ML17311A588
| ML17311A588 | |
| Person / Time | |
|---|---|
| Site: | Palo Verde  |
| Issue date: | 01/10/1995 |
| From: | Wong H NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML17311A585 | List: |
| References | |
| 50-528-94-34, 50-529-94-34, 50-530-94-34, NUDOCS 9501240405 | |
| Download: ML17311A588 (56) | |
See also: IR 05000528/1994034
Text
APPENDIX B
U.S.
NUCLEAR REGULATORY COMMISSION
REGION IV
Inspection
Report:
50-528/94-34
50-529/94-34
50-530/94-34
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:
Maricopa County,
Tonopah,
AZ
Inspection
Conducted:
October
16 through
November 30,
1994
Inspectors:
K. Johnston,
Senior Resident
Inspector
H. Freeman,
Resident
Inspector
A. MacDougall, Resident
Inspector
J.
Kramer, Resident
Inspector
Approved:
. J.
Wong,
C
,
eact
roJects
Branc
lo
ate
Ins ection
Summar
Areas
Ins ected
Units
1
2
and
3
Routine,
announced
inspection of plant
status,-onsite
response
to events,
operational
safety verification,
maintenance
observations,
onsite engineering,
plant support activities,
operations,
maintenance,
and engineering
followup, and licensee
event report
review.
In addition, TI 2515/126 concerning
the performance of online
maintenance
was performed.
Results
Units
1
2
and
3
0 erations
Conduct of operations
weaknesses
contributed to significant operator errors
involving balance of plant equipment during the Unit 2 restart following a
midcycle outage
(Section 2.2).
Licensee
management
took aggressive
actions to
address
these
weaknesses
by revising crew expectations,
reassigning
crew
leaders,
and making significant organization
changes
(Section 3.4).
9501240405 950fi7
ADOCK 05000528
I'<
il',
tl
Operations
management
made appropriate operability determinations
regarding
the erratic behavior of the pressurizer
head vent valves
and ultimately shut
down Unit
1 to repair the valves
(Section 3.6).
Additionally, operations
management
responded
appropriately to minor weaknesses
identified by
inspectors
in the application of temporary
procedure
changes
(special
variances,
Section 3.5), with the consistent
positioning of control rod
automatic controls
(Section 3.3)
and with the procedure
governing
independent
verifications (Section
2. 1. 1).
However, operations
management
was slow to
assess
the Technical Specification
(TS) implications of the failure of a
Unit
2 safety injection valve to operate
and communicate
these
implications to
the other two units (Section
2. 1.2).
The
NRC inspector's
identification of switched
lamp covers
on
a remote
shutdown
panel
in Unit 2 highlighted the need for operators
to be alert for
discrepant
conditions
(Section 3. 1).
Maintenance
'While most maintenance
activities observed
during the period were performed
satisfactorily,
the inspectors
found that
a weak procedure
was
used to perform
significant work on pressurizer vent'alves
even though these
procedure
weaknesses
had
been previously identified (Section 4).
The licensee
did not adequately
implement commitments'involving the control of
maintenance activities with an impact
on vital
DC power systems
in Unit 2 and
in the switchyard.
In'hree instances,
work in these
areas
was not adequately
controlled (Section 3.2).
The licensee
has
demonstrated
a carefully considered
approach
to the
performance of online maintenance
on risk significant equipment.
Additionally, enhancements
were planned to improve the ability for operations
and maintenance
personnel
to assess
the risk significance of emergent
work
(Section 7).
En ineerin
Engineering
provided strong support
and analyses
of the spent fuel pool
shuffle (Section
6. 1),
a
10 CFR Part 21 regarding certain motor-operated
valves '(MOV) (Section
2. 1.3),
and the risk assessment
for online maintenance
activities (Section 7).
However, they did not ensure that design
assumptions
for certain safety injection valves were adequately
communicated
to plant
operations
or factored into operating
procedures
and practice
(Section
2. 1.2).
Plant
Su
ort
Inspectors
observed
a successful
emergency
preparedness
drill in which the
licensee
demonstrated
good communications
between
the simulator, technical
support center,
and the emergency
operations facility (Section 5).
I
f
"I
Security took appropriate
actions
in response
to an access
badge left in a
locker
and
a degraded
access turnstile (Section 8).
Hang ement Overview
Hanagement's
efforts in the area of operations'xpectations
and changes
in
the crew staffing and organization
demonstrated
an understanding
of conduct of
operations'eaknesses.
The findings concerning
the implementation of commitments
on the control of
maintenance
on Unit 2 125-Vdc power
systems
and access
to the switchyard,
shortly after those
commitments
were made,
indicated that management
had not
adequately
communicated
these
commitments to site personnel.
Summar
of Ins ection Findin s:
~
Two unresolved
items were identified involving the failure of a Unit 2
safety injection valve to open during testing
(Section
2. 1.2)
and the
adequacy of maintenance
instructions for solenoid operated
valves
(Section 4.1).
~
One deviation
was identified involving three instances
where the
licensee
did not meet
commitments
made to control Unit 2 maintenance
activities involving 125-Vdc equipment,
and in the switchyard
(Section 3.2).
~
TI 2515/126 concerning
the performance of online maintenance
was
performed
(Section 7),
~
Violations 528/9422-01
and 530/9420-01
were closed.
~
Followup items 528/9355-02,
529/9402-02,
and 528/9348-04
were closed.
~
Licensee
Event Reports
528/94-05,
Revision
1, 529/91-05,
529/94-02,
530/94-05,
530/94-06,
and 530/94-07
were closed.
Attachments:
~
Attachment
A - Persons
Contacted
and Exit Heeting
~
Attachment
8 - List of Acronyms
I'
I
e
1
PLANT STATUS
1.1
Unit
1
Unit
1 operated
at 98 percent
power for most of the inspection period.
On
November
26,
1994,
the unit commenced
.a shutdown required
by the plant
TS to
repair two leaking pressurizer
steam
space
vent valves
(Section 3.6).
The
valves
were repaired
in Mode
4 and the reactor
was brought critical on
November 28.
The main generator
breakers
were closed
on November 30.
At the
end of the inspection period, reactor
power was
18 percent
and being raised to
98 percent.
1.2
Unit 2
Unit
2 began the inspection period in Mode
3 preparing for reactor startup
after completing
a midcycle outage for steam generator
tube inspections.
On
October
16,
1994,
the reactor
was
made critical and entered
Mode 1.
1
On October
17, the unit was placed
on the grid and
power raised to
19 percent.
The feedwater
system failed to swap over from the downcomer to the economizer
and power was reduced
to
15 percent to allow repairs.
The licensee
tripped
the turbine after
an operator error resulted
in a main feedwater
pump trip
(Section 2.2. 1),
and the reactor
was stabilized in Mode
2 at approximately
two
percent
power.
On October
19, the licensee
manually tripped the reactor
and entered
Mode
3
when they were unable to restore
a safety injection valve within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />
after it failed to open during testing
on October
16 (Section
2. 1).
The
following day, after completion of the valve repairs,
the unit was taken
critical, placed
on the grid,
and
began
power ascension
to
100 percent.
On October
29, the unit sustained
a reactor trip as
a result of a protection
system
equipment failure (Section 2).
On October 30, the licensee
completed
the equipment repairs
and the unit was taken critical.
On October 31, the
unit entered
Mode 1, placed
on the grid,
and
began
power ascension
to 100
percent
power.
The unit remained at essentially
100 percent
power through the
remainder of the inspection period.
1.3
Unit 3
Unit 3 began
the inspection period
and remained at essentially
100 percent
power until the unit began
a planned
on November 25.
The unit ended
the inspection period in Mode
5 conducting
a midcycle outage for steam
generator
tube inspections.
2
ONSITE RESPONSE
TO
EVENTS (93702)
2. 1
Unit 2 Hi
h Pressure
Safet
In ection Valve
Failed to 0 en Durin
~Testin
On October
15,
1994, Unit 2 conducted
a plant heatup to normal operating
temperature
and pressure.
