ML16342D435
| ML16342D435 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 09/12/1996 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML16342D434 | List: |
| References | |
| 50-275-96-16, 50-323-96-16, NUDOCS 9609230043 | |
| Download: ML16342D435 (96) | |
See also: IR 05000275/1996016
Text
ENCLOSURE 2
U.S.
NUCLEAR REGULATORY COMMISSION
REGION IV
Docket Nos.:
License Nos.:
Report No.:
Licensee:
Facility:
Location:
Dates:'nspectors:
50-275,
50-323
50-275/96016.
50-323/96016
Pacific Gas
and Electri'c Company
Diablo Canyon Nuclear
Power Plant.
Units
1 and
2
7 1/2 miles
NW of Avila Beach
Avila Beach, California
July 7 through August 17,
1996
M. Tschiltz, Senior Resident
Inspector
S.
Boynton. Resident
Inspector
C. Paulk, Division of Reactor Safety,
Region
IV
B. Olson, Project Inspector,
Region
IV
P. Goldberg, Division of Reactor Safety,
Region
IV
D. Corporandy,
Project Inspector,
Region
IV
Approved By:
R. Huey, Acting Chief, Branch
E
Division of Reactor Projects
ATTACHMENTS:
ATTACHMENT 1:
ATTACHMENT 2:
ATTACHMENT 3:
ATTACHMENT 4:
Partial List of Persons
Contacted
List oi Inspection
Procedures
Used
List of Items Opened.
Closed,
and Discussed
List of Acronyms
List of Attendees
at the July 1,
1996.
Predecisional
Enforcement
Conference
Apparent Violations as Originally Proposed for the
Predecisional
Enforcement
Conference
Licensee Presentation
Slides for the Predecisional
Enforcement
Conference
960'7230043
960'Vi2
ADOCK 05000275
8
EXECUTIVE SUMMARY
Diablo Canyon Nuclear
Power Plant, Units
1 and
2
NRC Inspection Report 50-275/96016;
50-323/96016
~0erations
~
Following the loss of instrumentation
on the primary and secondary
meteorological
towers,
the shift supervisor
and regulatory services
staff failed to recognize the requirement to report the event in
accordance
with plant procedures
and
A noncited violation
was identified (Section 01.2).
~
Both Units
1 and
2 operators
were challenged
by equipment
problems
following the dual unit trip caused
by a power grid disturbance.
Operator
response
to reduce
steam pressure
below the main steam safety
valve
(MSSV) lift pressure
on both units
and restore Unit 2 reactor
.coolant
pump
(RCP) seal injection flow was prompt and effective,
and
prevented additional complications
(Section 01.3).
~
Both the Unit 1 Control Operator
and Senior Control Operator
(SCO) were
observed escorting visitors while on watch.
The practice of allowing
the operator at the controls
and the
SCO to be assigned
the duties of
visitor escort creates
the potential to distract the operators
attention
from assigned
duties
and is considered to have
been
a weakness
in shift
management
oversight of'perator activities (Section 08.1).
~
Operators failed to properly position containment
fan cooler unit (CFCU)
2-3 and 2-4 speed selector
switches following surveillance testing which
resulted in the associated
high speed
breakers tripping following a bus
transf'er to startup
power due to a power
system grid disturbance.
A
noncited violation was identified (Section 01.3).
Maintenance'he
lift pressures
of three of'he five HSSVs associated
with Steam
Gener ator
(SG) 1-1 were improperly set .when personnel
performing. the
maintenance
inadvertently uti ized the valve-specific correction factors
for the
MSSVs associated
with SG 2-1 (Section E1.2).
Maintenance
personnel
did not adequately self-verify and review work
activities which resulted in one train of reactor trip permissive
P-4
being inoperable for a period of over 2 months.
A noncited violation
was identified (Section Hl.2).
Improper restoration
from maintenance
resulted in the failure to
maintain the TS required
number of Unit 1 reactor coolant leakage
detection
systems
in service.
A noncited violation was identified
(Section M8.1).
Licensee
management
was adequately tracking the backlog of corrective
maintenance:
however.
there were no organizational
goals set for
0
-2-
reducing the number of existing corrective maintenance
items
(Section, M6.1) .
~E
~
Engineers failed to perform
a prompt operability assessment
(POA) for
a
period of ten days after discovering
an emergency diesel
generator
(EDG)
voltage regulator with scorched resistor
and circuit board
and circuitry
different from that specified in the design
document.
A violation was
identified (Section E1.1).
Summar
of Plant Status
Re ort Details
Unit 1 began the period at
100 percent
power.
On July 12,
power
was reduced
to 50 percent to facilitate cleaning of the circulating water
Pump 1-2 tunnel.
