inspection period at 86 percent
power.
On May 15,
1994, the
turbine tripped
as
a result of a generator field trip (see Section 2.2)
and
'he
reactor
was stabilized at
12 percent
power.
On Nay 16, the unit was
placed
on the grid and achieved
86 percent
power on the next day.
On Nay 28,
the inadvertent
opening of a containment
spray valve caused
a trip of a
pump
(RCP)
and subsequent
reactor trip (see Section 2.3).
On
June 4, the reactor
was critical and placed
on the grid.
The unit returned to
86 percent
power the next day and remained at that power through the
inspection period.
Unit 3 began. the inspection period in Node
6 with the core reloaded.
The unit
entered
Node
5 on Nay 12.
On Nay 30, after entering
Mode 3 in preparation for
a reactor startup,
a level transmitter nozzle weld on Steam Generator
32 began
leaking.
The licensee
returned the unit to Node
5 to perform
a weld repair.
The licensee
conducted
the repairs
and
on June
7 placed the unit in Node 3.
On June 8, another
32 weld (on the blowdown sample line) began
leaking
(see Section 2.4).
Again, the unit was placed in Node
5 to conduct
a
weld repair
and ended the inspection period in this mode.
On June
10, the licensee
met with the
NRC for an enforcement
conference
held
at the Region IV office in Arlington, Texas.
The conference
was held to
discuss
the operability of the Unit 3B
EDG and the causes
of the damage
identified in March 1994.
2
ONSITE RESPONSE
TO EVENTS
(93702)
2.1
Fire in the Unit 2 Radiolo ical Controls Area Yard
Re uirin
Notice of
Unusual
Event
On Nay 18,
1994, at approximately 6:40 p.m., the Unit 2 control
room received
notification from security personnel
that
a small fir'e was burning within the
protected
area,
in the radiological controlled area yard.
The Unit 2 fire
team advisor,
a licensed reactor operator,
responded
to the fire.
When the
fire team advisor arrived at the scene,
the fire had
been extinguished
by the
security personnel
with a portable fire extinguisher.
~ e
1
!
0
l
i
A preliminary investigation revealed that
a portable light had apparently
fallen or blown over.
The light was
on some plastic covered
plywood.
The
heat from the light ignited
a small section of plastic
and plywood.
The fire
department
removed the plastic
and plywood and hosed
down the area.
Radiation
protection personnel
took air samples
and surveyed
personnel
and materials
as
work progressed
to ensure
there
was
no radiological
problems.
Based
on reports
from the operator
and security department,
the shift
supervisor
determined that the fire had burned longer than
10 minutes,
and
a
Notification of Unusual
Event
(NOUE) was declared
and terminated at 6:48 p.m.,
State
and local authorities
were notified 10 minutes later.
The inspector
observed
the fire area
and concluded that the damage
from the
fire appeared
to be minimal.
The licensee's
emergency'response
has
been
evaluated
in NRC Inspection
Report 50-528/94-19;
50-529/94-19;
50-530/94-19.
2.2.
Unit 2 Turbine Tri
on Loss of Turbine Generator Excitation
On Nay 14,
1994, Unit 2 experienced
a large load rejection when the main
generator
tripped from 86 percent
power.
The genet ator trip occurred
when
an
instrument failure caused
the generator
to lose field excitation.
The loss of
the main generator
resulted in a turbine trip and automatic reduction in
reactor
power to approximately 34 percent
power as designed
by the reactor
power cutback system.
Power
was eventually reduced to approximately
12
percent
by the operators.
The licensee initiated
a condition
report/disposition
request
(CRDR) to evaluate the generator trip and
subsequent
plant and operator
responses.
As operators
attempted to stabilize the plant at 20 percent
power,
steam
generator
water levels developed oscillations of approximately
20 percent of
the narrow range indication.
Operators
placed the feedwater control
system
(FWCS) in manual,
but were unable to control the level oscillations.
The
osci llations increased significantly.
The shift supervisor
decided to reduce
reactor
power from 25 percent to less than
16 percent
(the feedwater
economizer/downcomer
valve swap-over
point).
The
FWCS was returned to
automatic,
power was reduced,
and the level oscillations stopped after swap-
over occurred.
The plant experienced
problems with the steam
bypass control system
(SBCS)
during the plant stabilization.
Valve CV-1001 had
a an erratic response
and
occasionally stuck,
and Valve CV-1004 would not fully close until the valve's
handswitch
was taken to the off position.
The licensee initiated work
requests
to troubleshoot
these valves.
The inspector
noted the continuing problems the licensee
has experienced
with
the
FWCS and
SBCS.
The inspector
noted that problems with the
FWCS and
SBCS
contributed to complications
experienced
following the Unit 2 reactor trip on
May 28,
1994.
The inspector will review the licensee's
CRDR evaluation
as
part of a future routine inspection.
~ ~
I
l
0'
1
l
~
2.3
Unit 2 Reactor Tri
On Nay 28,
1994, at approximately
11: 15 a.m.,
a reactor trip occurred in
Unit 2 due to a low departure
from nucleate boiling ratio as the result of
1B tripping following a phase-to-phase
fault.
This fault occurred after
ISC technicians
removed
a relay in the wrong train of the engineered
safety
features
actuation
system,
causing
a containment
spray valve to open
and
borated water to drain from the reactor water tank into the containment
area
where the junction box was located.
The licensee initiated an investigation
of the event to determine the root cause of the errors, to evaluate
the
adequacy of operator
performance,
and to identify and correct
any equipment
problems
caused
by the water in the containment.
On Nay 30,
an
NRC special
team inspection
began
an independent
assessment
of the event.
The results of
the inspection will be documented
in NRC Inspection
Report 50-528/94-23;
50-529/94-23;
50-530/94-23.
Two other minor discrepancies
were identified during the reactor trip event.
First, control
room operators
noted that the rod bottom light and lower
electrical limit light for Control
Element Assembly
(CEA) 3 took approximately
10 seconds
to illuminate after the reactor trip.
Second,
a steam
bypass
control valve and
economizer valve did not go fully shut (slightly
open)
and required operator actions to shut the valve or isolate flow to the
valve.
Regarding the slow illumination of lights on
CEA 3, the operator actions in
identifying the slow illumination of the rod bottom light and initiation of
boration of the. reactor coolant system
(RCS) during the event were
appropriate.
The licensee
reviewed posttrip alarm printouts which showed that
all rods were within 10 inches of each other at the bottom of the core
3 seconds
after the reactor trip, indicating appropriate
response
by CEA 3.
The licensee
investigated
the position indication circuit and tested
the
circuit when the
CEA was actually moved
and could find no cause for the slow
rod bottom and lower electrical limit lights.
A rod drop test of CEA 3 was
conducted
during plant startup to verify proper drop time.
The
CEA met the
test acceptance criteria.
For the steam
bypass control
and feedwater valve problems,
operators
were
aware of prior problems with these
valves
and were able to take manual
actions
to prevent excessive
flow through the valves.