During the heatup,
operators
received
a 'safety
injection (SI) check valve high pressure
alarm and noted that the pressure
was
above the alarm setpoint of 1000 psig.
This alarm indicated that the check
valve between
(RCS)
and the SI line to
RCS Loop 2B
was leaking..
The following day,
a HPSI isolation valve,
two check valves
upstream of the line which had the high pressure
alarm, failed to open during
testing.
The initial investigation determined that the
HPSI valve
may have
failed due to pressure
binding.
The inspector
reviewed the following:
~
the operators
response
to the high pressure
alarm,
~
the licensee's
review of the failure of the
HPSI valve to open,
and
~
a
10 CFR Part 21 notification regarding
pressure
binding in HPSI valves
similar in design to the one that failed to open.
2. 1. 1
SI Line High Pressure
Alarm
On October
15, operators
in Unit 2 received
a SI check valve high pressure
alarm.
They noted that the pressure
was
above the alarm setpoint of 1000 psig
and determined that reactor coolant
Loop 2B had
a leaking check valve.
The
operators
determined that the leakage
was less
than the
one
gpm TS limit and
attempted
to lower the pressure
and clear the alarm in accordance
with the
response
procedure
(ARP).
The
ARP had
a step to open Valve SIB-628, the SI line to
Loop
2B drain
valve, to lower the pressure
and clear the alarm.
The inspector
noted that
applicable
procedures
seemed
to indicate that operators
should
be performing
an independent verification
( IV) of valve position every time drain Valve SIB-
628 was closed.
However,
the inspector
noted that while operators
opened
and
closed
Valve SIB-628 at least
ten times,
they performed
an
IV only the final
time the valve was closed.
The
ARP referenced
a procedure
governing
IYs for guidance
on when to perform
an IV.
The IV procedure
stated that
an
IV was not needed for control
board
manipulations with valve position indication (as
was the case with
Valve SIB-628) unless directed
by another
procedure.
While this apparent
logic loop provided confusion,
operators
met the intent of the IV procedure.
The licensee
determined that documenting
the
IV of valves operated
from the
control
room was not required
because
valve position indication was
/
0
continuously available.
The licensee
agreed that the
IV procedure
was poorly
written and subject to interpretation
and planned to revise the procedure.
The Unit 2 Operations
Department
leader also wrote
a night order to all three
units discussing this issue
and providing examples
on
how to properly perform
IVs.
As
a long term action,
the licensee
planned to review other operations
procedures
arid remove references
to the
IV procedure for valves operated
from
the control
room.
The inspector
concluded that these
actions
were
'ppropriate.
2'.2
HPSI Valve Failed
To Open
On October
16,
1994, Unit 2 HPSI Valve SIA-UV-627- failed to open during
a
routine
ASME Section
XI stroke test,
Unit
2 had just entered
Mode
1 following
a midcycle outage
when operators
performed the test.
Valve SIA-UV-627 is one
of eight
HPSI injection valves that opens
in conjunction with a HPSI
pump
actuation,
The licensee
decided to replace
the motor operator
on Valve SIA-UY-627 with a
'larger operator
in accordance
with an existing design modification package.
Additionally, th'ey replaced
the valve internals.
They were unable to complete
the repairs within the
72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> allowed outage
time and shutdown Unit 2 on
October
19.
Repairs
were completed,
and the Unit 2 returned to Mode
1 on
October
20.
The licensee
postulated
that Valve SIA-UV-627 may have
been pressure
bound.
Two of the three
between
the valve
and the
RCS were
known to have
been leaking.
As noted in Section
2. 1. 1, operators
had identified that
a
pressure
indicator at the discharge of the associated
valves
downstream of Valve SIA-UV-627, was reading at
RCS pressure
(roughly
2250 psig).
Licensee
valve engineers
noted that their calculations
supporting the operator
size for Valve SIA-UV-627 were based
on
a maximum differential pressure
across
the valve of 1050 psig.
The maximum differential pressure
had
been
based
on
a
high pressure
alarm of 1000 psig
on the pressure
indicator at the discharge of
the SIT.
On October
21, the inspector
noted that operators
in Units
1 and
3 were
.unaware of the problems with Valve SIA-UV-627 in Unit 2 and were not aware
that
a pressure
of greater
than
1000 psig at the SIT discharge
could imply
that the
HPSI injection valves might not open.
The inspector questioned
whether
an indication of greater
than
1000 psig would require operators
to
enter
TS 3.0.3 since the indication would apply to both
A and
B train HPSI
injection valves associated
with one cold leg injection path.
Operations
management
concurred that, until
an engineering
evaluation could
be completed,
this would be appropriate
and revised
the
ARP for downstream
pressure
alarms
requiring entry into TS 3.0,3
and actions to bleed pressure off the line.
The licensee
issued
Licensee
Event Report
on November
18,
1994
describing
the plant shutdown required
by the
TS.
The
LER indicated that
a
0
0
ropt cause
investigation
was underway to determine if Valve SIA-UV-627 was
pressure
bound.
The licensee
planned testing involving a mockup of the
components
that were removed.
Additionally, the licensee
informed the
inspector that they were investigating the following:
~
If the cause
was pressure
binding,
was the
HPSI injection line. for Cold
Leg
2B inoperable while downstream
pressure
was
above
1050 psig?
The
licensee
speculated
that the coincident
pump start could have
reduced differential pressure
prior to the injection valve operators
failing.
Why was the basis for Valve SIA-UV-627 not fully communicated
to
operations?
If the valve engineers
assumed
that downstream
pressure
-should
be less
than 3050 psig,
why was this not adequately
communicated
to operations?
~
Are there other
HOVs in which operations
assumptions
were made,
but not
adequately
communicated?
Based
on these
questions,
the inspector
considered
this to be
an Unresolved
Item (529/9434-01).
2. 1.3
Borg Warner Globe Valve
10 CFR Part 21 Notification
On October
25,
1994,
Borg Warner issued
a
10 CFR Part 21 notification
concerning
a potential
design
problem with 6-inch,
900 psi, motor-operated,
Y-globe valves.
The manufacturer
noted that the stem thrust required to close
the valve should
be based
on the differential pressure
(DP)
on the valve guide
area
instead of the nominal valve seat
area.
This conclusion
was
based
on
testing
conducted
during the Electric Power Research
Institute
(EPRI)
performance prediction program.
The tests
showed that,
in some cases,
the
required
stem thrust using the valve guide area
was more than double the value
calculated
using the seating
area.
As
a result,
the valves
may not have
enough thrust to close
under actual
DP conditions.
The inspector
asked
the licensee if they were
aware of the
notification and if they had
any of the susceptible
valves.
The inspector
was
informed that the
HPSI isolation valves,
including Valve SIA-UV-627, were
2-inch,
Borg Warner,
Y-globe valves
and were susceptible
to the problem.
There were eight of these
valves in each of the three units.
The licensee
had
previously determined
during
DP testing that the required thrust to close
these
valves
was higher than the calculated thrust in some cases.
The
licensee
found that in every case
the motor operators
had enough available
thrust to stroke the valves during design basis
DP testing.
The inspector
concluded that there
was not
a safety concern
because
the valves
had
been
successfully
DP tested.
1
The licensee
had previously issued
two condition reports/disposition
requests
(CRDRs) to evaluate
the problems identified during the
EPRI tests
and in the
(
,i
1
10 CFR Part 21 notification.
The licensee
determined that their
HOV 'program
required
a
CRDR to be initiated if the valve factor identified during
DP
testing
was greater
than
85 percent of the
assumed
valve factor.
The
inspector
reviewed the evaluations
and concluded that the licensee's
HOV
program compensated
for the
phenomena
described
in the
notification.
2.2
Unit
2 Restart
Followin
Hide cle Outa
e
On October
16,
1994, Unit 2 was in the final preparations
for restart
following a midcycle outage to inspect the steam generators.
During the
efforts to restart
the unit,
a number of conduct of operations
problems
resulted
in plant events.