The unit was returned to full power on July 15.
On July 19, Unit
1 was
curtailed to 48 percent
power to make repai rs on the main feedwater
Pump 1-1
control oil system.
Repairs
were completed
and the'nit returned to
100 percent
power on July 20.
On August 10, Unit 1 tripped on
12 kV bus
under voltage due to a significant grid disturbance
on, the Pacific Inter-tie.
The unit returned to Mode
1 on August
16 and reached
100 percent
on August 17.
Unit 2 began the period at 100 percent
power.
On August 1, the unit entered
Mode 2 and secured
the main turbine due to a nonisolable
leak on the turbine
electrohydraulic control oil system.
The unit returned to full power on
August 3.
On August 10, Unit 2 tripped due to loss of all four RCPs.
The
breakers
opened
due to an undervoltage condition on the
12 kV auxiliary and
startup
busses
caused
by
a significant grid disturbance
on the Pacific
Inter-tie.
The unit returned to Mode
1 on August
15 and was at 90 percent
power at the end of the inspection period.
I. 0 erations
Ol
Conduct of Operations
Ol. 1
General
Comments
71707
Using Inspection
Procedure
71707. the inspectors
conducted
frequent
reviews of ongoing plant operations.
In general,
the conduct of
operations
was professional
and safety-conscious.
01.2'oss
of Primar
. and Secondar
Meteorolo ical Instrumentation
a.
Ins ecti on Sco
e
71707
On July 27, the secondary
meteorological
tower instrumentation
was
damaged
as
a result of'
lightning strike to the tower.
The primary
meteorological
tower was concurrently out of service for extensive
maintenance.
The inspector
reviewed the licensee's
actions in response
to this event
and Procedure XIl.I02, Revision 3A, "Regulatory Reporting
Requirements
and Reporting Process."
b.
Observations
and Findin s
In response to the loss of the secondary
meteorological
instrumentation.
the licensee
established
a communication data link with the National
Weather Service Point Bouchon Station
and
made available
an onsite
portable meteorological station for local indications of weather
conditions.
Subsequently,
the licensee set
up several
additional
meteorological
instruments that could be polled from the control
room.
The secondary
meteorological
tower was returned to service
on July 30.
0
Procedure
XI1. ID2 provides specific guidelines for determining the
reporting requirement of various events.
In accordance
with
Attachment 8.6 of Procedure
XI1.ID2, a total loss of primary and
secondary
meteorological
tower indications requires
a 1-hour
nonemergency
report to the
NRC as directed
On
July 28, the inspectors
discus'sed
the loss of the meteorological
instrumentation with the shift supervisor
(SS)
and questioned
the need
for reporting the event to the
NRC.
The
SS determined that the event
was reportable in accordance
with Procedure
XI1. ID2.
A 1-hour
nonemergency
report was subsequently
made later that evening.
The
primary and secondary
meteorological
indications
had been unavailable
for 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br />
when the condition was reported.
Conclusions
The failure of the licensee to notify the
NRC within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> of the loss
of the secondary
meteorological
instrumentation
was not in accordance
with the licensee's
own regulatory reporting procedural
requirements
and
was also
a violation of 10 CFR 50.72.
However. since the licensee
promptly initiated actions to re-establish
meteorological
assessment
capabilities following the loss of the secondary
meteorological
tower
this violation is considered to be minor and is being treated
as
a
noncited violation. consistent with Section VII.B.1 of the
NRC
(NCV 50-275/96016-01;
50-323/96016-01).
Res
onse to the Dual Unit Tri
Followin
a
Ma or Grid Disturbance
Ins ection Sco
e
71707
Inspectors
responded to the site following notification of the dual unit
trip and observed control
room indications,
discussed
the event with
operators
and reviewed alarm summaries
and plant process
computer
recorded
parameter
data.
Observations
and Findin s
On August 10; at approximately 3:49 p.m.,
PDT,
a major grid disturbance,
which included the loss of the Pacific inter-tie,
caused
both Diablo
Canyon Units
1 and
2 reactors to trip.
Unit 1 tripped as
a result of
12 kV bus undervoltage,
while Unit 2 tripped due to the loss of RCPs
on
undervoltage trips.
Unit
1
remained in operation following the
transfer to startup
power.
Startup
power (230 kV) remained available following the trip and
provided offsite power to the site until restoration of the auxiliary
power
(500 kV) system.
After 500
kV power was restored'perators
realigned Units
1 and
2 electrical
systems to backfeed offsite power
from the 500
kV system.
Following the realignment.
Unit
1 busses
autotransferred
back to 230 kV power due to a ground in the protective
n
0
relay circuitry.
Auxiliary power to Unit
1 was declared
until the problem was isolated
and corrected.