The licensee
performed
a
calibration of the valve positi.oners
and verified proper operation of the
valves prior to unit startup.
While the manual
operator actions to close
valves did not adversely affect operator
response
to this event, there
has
been
a continuing need for operator intervention after plant transients
due to
the failure of these
valves to fully close
when required.
While the licensee
has
been pursuing long-term corrective actions,
minor problems remain.
This
matter warrants
continued
management
attention.
f
f
I
f
I
I
2.4
Unit 3 Steam Generator
32 Nozzle Leaks
On Nay 30,
1994, while the plant was in No'de 4 in preparation to start
up
following a refueling outage,
the licensee identified
a leaking weld in an
instrument nozzle for the Steam Generator
32 narrow range level instruments.
The nozzle
was welded inside the steam generator,
making inspection
and, repair
difficult.
The licensee
cut out
a segment of the nozzle,
performed
a weld
buildup
on the outside of the steam generator,
and welded
a new nozzle to the
outside of the steam generator.
The licensee
was able to determine through
a combination of ultrasonic,
visual,
and dye penetrant
examinations that neither the nozzle nor the steam
generator
contained
flaws.
The licensee
concluded that the weld must
be the
source of the leak,
although they were not able to directly inspect the weld.
On June 8,
1994, while the plant was in Node 3 following the weld repair, the
licensee identified another leaking weld in Steam Generator
32 in the nozzle
for the downcomer
sample line.
Following the end of the inspection period,
the licensee'inspected
the nozzle weld areas
using
a robot and concluded that
the leaks
were not caused
by the recently performed chemical
cleaning
process
and were probably caused
by inclusions in the weld.
The licensee
indicated
they would remove
and inspect the nozzle weld material in a future outage.
The inspector s will follow the licensee's
actions in the course of routine
inspections.
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'ctivities
and equ'ipment,
observation of control
room operations,
tours of the
facility, inter views
and discussions
with licensee
personnel,
independent
verification of safety system status
and Technical Specifications
(TS)
limiting conditions for operation, verification of corrective actions,
and
review of facility records.
3.'1
Loose Bolts on Control
Room Essential
Ventilation Fan
Unit
1
On Nay ll, 1994, the inspector
noted two loose bolts
on the Train
8 control
room essential
air handling unit fan during a routine plant walkdown.
The
essential
air handling unit was required to be operable
by the plant's
TS to
ensure control
room ventilation and habitability in an accident condition.
The system engineer
inspected
the component
and determined that the fan was
still operable
based
on the fact that ther e were still at least
10 bolts that
were securing the fan to the support saddle.
The inspector
agreed with the
licensee's initial operability determination.
il'r
I
f
j
The licensee initiated
a deficiency work order
(DFWO) to secure
the two loose
bolts.
Additionally, the licensee
inspected all six control
room essential
air handling unit fans
(two per unit) to,ensure all the bolts were tight.
No
other loose bolts were identified.
However, the lic'ensee
noted that there
was
a gap of about 1/8 to 1/4 inch on all the fans
between the flange of the fan
and the fan support saddle.
The installation drawings indicated that the
and the support saddle
should
be flush.
The licensee initiated
a
separate
DFWO to move the support saddle
and ensure that the flange
and
support were flush.,
The inspector
asked the licensee
whether the gap between the flange
and the
support saddle
increased
the shear
stresses
on the bolts.
The licensee
reviewed the seismic analysis of the bolts
and determined that there
was
a
large factor of safety in the analysis
and that any additional
shear stresses
and the two loose bolts would not significantly reduce the margin of safety.
The inspector
concluded that the licensee's
qualitative assessment
of the
structural integrity of the existing bolting configuration was reasonable.
Upon questioning
by the inspector,
the licensee stated that
a
CRDR had not
been initiated because
the evaluation
was adequately
addressed
in the
DFWOs.
The inspector
questioned
the licensee
as to whether
a root cause
analysis
was
conducted to determine
how the bolts
became
loose.
The licensee
conducted
a
search of the work orders
and determined that the fan was last worked in 1987.
At that time, the fan was
removed
and reinstalled.
The licensee
could not
conclusively determine if the bolts were incorrectly installed or if the bolt
nuts
became
loose
due to the gap between the flanges.
Based
on the inspectors
questions,
the licensee initiated
a
CRDR to document the condition of the fans
and the engineering evaluation of the structural integrity of the fan support.
The loose bolts
on the fan were properly aligned
and tightened
on June 9.
The
licensee
had not determined
a schedule to correct the problem with the
gap
between
saddle
and the flange of the fan.
The inspector
concluded that the
licensee
s investigation of the problem was thorough.
However, the inspector
was concerned that
a
CRDR was not promptly initiated. to document all the
actions
and to ensure that the right level of management
was involved in the
issue.
Additionally, the inspector
was concerned that the loose bolts
and the
problem with the gap between the flanges
was not identified earlier.
At the
exit meeting, the licensee
stated that the bolt problem appeared
to have
been
caused
during fan reinstallation in 1987.
3.2
~PRB
II
The inspector
attended
the regularly scheduled
PRB meeting held on Nay 18,
1994 (Neeting 94-17).'he first issue discussed
related to the status of a
welded tube plug in Steam Generator
32.
This tube plug had evidence of
leakage
during the Unit 3. midcycle outage
(December
1993)
and the
discussed
the issue at
a special
meeting
on December
14,
1993.
During that
meeting,
the
PRB allowed plant restart after reviewing
a safety evaluation
and
qualitative probabilistic risk assessment.
The
PRB reviewed
a plant
memorandum discussing
the Steam Generator
Group's
(SGGs) position that welded
l
l
I
j
t i,
0
1
1
l
0
-10-
4
plugs determined to have through-wall flaws must
be replaced.
Subsequently,
the chairman of the
PRB requested
that the
SGG present their position to the
PRB to determine
whether their position had changed
and
how that affected the
PRB's decision in December
1993 regarding plant restart.
In Heeting 94-17, the
PRB discussed
the
ASIDE Code requirements
(for both
mechanical
and welded plugs)
and
TS requirements
regarding pressure
boundary
leakage
and steam generator
leakage.
The
PRB concluded that,
although not
explicitly addressed
in the code, the intent was that plug replacement
is the
more conservative
course of action
and requested
the
SGG to gather additional
information and report back at
a future meeting.
The
PRB determined that
a
welded plug should
be considered
an extension of the
RCS pressure
boundary
and, therefore,
plug leakage
would be pressure
boundary leakage.
The
inspector
considered this course of action conservative.
Additionally, it was
noted
as
a strength that the
PRB chairman took the initiative'to=. schedule this
discussion
and revisit a previous
PRB decision
when additional information
became available.
This meeting
was evidence of a
PRB that discusses
issues
in-depth
and is proactive in ensuring that the correct decisions
are made.