Host significant were
a mai,n feedwater
pump trip
and
a main generator trip, both caused
by operator errors.
These
two events
are discussed
below.
In response
to these
events,
on October
20, operations
management
mandated
changes
to the conduct of operations
which they had planned to implement in
'late-November.
These
changes,
which were developed
through the re-engineering
study,
are discussed
in Section 3.4.
2.2.
1
Hain Feedwater
Pump Trip
On October
17,
1994, with reactor
power at
15 percent
and only one main
pump operating,
a reactor operator
(RO) performed
a main feedwater
pump stop valve test,
isolating
steam to the main feedwater
pump.
As
a
result,
level in both steam generators
decreased
rapidly,
and operators
manually initiated auxiliary feedwater
and manually tripped the main turbine
to prevent
a reactor trip on low steam generator level.
The licensee initiated
an investigation of this event.
The licensee
concluded
that the following factors lead to the turbine trip:
~
The
RO performed the stop valve test without using the procedure.
The
procedure
for performing the stop valve test required that the stop valve
test
be performed
when the feedwater
pump turbine is operating
on low
pressure
steam only and the high pressure
control valve is closed.
The
RO did not effectively communicate
his intentions to perform the test
to the rest of the operating
crew or shift management.
~
A test interlock, which should
have prevented
the stop valve from going
closed with the high pressure
control valve open,
was out of adjustment.
The licensee
performed several
corrective actions.
The licensee
removed the
RO involved in the event from licensed duties,
pending operations
management
review of the event.
The licensee
adjusted
and retested
the test interlock
associated
with the high pressure
stop valve.
Additionally, licensee
management
implemented
a set of immediate actions
in order to improve
formality, command
and control,
and performance of the control
room staff
(Section 3.4).,
The inspector
reviewed the licensee's
investigation
and corrective actions
and
concluded that licensee
management
took prompt
and thorough action in response
to the event.'ubsequent
to the implementation of the corrective actions,
the
inspector
noted
an
improvement
in the formality, and
command
and co'ntrol of
the control
room staff.
2.2.2
Generator Trip Due to Auxiliary Operator
(AO) Error
On October
16,
1994,
an
AO misinterpreted
an instruction from the Unit
1
control
room and attempted
to manually close
a motor-operated
disconnect
during the performance of an energy control center switching procedure
and
caused
a main generator trip during the unit startup.
At the time, the main
generator
was turning, but the field was not excited.
The Shift Supervisor
(SS)
had directed
the
AO to verify that the motor operator
mechanism
was
locked in the coupled position
and to remove the energy control center tag.
The licensee
performed
an evaluation to determine if there
was
any damage to
the generator
(which was running at
1800 revolutions per minute
and not
excited),
the motor-operated
disconnect,
and generator
output breakers.
The
licensee
determined
no damage
occurred
and synchronized
the generator
to the
grid several
hours later.
The licensee initiated
an investigation of the event
and identified the
following weaknesses
during the evolution:
~
the
AO did not attend the prejob briefing that was required
by the
sensitive
issues
manual,
~
the
AO did not use
an approved
procedure
to perform the task,
and
, ~
there
was
a communications
breakdown
between
the
AO in the switchyard
and
the Unit 2 control
room.
The inspector
reviewed the licensee's
investigation
and found it to be
thorough.
The inspector will monitor the licensee's
corrective actions
as
part of a future routine inspection.
2.3
Unit 2 Reactor Tri
On October
29,
1994,
the unit sustained
a reactor trip on
a low Departure
from
Nucleate Boiling trip signal
as
a result of an equipment failure when
Calculator
1 inserted
large penalty factors into the core protection
calculator.
The reactor trip was
an uncomplicated
reactor trip and the unit
was stabilized
in Mode 3.
0'
l
i,
-10-
The licensee
investigated
and repaired
CEA Calculator
1 by replacing the main
processor.
The licensee initiated
a
CRDR to evaluate
the main processor
board
failure.
On October 30, the unit was taken critical and returned to the grid
the following day.
The inspector
responded
to the plant following the unit trip.
The, inspector
determined that the operators
responded
adequately
to the reactor trip.
The
inspector
noted
no deficiencies
in the subsequent
reactor startup
and
power
ascension.
3
OPERATIONAL SAFETY VERIFICATION
(71707)
3. 1
Remote
Shutdown
Panel
Li hts
On October
17,
1994, during routine plant walkdown activities, the inspector
identified that indicating lamps
on the remote
shutdown
panel
(RSP) in Unit 2
were in'correct.
The inspector
observed that several
red, green,
and white
lamp covers,
which indicate equipment status,
had
been
swapped.
The inspector
'notified the
SS of the problem.
The inspector
and
SS inspected
both of the
Unit
2
and identified that approximately
20 lamp covers
had
been
swapped,
including changes
between
the two RSPs.
The licensee
subsequently
returned
the panels to their appropriate configuration
and confirmed operability of the
panels.
The licensee
inspected
the
in Units
1
and 3,
and other control
and
indication panels
in Unit 2, for proper
lamp covers,
and identified no
additional discrepancies.
The licensee initiated
a
CRDR and
an investigation
to determine
the cause of the event.
The investigation did not reveal
the
person
who swapped
the lamp covers.
The inspector
reviewed the scope of the
licensee's
investigation
and found it to be thorough.
As
a corrective action,
the licensee
issued
a night order to alert all
operations
personnel
to heighten
awareness
concerning
unusual
plant equipment
alignment
and suspicious activities.
The licensee
planned to install locking
mechanisms
on the rear doors to the
on all three units,
and modify the
operations
preventative
maintenance
procedures
to include drawings of the RSP.-
In addition,
the licensee
planned to install automatic control
access
devices
(ACAD) readers
at the
room entrances
in all three units.
The inspector
concluded that operations
performed
a prompt
and thorough
verification of other remote
panel light indications.
The inspector
determined that while the repositioning of the indicating lamps could
initially cause
operator confusion,
the configuration
posed
no significant
problem.'he
inspector concurred with the licensee's
conclusion that the
was operable
at all times.
3.2
Failure to Meet Commitments
Made in Unit 2 Batter
TS Amendment
On October
13,
1994,
the
NRC granted
an
amendment
to the Unit 2 TS 4.8.2. l.e,
"DC Sources - Operating," to allow Unit 2 to startup
and operate with 125-Vdc
I $
1
-11-
batteries
that did not meet the
TS surveillance
requirement for minimum
battery capacity
(NRC Inspection
Report 50-528/94-31;
50-529/94-31;
50-530/94-31).
In their request for the
TS amendment
dated October
9 and
supplemented
on October
12, the licensee
committed to provided the following
additional controls
and limitations on the maintenance
of important equipment:
~
(PRA) would be used to review al'l
125-Vdc
system
and related auxiliaries corrective/preventative
maintenance
work.
~
Access to the switchyard would be limited.
All emergent
switchyard work
would be reviewed
by the Unit
1 SS.
Shortly after the
TS amendment
was issued,
three events
took place which
demonstrated
that the licensee
had not fully implemented
these
commitments.
These
events,
discussed
in detail
below, involved:
~
The construction of scaffolding in two 125-Vdc equipment
rooms for
seal
work which was not reviewed prior to the start of work.
This was identified by the licensee.
An electrician,
while attempting to reposition
a control
board indication
socket,
caused
a short circuit to ground
on
a 125-Vdc bus.
NRC personnel
found
a switchyard gate
open after -hours
and
had apparently
been left open
by switchyard (nonutility) workers.
3.2. 1
Working in Two Vital 125-Vdc Equipment
Rooms Without
PRA Review
On October
18,
1994,
the licensee initiated the construction of scafFolding in
two Unit 2 125-Vdc equipment
rooms for penetration
seal
work.
On October
25,
the shift technical
advisor noted that the work had not been evaluated
by the
PRA group.
The
PRA group subsequently
reviewed the work and allowed it
to
continue in one equipment
room at
a time.
The licensee
subsequently
initiated
a
CRDR to assess
how this work was missed
and to initiate corrective actions.