Following the grid disturbance
the licensee
reported that
100 of 135
offsite early warning sirens
were without power.
Within several
hours
the licensee
restored
power to all sirens within the
10 mile emergency
planning zone.
Following the trip of the units, several
anomalies
were noted:
Unit 1:
MSSV RV-7 lifted following the trip at approximately
50 to
60 psig below its lift setpoint.
Operators
responded
by lowering steam
pressure to approximately
1000 psig using the atmospheric
dump valves at
which point the
MSSVs reseated
(details of MSSV testing are provided in
Section El.2).
Unit 2:
Operators
shut the main steam isolation valves to limit
cooldown.
MSSVs RV-4 and
RV-8 lifted approximately 50-60 psig below
thei r set lift pressure.
Operators
responded
by lowering steam pressure
to 1000 psig at which point the valves reseated.
RCP seal injection
flow decreased
to approximately
2 gpm per
pump apparently
due to
blockage of seal injection needle valves.
Operators
responded
by fully
opening
and then rethrottling the needle valves which restored
normal
seal injection flow.
CFCUs 2-3 and 2-4 breakers
tripped on overcurrent.
CFCUs 2-3 and 2-4 were, later determined to have
been selected to high
speed
and tripped as
a result of attempting to start directly in high
speed following the bus transfer.
Operators
had incorrectly left the
switches selected to high speed following the completion of surveillance
testing.
The licensee
concluded that the surveillance
procedure
was
adequate,
although it did not provide specific restoration
requirements.
Operating
Procedure
(OP) H-2: I, Revision 13,
"Containment
Fan Cooler
'nits
- Place in Service," the governing procedure for operation of
CFCUs, Section
6. 11, contains specific requirements
to reset
CFCU
selector
switches to slow speed after starting
a
CFCU in high speed.
It
was later verified by review of the associated
control schematics
that
CFCUs 2-3 and 2-4 would have started
and run in slow speed
as required
upon receipt of an engineered
safety feature
(ESF) start signal with the
selector
switches selected to the high speed position.
The fai lure of
the operators to properly restore the selector switches for CFCU 2-3 and
2-4 to the position required by OP H-2:I is
a violation of TS 6.8. 1.,
which requires that procedures
be established'mplemented
and
maintained covering the startup,
operation
and shutdown of the
containment cooling system.
Since this violation was self-revealing, of
minor safety consequence
and corrective actions to prevent recurrence
are being implemented
by the licensee, it is being treated
as
a noncited
violation, consistent with Section VII,B.1 of the
(NCV 50-323/96016-02).
-4-
08
08.1
b.
Conclusions
Operator
response
to the dual unit trip was good.
Both units were
quickly stabilized in Hode 3 despite
several
equipment
problems that
occurred which complicated operator
response
following the trips.
Improper restoration
from CFCU surveillance testing caused
CFCU 2-3 and
2-4 high speed
breakers to trip when the
CFCUs attempted to start
directly in high speed following an automatic
bus transfer.
Hiscellaneous
Operations
Issues
Control
Room Observations
Durin
Surveillance Testin
Ins ection Sco
e
71707
The inspector s observed
the normal conduct of routine evolutions in the
control
room during the performance of Rod Drop Surveillance testing.
Observations
and Findin s
During control
room observations
the inspector
noted that both the
CO
and the
SCO were escorting visitors.
The appropriateness
was questioned
and the Operations
Hanager
determined that it was inappropriate for
control
room operators to be assigned
the additional duty of escorting
visitors and directed that the visitor escort duties
be assigned to
another individual.
The guidance of OP1.DC10.
"General Authorities and
Responsibilities
of Operating Shift Personnel,"
specified that the
operator at the controls
must give full attention to the condition of
the plant at all times.
The guidance did not specifically address
the
performance of escort duties while on shift.
This procedure
incorporated the guidance of'egulatory Guide 1. 114,
"Guidance to
Operators
at the Controls
and Senior
Operators
in the Control
Room of a
Nuclear
Power Unit."
Conclusions
The practice of allowing the operator at the controls
and the
SCO to be
assigned
the additional duty of escorting visitors has the potential to
distract the operator
from assigned
duties
and was considered
a weakness
in shift management
oversight of operator acti vities.
Conduct of Maintenance
Maintenance
Observations
Ins ection Sco
e
62707
II. Maintenance
The inspectors
observed all or portions of the following work
activities:
~
CO 145794:
Replacement
of EDG 1-3 exhaust
bellows locknuts
Observations
and Findin s
The inspectors
found the work performed
under these activities to be
professional
and thorough.
All work observed
was performed with the
work package
present
and in active use.
Technicians
were experienced - .
and knowledgeable of their assigned
tasks.