4
NAINTENANCE OBSERVATIONS
(62703)
During the inspection period, the inspectors
observed
and reviewed the
selected
maintenance activity listed below to verify compliance with
regulatory requirements
and licensee
procedures,
required quality control
department
involvement,
proper use of safety tags,
proper equipment
alignment
and use of jumpers,
personnel
qualifications, appropriate radiation worker
practices,
calibrated test instruments,
and proper postmaintenance
testing.
Specifically, the inspectors
witnessed
portions of the following maintenance
activity:
4. 1
Reactor Tri
Breakers Failin
to Close Durin
Testin
Units 2 and
3
On Nay 25,
1994, the inspector
observed surveillance testing of the reactor
trip paths
per Surveillance Test Procedure
"PPS Functional Test-
RPS/ESFAS Logic," in Unit 2.
While testing the reactor protection
system
(RPS)
Channel
B trip path, reactor trip switchgear
Breaker
B failed to
close.
The inspector
observed
the technicians'ctions
and concluded that
they appropriately
stopped the test
and notified the assistant shift
supervisor that the breaker
had failed to close.
This event
was followed by
two more occurrences
of reactor trip breakers failing to close.
On Nay 26,
reactor trip switchgear
Breaker
C in Unit 2 failed to close.
Also, on Nay 27,
reactor trip switchgear
Breaker
B in Unit 3 failed to close during testing.
The reactor trip switchgear
breakers,
Type DS-416,
had recently
replaced
General Electric and other model Westinghouse
breakers
in all three
units.
On Nay 27,
representative
arrived at the site
and conducted
an
investigation of the three reactor trip breakers
which failed to close.
It
was determined that the root cause of failure for each of the three breakers
was associated
with the trip latch overlap adjustment
screw.
Specifically,
I
1
II
,
1
l
the vendor concluded that the normal four turns of the overlap adjustment
screw did not provide enough overlap of the trip latch onto the trip shaft for
these
breakers.
As
a result, vibration
induced
when the closing springs
were
energized to close the breaker
caused
the trip latch to trip free, thereby
leaving the breaker in a tripped condition.
The vendor
recommended
that the normal four full turns of the breaker trip
latch overlap adjustment
screw
be increased
to a limit of five full turns
when
required.
This additional overlap reduces
the mechanism's
sensitivity to
vibration and prevents
the breaker
from tripping free during the closing
stroke.
On June
1, the licensee
received
a letter from Westinghouse
authorizing the additional full turn.
also provided additional
testing requirements
to insure that increas'ing the .overlap does not prevent
the breaker
from opening
on demand.
The inspector
reviewed these testing
requirements
and concluded that they provide assur ance that the increased
overlap will not affect the ability of the breaker to open
on demand.
The
licensee
planned to incorporate these requirements'into
Procedure
"Naintenance of Westinghouse
Type DS-416 Reactor Trip Switchgear."
Test results
on the three reactor trip breakers
showed that the five-turn
adjustment corrected
the observed trip free closing condition.
In addition,
the response
times remained within acceptance criteria and the shunt
trip/undervoltage trip gaps were not increased.
The licensee
replaced
the two
Unit 2 breakers,
but the replacement
breakers
did not require more than four
full turns of the overlap adjustment
screw.
The Unit 3 breaker
was adjusted
to 5 turns
and returned to service.
The licensee
planned to adjust the other
- reactor trip switchgear
breakers
as required,
during future routine preventive
maintenance.
The inspector
concluded that the licensee
performed
an expedient
and detailed review of reactor trip breaker failure to close problems,
involving extensive
vendor support,
and initiated prompt corrective actions.
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
TS and approved
procedures.
5. 1
Auxiliar
Pum
Overs
eed Tri
Durin
Testin
- Unit 3
,
On June
8, at approximately 5:30 a.m., turbine-driven auxiliary feedwater
Pump AFA-P01 tripped
on mechanical
during surveillance testing
when
the Unit was in Node 3.
After the
pump trip, the licensee
declared
Pump AFA-
P01 inoperable
and entered
a 72-hour
TS action statement.
The licensee
established
the, normal
steam pressure
to test the
pump at approximately
4 a.m.
on June 8.
The inspector
reviewed the surveillance test procedure
and the normal
operating
procedure for Pump AFA-P01, attended
a meeting with operations
and
engineering to determine the cause of the overspeed trip, and discussed
the
f
l
l
f
(
f
-12-
event with plant supervision.
The licensee
concluded that the trip was caused
by steam
condensing
when it contacted
a relatively cold supply line and
entered'the
pump turbine.
The inspector
noted the following weaknesses
concerning the event:
Poor communications
between operators
led to inappropriately starting
Pump AFA-P01 with a cold steam supply line.
~
The licensee's
investigation of the event did not promptly determine the
temperature
of the steam lines prior to starting
Pump AFA-P01.
As a
result, the licensee
was slow to determine the root cause of the event.
The steam supply for Pump AFA-P01 is provided from two main steam lines
through separate
steam supply lines.
Each supply line has
an isolation valve
(SG-134
and SG-138) located near the main steam line.
These isolation valves
have small
bypass lines with solenoid operated
valves.
These valves are
normally closed
when the
pump is in standby.
The two lines travel to the room
for Pump AFA-POl, where they are joined.
The licensee
had previously identified that condensation
has occurred during
a
pump start
when the isolation valve was opened
and steam
was exposed to the
cold portion of the steam supply line.
The licensee
had found that the
turbine governor could not compensate
for the resultant water "slug" which has
caused
Pump AFA-POl to overspeed.
To address this, the licensee
has taken the
position that
Pump AFA-P01'as only operable with the temperature
of the
downstream portion of the steam supply lines greater
than 190'F.
In the past,
there
has
been sufficient leakage
past the isolation and bypass line valves to
keep the lines warm.
The licensee
also installed
an on-line temper ature
monitoring system
so that temperature
could be verified during operator
rounds.
The licensee
investigated
the June
event
and determined that
communications
weaknesses
were
a significant contributing factor.
Prior to
starting the surveillance test of Pump AFA-POl, the primary reactor operator
noted that the procedure
did not have any limitations for the temperature
of
the steam supply piping prior to starting the pump.
The reactor operator
(RO)
asked the assistant shift supervisor if the minim'um supply line temperature
of
193'F (described
in the normal operating procedure)
applied during the
performance of the test.
The assistant shift supervisor
noted that the
precaution section of the operating
procedure
stated that the steam supply
line temperature
was required to be greater
than
193~F.
The assistant shift
supervisor
then apparently
answered
"yes" to the RO's question concerning
whether the temperature limit applied to the test. The
RO thought that the
assistant shift supervisor
had actually verified that the supply line
temperature
was greater than 193'F.
The assistant shift supervisor
had not
verified the temperature,
and the actual
steam line temperature,
as measured
by the temperature
monitoring recorder,
was 77'F.
0
l
l
l
0
0
-13-
After this discussion,
the
RO opened the Train A steam supply Valve SG-134
and
started
AFA-POl.