The inspector
concluded that the licensee
took prompt action to stop work in
both equipment
rooms
and evaluate
the risk of the work being performed.
The
inspector
noted the alertness
of the
STA to recognize
the problem.
The
inspector
noted that this was
an example of the licensee's
work control
process failing to highlight work requiring
a
PRA review in accordance
with
the
TS Amendment
71,
3.2.2
Working on
a Control
Board Light Socket Without a Work Package
and
Without
PRA Review
On October
26,
1994,
the inspector
observed
arcing in a lamp socket while
performing
a control
board walkdown in Unit 2.
The inspector determined that
t
the arcing
was due to
an electrician attempting to insert the lamp socket
back
into its original position.
The electrician
subsequently
caused
a ground
on
a
e
-12-
vital 125-Vdc bus.
Operations
personnel
directed the electrician to pull the
lamp socket
away from the control
board
and the ground cleared.
The inspector questioned
the electrician
and determined that
he did not have
a
work order
(WO) to perform repairs
on the socket.
The licensee
indicated that
reinserting
1'amp sockets
was
a routine. activity and is often performed without
a
WO.
Based
on the comments,
the licensee
subsequently
issued
a WO'o
complete
the repairs to the
lamp socket
and
a
CRDR to evaluate
the
need for a
WO on this type of activity.
Nevertheless,
since operations
considered
the
lamp socket
work
a routine activity, the, inspector
concluded that the licensee
did not meet the commitments of the October
9 and
12,
1994 letters in that the
licensee
attempted corrective maintenance
on
an indicating lamp socket
powered
by
a 125-Vdc system without prior
PRA review.
3.2.3
Gate Left Open
On October 20,
1994, at approximately
6 p.m.,
NRC personnel
observed that
a
gate leading to the site switchyard
was open.
The gate
was to the main
switchyard, is controlled
by the Salt River Project
(SRP) electric utility,
and is outside the protected
area,
There did not appear
to be anyone working
in the switchyard at the time.
Site security
was notified and the gate
was
secured
at approximately 6:20 p.m.
The inspector notified plant management
of this finding and the licensee
performed
an investigation.
The licensee
contacted
SRP to determine
how the
gate
had
been left open.
SRP technicians
explained that the gate,
which is
not normally used for access
to the switchyard,
was probably opened to provide
a quicker route to the site cafeteria.
The technicians
had not closed the
gate
when they returned
from lunch, or closed it when they left the switchyard
for the day.
The licensee
evaluated this issue in light of their commitment to limit access
to the switchyards.
They determined that action
had
been taken to notify SRP
of the increased
sensitivity to switchyard activities.
However,
was not
notified formally nor at
a high level.
The licensee
subsequently
performed
followup discussions
with SRP management
to reiterate
the sensitivity towards
access
control.
As
a corrective action,
the licensee
was considering establishing
a primary
contact with SRP
and revising switchyard administrative controls procedures
to
ensure that appropriate
communications
are
made in the future when plant
conditions,
such
as the
RCS at midloop, warrant greater
switchyard control.
3.2.4
Conclusion
The inspector considered
these three incidents to be failures of the licensee
to adequately
implement the commitments
made to support the October
13,
1994,
TS amendment
(Deviation 529/9434-02).
l
l'
-13-
3.3
Unit 2 Control
Rods in Hanual
On November 3,
1994,
the inspector
noted that the Unit 2 mode select
switch
(HSS) for the reactor regulating
system
was in manual
sequential
(manual) vice
automatic sequential
(auto).
The inspector questioned
the operators
about the
position of 'the switch and determined that the
HSS was in manual
due to
temperature
oscillations
on the input to the reactor regulating
system
from a
hot leg temperature
instrument.
The inspector
noted that while
a similar
problem exists
on Unit 1, operators
in that unit maintain the
HSS in auto.
The inspector
asked operations
management
to explain the difference.
The licensee
determined that the Unit 2 operators
had placed the
HSS in manual
to address
the possibility of a single failure of the redundant
temperature
input causing
the control
element
assemblies
(CEAs) to automatically respond.
Unit
1 operators
had kept the
HSS in automatic to allow the
CEAs to respond to
a reactor cutback.
Operations
management
determined that the possibility of a
reactor cutback
was more likely than the failure of the temperature
input and
that the
HSS in both units should
be in auto.
The licensee
and Combustion
'ngineering '(CE) were continuing to monitor and evaluate
the hot leg
temperature
anomalies
that
have
been
occur ring on
CE reactors.
The licensee
had initiated
a
CRDR to document
and track the concern
(NRC Inspection
Report
50-528/94-22)
50-529/94-22;
50-530/94-22).
The inspector
agreed with the licensee's
conclusion to place the
HSS in auto.
While the inspector also noted that the licensee
had not addressed
the problem
until questioned
by the inspector,
the licensee
responded
quickly to the
questions.
3.4
0 erations
Standards
On October
19,
1994,
the Executive Vice President
issued
a memo to operations
listing several
revised expectations
for the conduct of operations
and stating
that
he expected
a step
change
in their performance
(NRC Inspection
Reports
50-528/94-34;
50-529/94-34;
50/530/94-34).
On November 7, the director of
operations
held
a day long seminar with the operations
department
leaders
emphasizing
management
expectations
for higher standards
of command
and
control, formality, and professionalism for the operations
department.
The
inspector
found the meeting to be very constructive.
A second
meeting
was
held
on November
18.
The inspector
noted
a general
improvement in control
room operations.
The
inspector
observed that the noise level
and
amount of control
room traffic was
reduced.
In addition,
the control
room communications
between
operations
and
other disciplines
imp'roved.
The inspector
noted that the licensee
intended to
implement the revised operations
standards
in November but implemented
them
earlier due to recent operations
events,
0'
-14-
3.5
S ecial Variance
on Unit 3 Steam Generator
Blowdown Alarm
On October
26,
1994,
the inspector
reviewed the special
variance log in
Unit 3.
Special
variances
are temporary
changes
to procedures
necessitated
by
plant conditions which are anticipated to be temporary.
The inspector
noted
that the log contained
an active special
variance that
was over 8 months old
and
one
on the steam generator
blowdown system alarm response
procedure.
The
licensee initiated the special
variance
on the alarm response
procedure
due to
a degraded
heat
exchanger.
A degraded
heat
exchanger
caused
a high
temperature
alarm in the control
room during certain
blowdown system
alignments
when ambient temperatures
were high.
Subsequent
to initiating this
variance,
the licensee
installed
a temporary modification which injected .cool
water into the ion exchanger
and the system
was
no longer affected
by ambient
temperatures.
The inspector questioned
the onshift crew to determine
whether
the variance to the alarm response
was still required following the
modification.
The
SS concluded that it was not and directed that it be
closed.
The inspector discussed
management's
expectations
concerning
the
use of
special
variances
with the Unit 3 operations
department
leader.
The
operations
department
leader stated that the special
variance
should
be used
when circumstances
required
a deviation from approved
procedures.
The
variation could
be used for a "one time" evolution or be used
numerous
times
until equipment repairs
allow the use of the original procedure.
The
operations
department
leader
expected
the
SS to be familiar with all active
special
variances
and committed to ensure that this was understood
by all
crews.
The inspector
concluded that the use of special
variances
was acceptable.
The
inspector
reviewed licensee
procedures
for special
variance,
for conduct of
operations,
and for shift turnover.
The inspector
noted that none of the
procedures
required
a review of the special
variance log.
Finally, the
inspector
concluded that the two special
variances
in question
had
no safety
impact
on the plant.
3.6
Unit
1
Pressurizer
Vent Valve 0 erabilit
Determination
On November 26,
1994,
the licensee
shut
down Unit
1 to comply with'he
TS
action statement
for an inoperable pressurizer
steam
space
vent path.
The
licensee
had declared
the vent path inoperable
due to the erratic behavior of
the pressurizer
vent Valve RC-103,
and the isolation valve to the reactor
drain tank (RDT), RC-105.