The
pump turbine reached
rated
speed of 3600 rpm and the
governor
was properly controlling turbine speed.
The operators
ran the
pump
for about
40 minutes
and then opened the Train
B steam supply Valve SG-138
as.
directed
by the test procedure.
As soon
as Valve SG-138 opened,
the turbine
slowed to 3400 rpm and then its speed
increased
quickly, which caused
a
mechanical
The inspector
noted that the test procedure for Pump AFA-POl required that the
operators verify system line-up in accordance
with the normal operating
procedure.
The normal operating
procedure
had precautions
to ensure that the
temperature
of the steam supply line was greater
than 193'F using either the
on-line temperature
recording
system or a hand held pyrometer prior to
starting the
pump.
The inspector
concluded that not ensuring the temperature
of the steam supply line was greater
than 193'F prior to starting the
pump was
a failure to follow procedures
(Violation 530/9420-01).
During the event investigation,
the licensee
did not establish that the steam
supply'ad not been adequately
warmed until the inspector
asked
what the
temperature
had
been prior to the
pump start.
The inspector
posed the
question following two licensee staff meetings,
which included plant
management
and engineering
involvement
and substantial
exploration of the
cause of the overspeed
condition.
Based
on the inspector's
question,
the operations
supervisor
then reviewed the
auxiliary operator
logs
and noted that the temperature
'was logged
as 77'F.
The auxiliary operator
had circled the reading
as "out of specification"
and
made
a comment
on the log sheet that the reading
was low due to not being at
normal operating
steam pressure.
The inspector
expressed
concern regarding the need to prompt the licensee
regarding the initial steam line temperature
given the history of overspeed
trip events
due to condensation
and the relatively short time between reaching
normal operating temperature
and starting the
pump.
The inspector
noted that
the appropriate individuals had
been involved in the event, review.
However,
it appeared
that these individuals had relied on assumptions
rather than the
evidence available in performing their initial review.
The inspector
concluded that, while those involved may have ultimately determined
the
correct root caus'e,.their initial review was not well conducted.
The licensee initiated
a
CRDR to document the event
and to evaluate
the
operator
communication errors which led to the event.
Plant engineering
had
initiated
a review of the adequacy of the design of the steam admission
system
to prevent water intrusion into the tur bine following concerns
raised
by the
testing
NRC team inspection
(NRC Inspection Report 50-528/94-12;
50-529/94-12;
50-530/94-12).
The details of the recent
overspeed trip event were included
in this evaluation which was scheduled to be completed
by the end of June.
A
review of this evaluation will be conducted
in a future inspection.
J
I
1
I
l
i
1
-14-
'.2- Other Surveillances
Observed
The inspectors
observed portions of the following surveillances
and determined
that they were performed acceptably:
Nain Steam Line Isolation Valves Surveillance 4.7. 1.5
42ST-2DF01 Diesel
Fuel Oil
Pump Operability 4.0.5
6
DIESEL GENERATOR PROBLENS
(62703,
61726)
6. 1
Rela
s Not Full
Inserted
Unit 3
On Nay 17,
1993, the Unit 3 control
room received
a trip alarm and
a high
priority alarm on
EDG 3A while the unit was in Node 5.
Because
the Train
B
diesel
generator
was inoperable
due to maintenance activities, the licensee
entered
the
TS action statement for having no diesel
generators
available.
Followup investigations
by the licensee
concluded that
EDG 3A would have
been
available in the emergency
mode.
The inspector
reviewed the licensee's
immediate actions
and concluded that they were appropriate.
The licensee
found that two power relay and controls technicians
had entered
the
EDG 3A control cabinet to trace wiring in preparation for work they
intended to perform later that day.
The technicians
did not have control
room
permission to perform the work, but intended to obtain the permission
following their wire tracing.
Mhile in the cabinet,
one of the technicians
caused
a poorly seated relay to break contact with its socket,
which caused
the alarm.
The licensee initiated an investigation
and discovered that
some of the
Agastat relays,
including the one that caused
the alarm, were not fully
inserted into their respective
sockets.
The circuit that caused
the alarm
reportedly fell out of the socket
when the personnel
removed the seismic
restraint.
The licensee
discovered that
some of the relays that were
purchased
were commercial
grade
and
had
an injection mold lip at the bottom of
the relay which made the relay difficult to insert into the socket.
The licensee
also reviewed the circumstances
surrounding the technicians
entering
a cabinet for the remaining available
EDG without permission
from the
control
room operators.
The licensee
discussed
the preliminary findings and
corrective actions with the inspector
and indicated that this was probably
an
isolated occurrence.
The licensee
noted that the two individuals involved
have
been counselled
and that the power relay and control,s organization
has
had 'subsequent
training concerning e'ntering safety-related
equipment without
permission.
The inspector questioned
the licensee
as to whether they had established
an
expectation for when plant personnel
can
and cannot enter cabinets for safety-
related
equipment without permission.
The inspector discussed
with members of
the unit staff their practices for entering safety-related
equipment cabinets
and discovered that there
was
no consistent
known management
expectation.
l
I
t'
-15-
Some individuals believed that it was acceptable
to open
and enter
a safety-
related cabinet without permission
as long as
no work was being performed,
while others
believed tPat they should obtain permission to open
any
cabinet/drawer
(safety-related
or nonsafety-related).
The inspector
discussed
the need to develop
a consistent
policy for opening
safety-related
equipment cabinets
and to disseminate this policy to the
employees.
At the exit meeting,
the licensee
stated that they would review
their position
and ensure that the appropriate
individuals were provided
guidance.
The inspector will review the licensee's
corrective actions to
develop
and disseminate
a policy regarding entering safety-related
equipment
and the results of the rel'ay insertion problem when the investigation is
complete (Inspection
Followup Item 530/9420-02).
6.2
Diesel
Generator
Exhaust
Fan Ino erable
Unit 2
On Nay 13,
1994, Unit 2 was in Node I when members of the
PRB determined that
the
EDG 2B had
been inoperable
from approximately 4:48 a.m.,
on Nay 10, to
approximately 8:55 a.m.,
on Nay 13.
For approximately
76 hours8.796296e-4 days <br />0.0211 hours <br />1.256614e-4 weeks <br />2.8918e-5 months <br />, the Train
B
EDG building essential
exhaust
fan had
been
removed
from service with its
breaker in the open position
and with the fan's discharge
damper wired in the
closed position.
The
PRB determined that the exhaust
fan had not been capable
of performing its specified support function.
The essential
exhaust
fan
starts automatically when the
EDG starts to ensure sufficient room cooling and
is considered
necessary
attendant
equipment required for the
EDG to perform
its intended safety function.
On Nay 10, operations
personnel
evaluated
opening the essential
exhaust
fan
breaker to support. preventive maintenance
on the exhaust
fan screen.
They
determined that the operability of the
EDG was not impacted
because
the
essential
exhaust
fan could
be reenergized
and manually started
should
operation
be required.