The leakage
from the pressurizer
vent path through
Valves
RC-103
and
105 was first noted in Hay 1994
and
had degraded
significantly since that time.
The inspector
reviewed the licensee's initial
as well as subsequent
evaluations.
In Hay 1994,
the licensee
observed
leakage
from the pressunzer
vent path
through Valves
RC-103
and
RC-105.
The leakage
caused
the pressure
in the
piping between
Valves
RC-103
and
RC-105,
as indicated
on pressure
instrument
(PI)
138, to stabilize at
RCS pressure
of 2250 psi.
The operators
noted that
l
,
t
-15--
after several
days,
the pressure
on PI-138 began to slowly lower until the
closed indication for Valve RC-103 went out.
The closed indication remained
out for about
10 seconds
and
th'en the pressure
on PI-138 increased
and again
stabilized at 2250 psi.
This sequence
occurred
once or twice
a day and then
would not reoccur for up to several
weeks.
The licensee
performed
an
operability determination to consider whether the pressurizer
vent path
was
"closed"
as required
by TS 3.4. 10 when the closed
(green
lamp) indication for
Valve RC-103
was not illuminated.
The licensee
determined that the vent path
was closed
because
redundant
Valve RC-105
had always indicated closed.
The
inspector
reviewed the initial operability determination
and
agreed with the
licensee's
conclusions.
On the morning of November 9, Unit
1 operators
received
several
reactor
coolant vessel
vent/seal
trouble alarms followed by an
RDT high pressure
alarm.
Operators
noted that the pressure
on PI-138 was oscillating between
2250 psi
and
1750 psi every
1 to'
minutes.
During the oscillations,
Valve
RC-103 closed indication would go out and the
open indication would illuminate
for a few seconds
when the indicated pressure
reached
the low end of the band.
The oscillations continued for several
hours until pressure
again stabilized
at 2250 psi.
The licensee
reviewed the original operability determination
and
concluded that the condition observed
on November
9 did not cause
the system
to be inoperable
because
Valve RC-105 always
had indicated shut.
The
inspector
agreed with the licensee's
The inspector
also concluded that the licensee
was sensitive to the changes
in the observed
behavior of the pressurizer
vent system
and the
need to continually assess
the
operability of the system.
On November
24, oscillations
on PI-138
and additional
RDT high pressure
alarms
were noted.
However, rapid pressure
increases
were noted in the
RDT and
on
several
occasions
the closed position indication
on Valve RC-105 was lost,
The licensee
subsequently
questioned
the continued operability of the
pressurizer
vent path since both Valves
RC-103
and
RC-105 lost closed
indication at the
same time.
The licensee
decided to manually isolate the
pressurizer
vent path
and entered
a 72-hour
TS shutdown action statement.
On
November
26, the licensee
shut
down the plant
and entered
Mode
4 to repair
Valves
RC-103
and
RC-105 (Section
4. 1).
The inspector concluded that the
licensee
made
a safe,
conservative
decision to declare
the system inoperable
and shut
down the plant to perform repairs.
4
MAINTENANCE OBSERVATIONS
(62703)
4. 1
Re air of Unit
1 Pressurizer
Head Vent Valves
As noted in Section 3.6 of this report,
on November
26,
1994, Unit
1 shut
down
to repair the pressurizer
head. vent Valve (RC-103)
and the head vent system
drain valve to the
RDT (Valve RC-105).
Both valves
had
been
behaving
erratically.
The inspector
observed
the disassembly
and reassembly
of both valves
and the
subsequent
reinstallation.
The valves. were Target
Rock solenoid-operated
f
I
-16-
valves.
The inspector
observed
that during the reassembly
work, the mechanics
relied heavily on instructions contained
in the vendor technical
manual
(VTH).
However, the
VTH instructions
were not referred to in the maintenance
WO.
.The
WO, which included
some detail
on how to reassemble
the valves,
was
a "model
WO" for the repair of Target
Rock solenoid-operated
valves.
The specific steps
in the
VTH which the mechanics
used involved the'etting of
the stroke
on the pilot valve.
The setup
involved an iterative step in which
tolerances
were measured,
a calculation
was performed to determine if the
stroke
met
an acceptance
criteria,
and, if the acceptance
criteria was not
met, considerable
disassembly,
adjustment,
and reassembly
were performed.
The
inspector
observed that while the mechanics
had
some difficulty initially
doing the iterative step,
they ultimately accomplished
the setup for both
valves.
In
NRC Inspection
Report 50-530/94-26,
the inspector
observed
maintenance
on
an identical valve used
as
a containment isolation valve in the main steam
system
in Unit 3.
The inspector
had identical findings regarding the quality
of the
WO which was derived from the
same
"model
WO".
At that time, the
licensee
had noted that the setup of the pilot valves in Target
Rock valves
was critical to ensure
valve reliability.
In response
to the concern
in
NRC
Inspection
Report 50-530/94-26,
the maintenance
management
initiated
a
CRDR
which included
an action to improve the
"model
WO."
The licensee
had suspected
that the erratic behavior of both Valves
RC-103
and
RC-105 was attributable to degraded pilot valve seats.
This proved to be true
during the disassembly
of the valves.
The inspector
noted that with the
previous
concerns
regarding the quality of the
WO, the importance of the pilot
valve setting,
and the fact that erratic behavior of these
valves
had required
a plant shutdown,
maintenance
had not ensured that the
WOs for Valves
RC-103
and
RC-105
had
been modified.
The inspector
planned to perform
a followup inspection to determine
why
problems continue to exist in the
WO.
This is
an Unresolved
Item (Unresolved
Item 528/9434-03).
4.2
Other Haintenance Activities Observed
The inspector
observed
portions of the maintenance activities noted
below.
The activities observed
were performed appropriately.
~
Unit
1 pressurizer
spray valve maintenance,
~
Unit
1 removal of spent fuel pool blocking plates,
and
~
Unit 2 fire penetration
seal repairs.
-17-
5
DRILL (71750)
On November
16,
1994,
the inspector
observed
the licensee
performance
during
an emergency
plan drill.
The drill involved activation of the emergency
operations facility (EOF)
and limited State
and local participation.
The
drill scenario
involved
tube rupture,
main steam line break,
and
some fuel failure, resulting in an offsite release.
The resident
inspector
responded
to the simulator
and the senior resident
responded
to the technical
support center
(TSC).
The resident
inspector noted
the operations
advisor did
a good job maintaining the professionalism of the
operations
crew.
The inspector noted adequate
communications
between
the
operations
personnel
and
between
the simulator
and the TSC.
The senior resident
observed
good communications
between
the
TSC staff,
and
the simulator
and
EOF.
The inspector
noted that this was the most extensive
drill in which the emergency
response facility data acquisition display system
was used.
The computer
system,
which provided real time plant data to the
and
EOF,
appeared
to provide
a substantial
amount of information, easing
the
event diagnosis
and response.
6
ONSITE ENGINEERING (37551)
6. 1
S ent fuel Assembl
Reshuffle
A TS amendment
was recently approved to change
the storage
mode of the spent
fuel pool from a checkerboard
mode
(two out of every four locations with spent
fuel) to
a high density
mode.
The revised
TS a'llowed the spent fuel pool to
be divided into three regions.
Region I of the spent fuel pool remained
a
checkerboard
pattern,
region
2 would be
a three out of four pattern,
and
region
3 would be
a four out of four storage pattern.
The fuel assemblies
were stored
in the region
based
on fuel burnup, with the highest
burnup
assemblies
stored
in the four out of four region.
On November 9, the inspector
reviewed the
WO and observed
portions of the
reconfiguring of the spent fuel pool in Unit 1.
The inspector
noted
good
radiological
and foreign material controls
by the personnel
involved in the
evolution.
The inspector concluded that engineering
had established
appropriate
controls
to determine
the proper location of the fuel assemblies.
Additionally, the
inspector
noted that
a
100 percent verification of the proper configuration of
the spent fuel pool
was conducted after the movement of all the spent. fuel
assemblies,
0
-18-
7
EVALUATION OF ONLINE NAINTENANCE (TI 2515/126)
The inspector
reviewed the licensee's
process for scheduling online
maintenance.