Operations
personnel
failed to consider the essential
exhaust
fan as necessary
attendant
equipment that was required for the
EDG to
perform its intended safety function.
Subsequently,
maintenance
workers wired
the fan's discharge
damper shut to serve
as
a foreign material
boundary.
This
step
was not communicated to the operators.
The inspector
noted that this event
was similar to another
example of where
the licensee
removed
a system
from service that had
an automatic design
function to support the
EDG without an operability evaluation
(see
NRC
Inspection Report 50-528/94-02;
50-529/94-02;
50-530/94-02).
Previously the
licensee isolated the jacket water expansion tank automatic make-up valve in
Unit I, an automatic function described
in the Updated Final Safety Analysis
Report
(UFSAR).
The licensee
submitted
Licensee
Event Report 94-001-00
as
a result of the
and attendant
equipment
being inoperable
for. greater
than
72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />.
The
inspector will evaluate this event
and the licensee's
actions in a future
review of the licensee
event report.
1
(l
l
I
I
)
)
-16-
0
6.3
Application of RG 1.108 Regarding Valid Tests
and Failures
On June
2,
1994, the licensee
discussed their interpretation of RG 1. 108,
regarding diesel
generator reliability testing, with the inspector,
the Office
of Nuclear Reactor Regulation
(NRR) Project Nanager,
and the Chief of the
Electrical Engineeting
Branch,
NRR.
The licensee
had concluded that only the
monthly diesel
generator
surveillance tests
should
be included
as valid tests
under
RG 1. 108.
The inspector
and
NRR concluded that all surveillance tests,
not just the monthly surveillance tests,
should
be included
as valid tests.
The licensee
committed to reassess
their position
and evaluate
the impact on
the reliability trends of the diesel
generators.
On June
6, the licensee
informed the inspector that they had not included
several
diesel
generator tests
as valid tests
because
the start times of the
diesels
had not been recorded.
The licensee
observed
and the inspector
concurred,
that
RG 1. 108 did not include start times
as
a criteria for
recording valid tests.
The licensee
committed to reassess
this issue
and to
evaluate the impact on the reliability trends of the diesel
generators.
The
inspector will review the licensee's
evaluation in a future inspection
(Inspection
Followup Item 530/9420-03).
7
CONTAINMENT PENETRATION TESTING AND VALVE LINEUP Unit 3
(61715)
The inspector
conducted
a review of the licensee's
testing
and procedures
to
verify that the
TS requirement to maintain primary containment integrity prior
to entering
Node 4 operations
was satisfied in Unit 3.
The inspector
conducted
the review while Unit 3 was in Node
5 at less than 210'F.
The inspector
reviewed the local leak rate test
(LLRT) results of selected
mechanical
and electrical penetrations,
verified the operability of the
system,
reviewed the basis for the surveillance test
(ST) procedure for
ensuring that manual
valves associated
with containment penetrations
are
verified closed,
and observed the
LLRT of the containment
personnel
airlock.
The inspector
concluded that all the penetrations
that were required to be
tested successfully
passed
an as-left
LLRT.
The inspector also noted that the
testing of the personnel
airlock was appropriately conducted.
The inspector
noted that all the required tests to verify operability of the
CS system were
satisfactorily completed
and appropriately reviewed.
The inspector identified several
manual
vent and drain valves in penetrations
that were Type
C leak tested that were not verified closed
as part of the
containment integrity ST.
The licensee initiated
a review of the
ST procedure
and determined that the
ST met the intent of the TS.
However, the licensee
also noted several
inconsistencies
in the procedure
and was conducting
a
detailed evaluation of the design basis for verifying containment integrity.
The inspector
agreed with the licensee's
conclusions
and noted that the
licensee initial response
to the inspector's
questions
was
good.
A discussion
of the details of this finding follows.
~ ~
-17-
7.1
Verification of Containment
Vent and Drain Valve Positions
The inspector
reviewed
Procedure
43ST-3ZZ13 "Containment Integrity-
4.6. l.la."
The purpose of the procedure
was to meet
TS 4.6. l.l.a
which requires that primary containment integrity shall
be demonstrated,
"at
least
once per 31 days
by verifying that all penetrations
not capable of being
closed
by OPERABLE containment
automatic isolation valves
and required to be
closed during accident conditions are closed
by valves, blind flanges,
or
deactivated
automatic valves secured
in their actuated
position except
as
provided in Table 3.6. 1."
The inspector
reviewed the
UFSAR to determine which penetrations
have valves
that would receive
a containment isolation actuation signal,
a containment
spray actuation signal,
or
a containment
purge isolation actuation signal to
shut during an accident condition.
The inspector
determined that all the
manual
vent and drain valves connected to penetrations
that received
a
containment isolation actuation signal,
a containment
spray actuation signal,
or a containment
purge isolation actuation signal
were included in the
procedure.
The inspector
noted that the
ST procedure
also included manual
vent
and drain
valves associated
with penetrations
that were listed in the
UFSAR and .the
TS
as containment isolation valves that were required to be open during accident
conditions.
Even though these penetration
were required to open, they were
Type
C tested to ensure that the isolation valves were leak tight.
Based
on
this observation,
the inspector
reasoned
that penetrations
that were Type
C
tested to ensure .the leak tightness of the penetration
should also have the
manual
valves verified shut in the ST.
To verify this assumption,
the
inspector
reviewed the valve configurations for all the penetrations
that were
Type
C tested
and identified four manual
vent and drain valves associated
with
the shutdown cooling
(SDC)
and the .high pressure
safety injection (HPSI)
system long-term recirculation path penetrations
26, 27, 67,
and
77)'hat were not verified closed in the ST.
7.2
Licensee
Evaluation
The inspector
discussed
the apparent
omission of the manual
vent and drain
valves associated
with the
SDC and
HPSI long term recirculation valves with
the shift operations
crew.
The licensee initiated
CRDR 3-4-0331 to evaluate
the condition and determined that the manual
valves connected to the
in question
(26, 27, 67,
and 77) were not required to comply with
TS 4.6. l.Ia.
The basis for the licensee's
determination
was that the valves
were connected to penetrations
that were required to be open during, accident
conditions
and
TS 4.6. 1. la applied to valves that were required to be closed
during accident conditions.
Additionally, Section 6.4.2. 1 of the Combustion
Engineering standard
safety analysis report stated that the
SDC and safety
injection systems
were considered to be an extension of the 'containment
pressure
boundary.
Therefore,
containment integrity would be verified by
ensuring the. integrity of the safety injection and
HPSI systems.
The
3
I
I
I
I
e
-18-
inspector
agreed with the licensee's
conclusion that the
SDC and
HPSI long-
term recirculation penetrations
were not required to be checked to meet the
intent of the TS.
The licensee
reviewed the configuration
and design
basis for all the
mechanical
and did not identify any manual
valves that should
be
verified to meet the intent of the
TS that were not included in the ST.