The inspector
reviewed the licensee's
procedures,
interviewed
individuals from site scheduling,
the
group probable risk assessment,
and
operations.
The purpose of this inspection
was to review maintenance
scheduling activities with respect
to risk management
in accordance'ith
NRC
Inspection
Manual TI 2515/126.
7. 1
Schedulin
Process
Develo ment
The inspector
reviewed the scheduling
and planning processes
for online
maintenance
with members of the site scheduling organization.
The licensee
uses
a 12-week scheduling matrix for planning maintenance.
The licensee
chose
a 12-week vices
a 13-week schedule
so that maintenance
and testing activities
could
be performed
on the
same
day of the week while meeting monthly frequency
requirements.
The matrix coordinates
the scheduling of planned
maintenance,
and surveillance testing
by the week.
The licensee
developed
the matrix
'starting with surveillance testing requirements.
Systems that required
attendant
equipment to be operable
were grouped together.
The licensee
spaced
the testing
so that redundant train components
were not tested
during the
same
week.
The licensee
used
these
scheduled
test
windows to coincide with planned
outage
availabilities.
This way, the licensee
was able to use
a scheduled
routine
surveillance test to demonstrate
system operability following maintenance.
Finally, the licensee
reviewed all preventative
maintenance activities.
The
most frequent activity was
used to develop the online maintenance
window.
For
example,
each
emergency
diesel
generator
(fDG) is tested
on
a monthly basis
and the most frequent preventative
maintenance
is required every
6 months.
Therefore,
the licensee
has
an online maintenance
window every
6 months or
every other
12 week schedule.
Additionally, maintenance
required
on the spray
pond or other systems
required for fDG operability are scheduled
concurrently
with the
fDG.
7.2
Schedulin
Process
Risk Assessment
0
The
PRA group reviewed the scheduling matrix to ensure that the scheduled
activities did not introduce unacceptable
risks to the plant.
The
PRA group
reviewed the matrix for unacceptable
risk combinations.
This included
a
review to ensure that divergent
systems
necessary
for accident mitigation were
not scheduled
concurrently.
For example,
the matrix schedules
activities
on
the high pressure
SI, low pressure
SI,
and containment
spray
pumps
on separate
weeks since they all may be required for similar accident mitigation purposes.
7.3
~Schedulin
Using the
12-week scheduling matrix, the licensee lists routine work requests,
preventative
maintenance,
and
WOs to the assigned
outage
window.
Four weeks
prior to the work week,
the licensee
performs
an item-by-item review.
Two
i
-19-
weeks prior to the work week,
the licensee
conducts
another
review and
develops
a final approved
work list.
Finally, site scheduling
conducts daily
meetings with operations
and maintenance
to review the items scheduled
for the
next
24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
I
The impact'f emergent
work activities
on core
damage
frequency ultimately is
the responsibility of the operating
crew.
The licensee
procedure for hanging
a clearance
requires
the operating
crew check that redundant trains of
equipment
are available prior to hanging the clearance.
Additionally, if one
train of equipment is already out of service for maintenance
when
a component
fails, the operating
crew is responsible
to initiate the required
TS action
statement.
Finally, the inspector
has
observed
that the operating
crews
inform their management
of all priority one
and
two work requests
to ensure
that management
is involved in the decision
making process.
Priority one
and
two work requests
are those that require
immediate action to protect
individuals and/or equipment
and those that must
be worked within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />
due
to
a TS.
The
PRA group was involved in the scheduling
process
mainly in the development
and approval of the
12-week scheduling matrix.
However, if unusual
circumstances
occur, site scheduling
may request
that the
PRA group review the
weekly schedule.
For example,
Unit 2 has
a degraded
condition on all four
class
1E batteries
and
has committed to the
NRC to minimize activities
affecting the class electrical
system
(Section 3.2).
The
PRA group reviews
and approves
the scheduled
work activities which may impact the electrical
system.
The inspector
concluded that this approach
required site scheduling
to identify unusual
circumstances
and
had the potential to overlook less
obvious interactions.
7.4
Future Schedulin
Develo ments
0
The licensee
was in the process
of implementing
an integrated
scheduling
matrix that coordinated
the testing
and maintenance activities of all three
units.
This was
a result from the recently reengineered
maintenance
organization
where the licensee
combined three
separate
organizations
into
one.
The licensee
anticipates
two benefits to the integrated
schedule.
First, it should reduce
the manpower
burden
on the maintenance
organization
by
ensuring
maintenance
activities are not scheduled
on the
same
system in more
than
one unit.
And second, it should
reduce
the probability that the
same
type of human error-caused
failure could
be introduced in more than
one unit.
Finally, the licensee
was in the process of developing
a visually oriented
matrix for determining the risk associated
with different combinations of
system unavailabilities.
The licensee
performed
a detailed
PRA review of
potential
combinations
and categorized
each
one according to the associated
increase
in core
damage
frequency
(CDF).
The licensee
used the
guidelines
to develop the following coding
scheme:
green or acceptable
activities increase
the
CDF by less
than
10E-6, yellow or caution activities
increase
the
CDF by between
and
and red or dangerous
activities
increase
the
CDF by more than
The matrix also
shows which combinations
If
1
fi
I
1
0
-20-
would place the unit in
a
TS 3.0.3 action statement
and require
immediate
action to restore
systems
or shut
down the unit.
The licensee
intended to
implement this matrix by February
1995.
The inspector
reviewed
a draft of
this matrix and found it to be easy to understand
and apply to plant
configurations.
7,5
Conclusions
The inspector
concluded that the licensee
had
a process for scheduling online
maintenance
that appropriately
considered risk aspects
of those activities.
The licensee's
program involved the
use of a simple matrix system that
had
been thoroughly reviewed
by the risk assessment
group.
The inspector
concluded that the licensee
reduced
system
outage
times
by performing online
maintenance
on systems
concurrent with their required preventative
maintenance
and with maintenance
on their attendant
systems,
The inspector
noted that the
PRA group was not involved in maintenance
scheduling
on
a daily basis
unless
requested
and that this had the potential for missing risk significant
'ombinations.
However,
the inspector
concluded that the licensee's
planned
implementation of a visually-oriented risk matrix would lessen this potential.
8
PLANT SUPPORT
(71750)
8.1
Securit
Findin
s
On November
15,
1994,
the inspector
observed
a contract
employee
not wearing
an
ACAD inside the protected
area.
The inspector notified security,
who
subsequently
took custody of the person.
The inspector
noted that the contract
employee failed to properly wear the
ACAD, failed to notify security of the misplaced
ACAD, and did not
know of the
requirement
to notify security
when misplacing
an
ACAD,
The inspector determined that security performed adequately
when informed
about the lost ACAD,
Security determined that the worker had left the
ACAD in
a locker.
Although the worker had received
access
training
4 months prior to
this protected
area entry, this had
been the worker's first entry into the
protected
area.
The inspector
noted that the licensee
counseled
the
individual about
ACAD control.
The inspector
determined
the safety
significance of the event
was low.
8.2
Protected
Area Access Turnstile
After entering the protected
area,
the inspector
noted that the turnstile
continued to rotate
beyond the 120'F rotation necessary
for protected
area
access.
The inspector
continued to rotate the turnstile until it latched
after rotating approximately
240~F.
The inspector notified security
concerning
the turnstile.
The turnstiles
are full height access
devices that
allow one person to enter for every 120'F of rotation.
,jl
-21-
A security sergeant
explained that the turnstile latch was designed
to be
free'ong
enough for one person to enter
and then relatch
when the turnstile
reached 120'f rotation.
The sergeant
demonstrated
that if the latch
becomes
worn and the turnstile were rotated quickly enough,
that it can
be rotated
beyond 120'here it will not be aligned
and cannot latch.
However, if the
turnstile were left in this position for a few seconds, it would latch after
an additional 60'f rotation.