However, the licensee
did identify several
other penetrations
that were
required to be open during accident conditions that had the vent and drain
valves verified shut in the
ST procedure.
The licensee
was reviewing the
design basis for all the penetrations
that were required to be open to
determine if the vent
and drains should
be included in the
ST or in other
operations
procedures.
The inspector will review the results of the
CRDR
during an ongoing inspection of containment integrity.
8
DESIGN CHANGES IN ACCORDANCE WITH 10 CFR 50.59
(37001)
The inspector
reviewed the licensee's
procedures
and training records for
conducting
10 CFR 50.59 evaluations.
The inspector
found that the procedures
were generally well written and contained
adequate detail.
The inspector also
found that the qualification requirements
for personnel
who screened,
evaluated,
and reviewed
10 CFR 50.59 evaluations
were significant in terms of
experience
required
and factored in increasing
requirements
for the evaluator
and reviewer positions.
The inspector
learned that no requalification
training was required for personnel
performing
10 CFR 50.59 screenings
and
evaluations
and pointed out that Procedure
01PR-ONS04 clearly presumed
a
requalification program.
The licensee
stated that
an internal
assessment
had
identified
a similar finding regarding requalification training,
and the
finding was being evaluated.
The inspector reviewed the previous
two assessments
of the
program performed
by the licensee
and
had no comments
on the first assessment,
performed in November
1991.
The second
assessment
had recently
been
completed
(April 1994)
by the site's
Independent
Safety Evaluation
Group (ISEG).
Although the assessment
report was not yet complete,
the inspector
reviewed
the team's
summary slides
and found that the assessment
appeared to be
thorough
and documented
several significant findings.
One finding was that,
from the screenings
sampled,
approximately
30 percent of the negative
screenings
(meaning
a
10 CFR 50.59 evaluation
was not necessary)
should
have
been positive.
The inspector
reviewed several
of the deficient screenings
and, in general,
concurred with the
ISEG assessment.
As
a result of ISEG's
.finding,
a
CRDR was generated
to perform the
10 CFR 50.59 evaluations for the
deficient screenings
and to perform
a root cause evaluation of this weakness.
The inspector
reviewed the licensee's
10 CFR 50.59 annual report of plant and
procedure
changes,
dated
June
24,
1993,
and selected
approximately
20 change
packages
from a variety of procedure,
design change,
and modification
categories
for a detailed
review.
In general,
the inspector
found that the
summaries
provided in the
10 CFR 50.59 annual report were adequately detailed
l
I
I
i
-19-
in describing the change to the facility or procedure.
The inspector verified
'that only qualified individuals performed the
10 CFR 50.59 screenings
and
evaluations.
Overall, the inspector
found that the
10 CFR 50.59 evaluations
adequately
supported
the conclusions that were drawn.
An overall strength
was the
extensive listing of licensing basis
documents that were reviewed in preparing
the evaluation,
which indicated
a strong
commitment to ensure that facility
changes
were properly evaluated.
Several
design
changes
were reviewed that had two 10 CFR 50.59 evaluations
(one for plant operations
and
a separate
evaluation for plant implementation).
Although these evaluations
could be combined, this approach
was considered
a
strength
because
the operations
evaluation could focus
on the various system
and equipment requirements
for different operational
modes,
and the
implementation evaluation could focus
on other concerr
<. (e.g.,
shutdown risk
factors).
The inspector
reviewed approximately
seven
design
change
packages
to verify
that drawings
and
UFSAR sections
were updated.
The inspector
found that the
UFSAR had not been
updated for two design
changes
which had
been
completed in
Narch
and August 1993.
The licensee initiated a
CRDR to review each closed
design
change
package for the last several
years to identify any changes that
have not been incorporated into licensing basis
documents.
The inspector will
review the licensee's
CRDR in a future inspection report (Inspection
Followup
Item 528/9420-04).
One of the de'sign
change
packages
reviewed,
PJ-Sg-065,
provided for
functional separation
of the condenser
exhaust effluent radiation monitors,
by
rerouting the condenser
exhaust to the plant vent.
The
10 CFR 50.59 screening
correctly identified that the
TS were affected (Figure 5. 1-3 depicts
a
separate
release
path for the condenser air removal
system);
however, the
evaluation
was performed
and the modifications were completed
on all three
units prior to a TS change
being submitted
and approved
by the
NRC.
Although
the significance of the required
TS change
was low in that it involved a
change to a TS figure, procedure
"10 CFR 50.59 Screening
and
Evaluation,"
was not followed.
The failure to follow procedures
is
a
violation (Violation 528/9420-05).
Evaluations that lacked information or wei e not adequately detailed
included
the following:
~
Temporary Nodification 89-CP-076
removed
handwheels
for four containment
purge valves,
but the evaluation did not address
whether
manual
operation of the valves
was
assumed
in the
UFSAR or emergency
procedures.
Design
Change
package
NC-041 upgraded
valve operator motors
and gear
sets
on two nuclear cooling water containment isolation valves.
The
l
0
l
I
1
-20-
design
package only stated that the
TS requirements
were not affected
and did not discuss
the potential
impact of too fast of a stroke time on
valve reliability or system response.
~
Limited Design
Change
Package
RJ-048
implemented
a core operating limit
supervisory
system software
change to automate the monitoring of the
Azimuthal
Power Tilt TS limit.
The
10 CFR 50.59 screening
indicated
that the change did not involve a test
described
in the
UFSAR; however,
the screening
referenced
a preoperational
test of core operating limit
supervisory
system, defined'in
UFSAR Section
14.2. 12, to verify proper
system operation.
moreover,
the evaluation
used conflicting logic in
answering questions
regarding
an increase
in either the probability or
consequences
of an accident.
In each case,
the inspector discussed
the evaluation with cognizant licensee
personnel,
reviewed applicable
documentation
and test results,
and determined
'hat the changes
were either
bounded
by existing analysis
or did not affect
safety.
9
FOLLONlP OPERATIONS
(92901)
9.1
0 en
Violation 529 9348-02:
Failure to Follow Procedures
for RCS
~
~
~
~
~
~
~Nakeu
This violation involved the failure to return the reactor
makeup water
controller to automatic following a dilution evolution.
On November 23,
1993,
the inspector noted shortly after shift turnover that the reactor
makeup water
controller was in manual.
The procedure directed the operator to place the
controller in automatic following manual operation.
The primary operator
subsequently
placed the controller to automatic
when informed by the
inspector.
The inspector reviewed the licensee's
response
to the violation and noted that
the licensee's
corrective actions
included
an operations
management brief
given to each
crew on the incident and management's
expectations
for
procedural
adherence.
In addition, positive discipline was administered to
the individual involved in the event.
R
On Nay 9,
1994, the inspector
observed that the Unit 2 reactor
makeup water to
volume control tank flow controller
(CHN-FIC 210X) was in manual.
Mhen asked,
the
RO stated that
he chose to leave the controller in manual
because
he was
performing
an
RCS dilution every 45 to 60 minutes.