The inspector
noted that if an unauthorized
individual attempted
entry in this case,
the individual would be locked
between
the security bars.
Security tested
the turnstile several
times
and noted that while it usually
did latch after 120', it did on occasion
continue to 240'.
Security promptly
locked the turnstile in position
and initiated
a work request.
The inspector
concluded that the licensee
took appropriate
action.
9
FOLLOWUP OPERATIONS (92901)
9. 1
0 erabilit
Determination
Procedure
Im lementation
During
a routine tour of the plant the inspector
noted that the automatic
makeup valve for the essential
cooling,water expansion
tank was isolated.
The
inspector
noted that the automatic
makeup function was described
in the
Updated Final Safety Analysis Report
(UFSAR)
and that the valve was isolated
because it failed to stop flow.
The inspector discussed
the existing
condition with the system engineer
and noted that
a
WO and
a
CRDR had
been
written documenting
the condition
and that
a
10 CFR 50.59 safety evaluation
was performed to address
the fact that the automatic
makeup function was
disabled,
the inspector
concluded that the manual
actions
were acceptable
substitutes.
The inspector
conducted
a review of the licensee's
procedure
and noted that
a new appendix
was
added that outlined the required
reviews
when
a manual
action
was substituted for an automatic function.
The
inspector
reviewed the operability determination
log and- noted that the
actions outlined in Appendix
C of the operability determination
procedure
were
not performed for the essential
water surge tank automatic
makeup valve,
't
The inspector
asked
the licensee
why an operability determination
was not
performed.
The licensee
determined that Appendix
C was not part of the
procedure
when the condition was identified in Hay 1994.
At that time,
an
o'perability determination
was not required since existing documentation
in the
form of the
CRDR and
10 CFR 50.59 safety evaluation
already existed.
Appendix
C was
added to the operability determination
procedure
in July 1994.
However,
the body of the procedure
was not updated to describe
the purpose of
Appendix C,
and
when the actions
were required.
The inspector discussed
the apparent
oversight in the operability
determination
procedure with operations
management.
The inspector
reviewed
the manual
actions for essential
water surge tank makeup
and found them to be
consistent
with the manual
action criteria discussed
in Appendix
C of the
J'
'I
'f
i
'fir
-22-
licensee's
operability procedure.
The licensee
agreed to review the procedure
to determine if any improvements
were needed.
The licensee
also
agreed
to
review all the automatic functions that were bypassed
by manual
actions to
ensure that the intent of the Appendix
C actions
were being met.
The
inspector
concluded that management's
response
to this concern
was prudent.
9.2
Violation 528 9422-01
Closed
and
LER 528 94-05 Revision
1
Closed
Failure to Ad ust Core Protection Calculator
Delta-T Power
This violation involved the failure to declare
channel
"D" CPC inoperable
when
a
TS calibration check could not be adequately
performed
due to large swings
in calculated
thermal
power.
The licensee
determined that the root cause of the violation was
a failure of
management
to aggressively
pursue all aspects
of a known problem with
fluctuations in hot leg temperature
(T).
Although engineering
had identified
the root cause of the
T fluctuations,
management
did not pursue whether
operators
could conduct
a reasonable
channel calibration check between
thermal
power and secondary calorimetric power with the existing fluctuations in T.
The licensee
conducted
a review of the four channels of CPCs in all three
units to determine
whether they were operable with the
T fluctuations.
They
determined that calculated
thermal
power fluctuated
about 4-7 percent
on all
12 CPCs.
The licensee
conducted
an evaluation
which determined that the
fluctuations in thermal
power were caused
by the
known fluctuations in T.
Additionally, engineering
determined that the calculated
mean of the recorded
thermal
power was within H percent of the actual
power and that the values
recorded
by the operators
during tests
were very close to the mathematical
mean of the thermal
power.
Based
on this review, the licensee
determined that
all the
The inspector
agreed with the licensee's
conclusion.
The operators
had
been performing
an average of the
CPC digital display of
thermal
power
and recording that value during the channel calibration check.
The licensee
issued
a new procedure,
"Dampening/Averaging of Instrument
Fluctuations," to provide operators
with formal guidance
on
how to record
a
specific parameter, when the corresponding
instrument oscillates
excessively.
The procedure directs the operators
to record the displayed
thermal
power
every minute for a total of 10 minutest
These
values
are then
averaged
and
compared to the secondary calorimetric power to see if a channel
calibration
is required.
The inspector
noted that this procedure
should
he] p ensure
consistency
in the performance of the channel calibration check.
The licensee
used
a new analytical
model to verify that the fluctuations in T
were being caused
by the
known phenomena
of hot leg stratification.
During
this review, engineering
noted that
some utilities either average
the value of
T before
sending
the signal to the
CPC or have all the
T temperature
detectors
in the upper portion of the hot leg.
The licensee
was evaluating
moving the two lower T temperature
detectors
into the two spare wells in the
0-
-23-
upper part of the hot leg to minimize the fluctuations in T>.
The inspector
concluded that engineering
had conducted
a thorough review of the issue.
9.3
Closed
Violation 530 9420-01:
Im ro er Startin
of Steam-Driven
Auxiliar
Pum
This violation involved Unit 3 operators
that did not follow the procedure
when they attempted
to start the steam-driven
AFW pump with a cold steam
supply line.
As
a result,
condensation
formed when the supply steam contacted
the'0'F
steam piping and caused
the
pump to trip on mechanical
The licensee
determined that poor communication
between
the
RO and control
room supervisor
caused this event.
The licensee
recently
implemented
new
communications
and conduct of operations
standards
to minimize communication
errors.
The inspector
noted that communications
standards
had previously
existed,
but were not being consistently
used during routine evolutions.
The
inspector
noted that management
was emphasizing their expectations
for proper
communications
during all levels of control
room activities (Section 3).
The
inspector recently observed
good closed
loop communication during various
types of control
room evolutions indicating that these
expectations
were being
implemented.
The licensee
also determined that
a procedural
weakness
contributed to the
error.
Specifically, the surveillance test procedure
did not include
steam
line temperature limits in the precautions
and limitations section of the
procedure.
The licensee
updated
the surveillance test to clearly describe
the
steam line temperature
requirements.
An appendix of the procedure
was also
created
to describe
the locations for taking the temperature
of the steam
line.
The inspector
reviewed the procedures
and concluded that the procedure
clearly described
the precautions
and contingencies for low steam line
temperatures.
10
FOLLOWUP - MAINTENANCE (92902)
10. 1
Closed
Followu
Item 528 9355-02:
Pressurizer
S ra
Valve
Maintenance
This open item involved the failure to incorporate
engineering
information
concerning
a revised
bench set into a
WO to calibrate the pressurizer
spray
valve.
The licensee
had previously determined that the bench set information
in engineering
evaluation
requests
(EERs) from 1987
and
1988 had not been
updated
in the equipment
information data
base.
As
a result,
the appropriate
bench set
was not included in the
WO instructions.
The licensee
wrote
a
CROR in response
to the inspector's
concerns
to evaluate
the root cause of the problem
and to determine if there were
any programmatic
weaknesses
that contributed to the error.
The licensee
determined that the
EER procedure
was revised
in 1991 to require
a
CRDR to document
and track any
corrective actions resulting
from EERs.
Prior to 1991, actions resulting from
nonsafety-related
EERs were not formally documented
and tracked,
In the case
of the pressurizer
spray valve,
a
WO was written to update
the nameplate
data
but, was later canceled.
In 1991,
the licensee
conducted
a review of 20 percent of all the closed
design
change
and determined that only two out of 33 followup actions
were not tracked.
The licensee
canceled
the
EER program in September
1994
and
all engineering
information has since
been
processed
using the
CRDR program.
The licensee
concluded that based
on the previous review of EERs,
an
additional
review of all
EERs prior to 1991
was not required.
The licensee
also concluded that the current process
was adequate
to trend, track,
and
transfer design information, lessons
learned,
and followup actions into plant
maintenance
and practice.
The inspector
reviewed the licensee's
evaluation
and
agreed with the licensee's
conclusions.