The inspector
asked the
if he was
aware of a previous Unit 2 violation for not lining up the makeup
system for automatic operation.
The
RO responded that he was unaware of any
previous event involving the reactor
makeup system.
The
RO reviewed the
chemical
and volume control system
(CVCS) normal operations
procedure
(420P-
2CH01)
and subsequently
returned the flow controller to automatic.
I
l
'
1
-21-
The inspector
discussed this with Unit 2 operations
management.
Operations
management
stated that it was their expectation that, if an evolution,
such
as
dilution of the
RCS,
was performed
on
a relatively frequent basis, it
was'cceptable
for the operators to leave the flow controller in manual after the
completion of the dilution and not return the controller to automatic.
In
addition, operations
management
stated that they expected
the operators
to
return the controller to automatic prior to shift turnover or document in the
shift turnover log that the controller was left in manual.
The inspector
noted that this interpretation
was consistent with the licensee's
discussion
in their response
to the previous notice of violation.
In response
to the
RO not being aware of the previous violation concerning
reactor
makeup system,
the licensee
generated
a
CRDR to determine
how
information and actions that are taken to resolve previously identified
concerns
are communicated to department
members
and to determine if these
methods
are adequate
to prevent similar occurrences
in the future.
The
licensee
subsequently
determined that the operator
had received training on
management's
expectation
regarding the flow controller, but did not remember
the training.
The inspector
noted that, inthe response
to the violation, the licensee
did
not identify any procedural
problems.
The inspector
reviewed Procedure
420P-
2CHOl and concluded that management's
expectation
concerning the flow-
controller
was unclear.
Following the Nay 9 event,
the licensee
generated
a
temporarily approved
procedure
action
(TAPA) to direct operators
to line up
the makeup system for automatic or manual operation per the shift supervisor's
or assistant shift supervisor 's direction.
The inspector
concluded that this
revision allowed operators
the option to leave the controller in manual;
however, it did not clarify management's
expectation for returning the
controller to automatic prior to shift change or document the condition during
turnover.
The inspector
reviewed the
CVCS normal operations
procedures
for all three
units
and discovered that only the Unit 2 procedure
had
been revised to
incorporate the TAPA.
The licensee
indicated that the Unit
1 and Unit 3
procedures
would not be changed until the next procedure revision in early
1995.
The inspector questioned
the specific individual responsible for the
CVCS normal operations
procedure
and discovered that
he was not aware that the
procedure
change resulted
from a violation.
The licensee
changed their
priorities and decided to issue the procedure revisions incorporating the
TAPAs.
This violation will remain open until the inspector reviews the
procedures
to ensure the change is incorporated.
9.2
Closed
Violation 528 9348-05:
Inadvertent
RCS Draindown
This violation involved the inadvertent draining of the
RCS level from the
reactor vessel
flange level (about
114 feet) into the reduced
inventory level
(< 111 feet)
on November 3,
1993, while the plant was in a refueling outage.
The primary reactor operator
performing the evolution was distracted
during
the evolution and did not monitor the
RCS level for approximately
8 minutes.
i
l
j
1
)
-22-
The licensee initiated several
immediate corrective actions,
documented
in NRC
Inspection
Report 50-528/93-48;
50-529/93-48;
50-530/93-48,
and discussed the,
broader
performance
issues
surrounding the event at
an
NRC management
meeting.
One of the key issues
the licensee
emphasized
at the meeting
was that shift
supervision
allowed important evolutions to be conducted without maintaining
expected
communication standards.
To address this issue,
the licensee
scheduled'igh
intensity training (HIT) in the simulator.
This training was
designed to improve team communication skills.
The inspector
observed
portions of the HIT and found that it emphasized
both communications
and
teamwork
(see
NRC Inspection Report 94-09).-
The inspector also noted that, during the Unit 3 steam generator
inspection
outage in December of 1993
and the Unit 2 midcycle outage in January
1994,
an
alarm was installed that would annunciate
when the
RCS level
was less than
ill feet.
The inspector noted that this alarm and the previous corrective
actions
should prevent
a similar problem with inadvertently draining the
RCS.
The inspector
noted that management's
efforts to improve crew communications
needed to continue
based
on recent errors in Units 2 and 3.
For example,
poor
.communications
between operators
and technicians
in Unit 2 contributed to
inadvertently opening
a containment
spray valve and
a reactor trip (see
paragraph 2.3).
In Unit 3, weak communications
between operators
resulted in
~
~
~
~
~
~
~
~
~
~
starting the turbine-driven auxiliary feedwater
pump with cold steam lines.
This resulted in an overspeed trip of the turbine due to.condensation
in the
steam lines
(see
paragr aph 5. 1).
The inspector will continue to monitor communications to ensure that
communications
during control
room evolutions
are precise
and meet
management's
expectations.
10
FOLLOWUP NAINTENANCE
(92902)
10. 1
0 en
Ins ection Followu
Item 528 9355-02:
Pressurizer
S ra
Valve
Naintenance
This open item involved the engineering resolution of repeated
problems with
maintenance
on the pressurizer
spray valve air operators
in Unit l.
Specifically, the inspector
was concerned
about apparent
weaknesses
that
prevented
engineering
information from being factored into the maintenance
work orders.
The- licensee identified two factors that contributed to the work order not
referencing current engineering
information regarding the pressurizer
spray
valve air operator
bench set.
The first problem was that the revised
bench
'set pressures
were not incorporated into the engineering
comments
screen
in
the component information computer data
base.
The second
problem was that the
revised
bench set pressures
that had
been incorporated into a preventative
maintenance
(PN) task
had not been carried over into a maintenance
task that
replaced
the
PN task.
The licensee
concluded that the distribution of
'
1
l
f
I
l
f
-23-
information between
engineering
and the work planners
should improve with the
maintenance
team concept
being implemented
as part of the reengineering
effort.
The inspector
noted that the licensee's
evaluation of the communication
problems
was not incorporated into the formal
CRDR evaluation
process.
Additionally, the licensee
did not address
the reasons
why the information
screen
was not updated
and if there were other
PN tasks that were converted to
maintenance
task with inadequate
or missing information.
The licensee
believed the reengineering
process
would prevent these errors in the future
without understanding
the specific barriers to ensure that information was
properly incorporated into work documents.
The licensee initiated
a
CRDR to
address
these
issues.
This item will remain
open pending the inspector's
review, of the licensee's
CRDR.
ll
FOLLONJP ENGINEERING/TECHNICAL SUPPORT
(92903)
11. 1
Sam le Valve Crackin
In NRC Inspection
Report 50-528/94-02;
50-529/94-02;
50-530/94-02,
the
inspector discussed
the licensee's
discovery of cracks in two Unit 2
RCS hot
leg sample line valves
(SSAUV-203 and SSBUV-200).
At the end of the
inspection period, the licensee
had not determined the cause of the cracks.
The licensee
subsequently
determined the cause of the cracks
and has initiated
appropriate corrective actions.
The
RCS sample valves were supplied
by Valcor Engineeri.ng Corporation
and were
essentially stainless
steel
blocks weighing approximately
40 pounds.
Inlet
and outlet ports were drilled into the valve body.
Additionally, a larger
diameter cylindrical area
was provided for a threaded
and the valve
internals.
The cracks circumscribed the seat
area at the base of the larger.
cylindrical area.
Additionally, cracks were found initiating from areas
of
stress
concentrations
at the inlet and outlet ports.
The licensee
determined that the cracks
were caused
by thermal cycling.
The
normally closed valves were routinely opened to draw samples
from the
RCS.
This resulted in thermal cycles from approximately
100'F to 600'F.
This
caused
high stress,
low cycle fatigue.
The licensee
determined that
sufficient thermal stress
was created
during valve heat
up to compressively
yield the material.
The licensee
evaluated
the cracks to determine if.it had affected the
operability of these
valves
and other similar valves.
The licensee
concluded
that operability was'ot affected.
The licensee
determined that, at the
present rate of growth, the cracks would not affect valve integrity for at
least
5 years of operation.
Additionally, the Unit 2
RCS hot leg sample
valves
had experienced
the greatest
number of thermal cycles.
>
1
f
J
-24-
Inspections of Unit 3 valves were performed during the refueling outage.
The
licensee
discovered that both
RCS hot leg sample valves exhibited similar
cracking.
Additionally, they discover ed that
a sample valve for the
pressurizer
steam
space
(SSA
UV 205)
had crack indications.
The licensee
estimated that this particular valve would have experienced
I/10th of the
number of cycles
as the hot leg sample valves
and speculated
that the crack
depth would be proportionally shallower.
At the end of the inspection period,
the licensee
was examining the valve to determine the depth of these
indications.
In response
to the inspector's
questions,
the licensee
reviewed their
purchasing
documentation
and determined that the purchase
order for the
sample valves specified
maximum temperature,
but did not indicate that the
valves would be exposed to thermal cycles.
The inspector
asked if the vendor
would have provided these particular valves if the purchase
order
had
indicated- exposure to thermal cycles.
The licensee
indicated that the vendor
stated that the valve is an excellent design
and is appropriate
even in a
thermal cycling application.
Also, the vendor stated that cracks
have not
been
observed
in similar valves at other nuclear facilities.
The licensee
investigated
the nuclear plant reliability data system data
base
and found no similar failures.
They also discussed this with other licensees
identified in the nuclear plant reliability data
system data
base
and found
that none
had
used these valves in similar applications.
The licensee
also
discussed
these findings on the Nuclear Network.
The inspector
found these
actions to be appropriate.
11.2
Closed
Ins 'ection Followu
Item 528 9409-02:
Blowdown
S stem Water
Hammer Event
This item concerned
a water
hammer event in the Unit I steam generator
blowdown system which occurred
when the blowdown system containment isolation
valves were opened to return the blowdown system to operation.
The blowdown
system
had
been depressurized
for maintenance.
The inspector
noted that the operating
procedure for the system included
a
caution that
a water
hammer could occur when opening the containment isolation
valves.
The inspector
was concerned that this step indicated that water
hammers in the blowdown system
may be routine and that appropriate corrective
actions to prevent the water
hammer events
had not been taken.
In addition,
the inspector questioned
whether the licensee
had evaluated
the hydraulic
impact on the system
components.
The licensee
reviewed the available work history to determine the last time
a
blowdown system water
hammer
had occurred.
They determined that the most
recent event
had occurred in mid-1990 and that the procedure caution
had
been
added to following the event.
The system engineers
recalled similar events
during initial plant startup.
They also identified
a modification, which had
been in a hold status,
to modify the blowdown system to prevent
a water
hammer
event.
The inspector
concluded that these
events
appeared
to be infrequent.
0
I
I
'
e
-25-
The licensee
performed
an evaluation of the stresses
placed
on the blowdown
system during the water
hammer event.
They determined,
based
on the force
required to cause
the
damage to the pipe supports,
that the forces
had not
exceeded
the allowable forces for the system.
The licensee initiated
a design
change request to modify the system to include
a bypass line, with three series isolation valves,
around the outboard
containment isolation valves in all three units.
The licensee
plans to revise
the procedure for returning the blowdown system to operation to require that
the bypass line be opened to equalize
system pressure prior to opening the
containment isolation valves.
The licensee
planned to expedite this
modification in all units.
The inspector
reviewed the
CRDR associated
with this event
and noted that it
addressed
the engineering
aspects
of the event,
but did not address
the
operations
aspects.
In response
to this comment,
engineering
issued
a memo to
all operations
managers,
informing them that taking the blowdown system out'f
service with the units at power could lead to a water
hammer event
when the
system is returned to service.
The inspector also learned that
a night order
had
been written to advise operations
personnel
of this event
and that initial
operator training includes
sessions
which focus
on identifying and preventing
water
hammer events.
This item is closed.
4
'I
1
l
l
I
il,
r
0
1
PERSONS
CONTACTED
ATTACHMENT
1. 1
Arizona Public Service
Com an
J. Bailey, Assistant Vice'President;
Nuclear Engineering
P. Coffin, Engineer,
Nuclear Regulatory Affairs
E. Dutton, Supervisor, guality Control, Unit 2
A. Fakhar,
Manager, Site Mechanical
Engineering
R. Flood, Plant Manager,
Unit 2
D. Garchow, Director, Site Technical
Support
W. Ide, Plant Manager,
Unit
1
A. Krainik, Manager,
Nuclear Regulatory Affairs
D. Larkin, Senior Engineer,
Nuclear Regulatory Affairs
J.
Levine, Vice President,
Nuclear Production
D. Mauldin, Director, Site Maintenance
and
Modifi'cations'.
Russo,
Manager,
Maintenance
Nuclear Assurance
J. Scott, Assistant Plant Manager,
Unit 3
C.
Seaman,
Director, Nuclear Assurance
G. Shanker,
Department
Leader,
Engineering Nuclear Assurance
E. Simpson,
Vice President,
Nuclear Support
J. Velotta, Director, Training
P. Wiley, Manager,
Operations,
Unit 2
1.2
NRC Personnel
B. Olson, Project Inspector',
WCFO
1.3
Others
J. Draper, Site Representative,
Southern California Edison
B. Drost, Engineering .and Operations
Committee Alt., Salt River Project
F. Gowers, Site Representative,
El
Paso Electric
All personnel
listed were in attendance
at the exit meeting held with the
NRC
resident
inspectors
on June
15,
1994.
2
EXIT MEETING
An exit meeting
was conducted
on June
15,
1994.
inspectors
summarized
the scope
and findings of
acknowledged
the inspection findings documented
did not identify as proprietary any information
the inspectors.
During this meeting,
the
.
the report.
The licensee-
in this report.
The licensee
provided to, or reviewed by,
L
d)
J
I
t
f
l
j
l,'
I
0