11
FOLLOWUP ENGINEERING/TECHNICAL SUPPORT
(92903)
11. 1
Followu
Item 529 9402-02
Closed
Pum
Seal
Leaka
e - Unit 2
'This item involved the
pump seal
leakage limits.
The inspector
had noted
that the
UFSAR had stated that the
maximum allowed seal
leakage
was
100 cc/hr
and that the licensee
did not have
a surveillance test which limited the seal
leakage
to this value.
The inspector
had noted that the licensee's
surveillance test for the emergency
core cooling system
(ECCS) limited the
entire
system to less
than
1 gallon per minute,
but that there
were
no
specific limits for pump seal
leakage.
The inspector
had questioned
the
licensee
on the basis for the limit and whether the licensee
adequately
measured
this limit.
The inspector
reviewed the licensee's
response
to this issue.
The licensee
stated that the
100 cc/hr seal
leakage limit was
a design input assumption
used for calculating
dose
and
was
an input to the radioactive
waste drainage
system.
The
UFSAR states
that the radioactive
waste drainage
system
was
designed
to handle
up to 50 gallons per minute
and could easily handle the
ECCS leakage of
1 gallon per minute.
The licensee
had contacted
CE for the
basis for the limits.
CE stated that the limit was simply used for design
input used during evaluation for habitability and dose calculations.
During the review of the questions
raised
by the followup item, the licensee
identified that its dose
assessment,
Calculation
13-NC-ZY-209,
assumed
1500
cc/hr leakage
from the
ECCS,
however,
the licensee's
TS Surveillance
requirements
limited the total
ECCS leakage
at
1 gallon per minute.
The
licensee
reviewed the
UFSAR dose
assessment
and determined that it included
enough margin to increase
the
ECCS leakage
from 1500 cc/hr to
1 gallon per
minute
and not exceed
the
10 CFR Part 100 limits for offsite dose.
The
.
inspector estimated
the increase
in offsite dose
was caused
by raising the
ECCS allowable leakage
from 1500 cc/hr to
1 gal/min
and concurred that the
dose did not exceed
Part
100 limits.
The licensee
reviewed the past
15 performances
of the
ECCS leakage testing
and
noted that the total
leakage
was greater
than
1500 cc/hr on five occasions,
l
i
-25-
but always
remained
below
1 gallon per minute.
The licensee
conservatively
decided to revise the surveillance test
acceptance
criteria to z 1500 cc/hr
for the
ECCS.
The inspector
reviewed the surveillance test procedures
for
each unit and confirmed that the acceptance
criteria had
been revised to
~ 1500 cc/hr.
The inspector
concluded that the licensee
met
limits and the surveillance test adequately
measured
the
ECCS leakage.
11.2
Followu
Item 528 9348-04
CLOSED
0
This open item involved an evaluation of the apparent differences
between
setpoint testing using the Furmanite "Trevitest" and the Westinghouse
live
steam test
methods
to determine if these differences
could affect the
operability of MSSVs.
The licensee
determined
there
was
an apparent offset
between
the two methods
during
MSSV testing in September
1993.
In March
1994,
Furmanite
informed the
NRC that the valve disk seat
area for the Dresser
Model
3707R safety valve used
in the equation to calculate
the lift setpoint
should
be changed
to match the actual
setpoint during full pressure lift
tests.
This change
was
based
on comparison testing
conducted
at the
Westinghouse test facility between
the two approved test methods.
On August 11,
1994,
the
NRC issued
"Inaccuracy of Safety Valve Set
Pressure
Determinations
Using Assist Devices."
The
IN described
how the
change
in valve disc area
would cause
the calculated lift setpoint to be about
one percent
lower.
In response
to the IN, several utilities had to declare
because
the revised lift setpoint
using the
new seat
area
was
outside
the
band required
by the plant TS.
Operability of the
MSSVs at Palo
Verde was not affected
because
the licensee
had already factored in the offset
into their equation for determining the lift setpoint.
Additionally, the
licensee
had
used
the live steam tests
at the Westinghouse facility to
establish operability of the
The inspector
noted that the licensee
had
aggressively
pursued
the resol,ution of these testing differences.
12
ON-SITE REVIEW OF LERs
12, 1
LER 529 91-005
CLOSED
HSSV Set oint Out-of-Tolerance
This
LER involved
a significant number of MSSVs having setpoints
during
testing outside the al percent
band required
by the plant TS.
On Hay 16,
1994,
the
NRC issued
a
TS amendment
to increase
the
MSSV setpoint tolerance
from +I percent to a3 percent.
This
LER is closed
based
on the change
in
allowable setpoint tolerance.
13
IN-.OFFICE REVIEW OF LERs
(90712)
The following LERs were closed
based
on
an in-office review:
~
LER 529/94-02:
Reactor Trip and Containment
Spray
~
LER 530/94-05:
Reactor Trip Caused
by Degradation of Feedwater
Flow
Q,y
l
,~
I
~
-26-
LER 530/94-06:
Shutdown Cooling Loop Inoperable
Due to
Pump Breaker
Racked in the Test Position
LER 530/94-07:
Reactor Trip Caused
by an Increase
in Hain Feedwater
Flow
I
4
ATTACHMENT 1
1
Persons
Contacted
1. 1
Arizona Public Service
Com an
J.
- D
- W.
- A.
- D
- J
- D
- W.
- 'G
F,
- G
D.
- R.
J,
Bailey, Vice President,
Nuclear Engineering
and Projects
Garchow, Director, Engineering
Grabo,
Section
Leader Compliance,
Nuclear Regulatory Affairs
Ide, Director, Operations
Krainik, Department
Leader,
Nuclear Regulatory Affairs
Larkin, Senior Engineer,
Nuclear Regulatory Affairs
Levine, Vice President,
Nuclear Production
Mauldin, Director, Site Maintenance
and Modifications
Montefour, Senior Representative,
Management
Services
Overbeck, Assistant to the Vice President
Nuclear Production
Riedel,
Department
Leader,
Unit 2 Operations
Seaman,
Director, Nuclear Assurance
Shanker,
Department
Leader,
Nuclear Assurance
Engineering
Smith,
Department
Leader,
Unit
1 Operations
Stroud,
Consultant,
Nuclear Regulatory Affairs
Taylor, Department
Leader,
Unit 3 Operations
1.2
NRC Personnel
- K. Johnston,
Senior Resident
Inspector
H. Freeman,
Resident
Inspector
- J.
Kramer, Resident
Inspector
- A. MacDougall,
Resident
Inspector
1.3
Others
- J.
Draper, Site Representative,
Southern California Edison
- R. Henry, Site Representative,
Salt River Project
- F.
Gowers, Site Representative,
El
Paso Electric
- denotes
personnel
listed above
who attended
the Exit meeting held on
November 30,
1994.
2
EXIT MEETING
An exit meeting
was conducted
on November 30,
1994.
During this meeting,
the
inspector
summarized
the scope
and findings of the report.
The licensee
acknowledged
the inspection findings documented
in this report.
The licensee
did not identify as proprietary
any information provided to, or reviewed by,
the inspectors.
i
I
I
ATTACHMENT 2
CRDR
DP
IN
IV
LER
MSS
RDT
TH
TS
VTM
automatic control
access
device
auxiliary operator
response
procedure
core
damage
frequency
Combustion
Engineering
control element
assembly
core protection calculator
Condition Report/Disposition
Request
differential pressure
energy control center
emergency
core cooling system
emergency 'diesel
generator
engineering
evaluation
request
emergency
operations facility
Electric Power Research
Institute
high pressure
safety injection
Information Notice
independent verification
Licensee
Event Report
motor operated
valve
mode select
switch
pressure
instrument
reactor drain tank
reactor operator
reactor regulating
system
remote
shutdown
panel
safety injection
safety injection tank
Salt. River Project
Shift Supervisor
hot leg temperature
Technical Specifications
technical
support center
Updated
Final Safety Analysis Report
vendor technical
manual
work order
1
i'
I
l
~