The inspectors
observed
system engineers
monitoring job progress
and that quality control
personnel
were present
when required
by the procedure.
When'pplicable,
appropriate radiation control measures
were in place.
In addition, selected
maintenance
observations
are discussed
below.
Im ro er Material Utilized for Auxiliar Salt Water
ASW
Pum
Shaft
Ke
Ins ection Sco
e
62707
On August 3, during refurbishment of the Unit 2 ASW pump that had been
replaced
during 2R7, the licensee
discovered
one of the
pump shaft
was missing one of its two torsion keys.
The inspector
reviewed the licensee's
actions taken to determine the root cause of the
missing key and the licensee's
evaluation of the condition of the
installed
ASW pumps.
Observations
and Findin s
fach of the
ASW pumps are vertically mounted wet pit pumps with the
impeller coupled to the
pump motor through
a three-piece
extended shaft
arrangement.
The three
segments
of the
pump shaft are coupled by means
of a thrust ring and two torsion keys fitted into a coupling sleeve.
As
these coupling arrangements
are normally exposed to a saltwater
environment,
a corrosion resistant
stainless
steel alloy, trade
name
Hastelloy, is used in their fabrication.
The licensee
evaluated
the as-found condition of the
pump shaft
and
determined that
a key had been installed in the coupling;
howevers
the
key installed
was most likely fabricated
from carbon steel.
During the
time the
pump was installed in the system,
the key had dissolved in the
0
0
-6-
corrosive saltwater environment.
As
a consequence,
the
pump shaft was
allowed to rotate inside the coupling sleeve before galling between the
two components
caused
them to fuse together.
The licensee
had not
completed its evaluation of the impact of the as-found condition on past
operabi lity of the
pump at the end of this inspection period.
In evaluating'the
root cause of installing
a key of the wrong material.
the licensee
looked at (1) the key configuration of the entire
ASW
pump-motor
assembly,
(2) previous work packages
associated
with the
refurbishment of each of the
ASW pumps,
including procurement of any
replacement
keys,
(3) work practices
during pump refurbishment,
including material accountability,
and (4) warehouse
stocking of the
different keys.
The licensee
did not identify any process
inadequacies
in 'these
areas
and concluded that this was
an isolated event
and it was unlikely that
an improper key was
used in any of the currently installed
ASW pumps.
The licensee
supported their conclusion
by showing that the key
configuration of the
ASW pump shaft does not allow for interchanging the
various keys.
Thus, the improper key had been installed
as
a
replacement.
A survey of the warehouse
found that the carbon steel
keys
and Hastelloy keys are in separate
locations.
Additionally,
a magnet
test found that there were no carbon steel
keys inadvertently stored in
the bin for the nonmagnetic Hastelloy key.
The licensee
found that
three carbon steel
keys
had been pulled from stock for refurbishment of
the
ASW pumps
and was confident that the keys
had been installed in the
motor-to-shaft couplings.
Conclusions
The licensee
took reasonable
steps to determine whether or not
a
degraded
condition exists in the currently installed
ASW pumps.
However,
a definitive conclusion
on the current configuration could not
be made without visual observation of the
pump shaft keys.
Therefore,
the licensee's
past operability evaluation
and long-term corrective
actions will be examined
as
an inspection followup item
(IFI 50-275/96016-03:50-323/96016-03).
Ino erable Reactor Tri
Permissive
P-4
Ins ection
Sco
e
92902
The inspectors
reviewed the work order and maintenance
procedure
utilized for the replacement of solenoid valve (SV)-171. the applicable
TS and operator
emergency
procedures.
Observations
and Findin s
On August 2.
a maintenance
technician
found
a lifted lead associated
with Train 8 of Unit 2 reactor trip Permissive
P-4.
The lifted lead
0
K
would have precluded
a Train 8 turbine trip upon opening of the Train 8
reactor trip breakers.
The licensee
determined that the lead
had not
been landed after restoration
from maintenance activities on May 22,
1996.
As
a result. the
TS requiring two operable trains of the reactor
trip permissive
had not been
met when Unit 2 entered
Hode
1 on May 24
and during subsequent
plant operation.
The licensee
reviewed the work order for the work performed
'on Hay 22
and identified that Step 8.6.26.f. 'of Procedure
HP I-1.36-1,
"Main
Turbine Control Integrated
Functional Test," to land the lifted leads
had not been initialed as completed.
The licensee's
root cause
determination
indicated that inadequate self-verification during the
activity and inadequate
postwork review contributed to the event.
Proposed corrective actions
included revising the maintenance
procedure
to enter
the
TS action statement
when the lead is lifted. training
maintenance
personnel
on the event,
and counseling
those involved.
Train A of reactor trip permissive
P-4 had been inhibited for
surveillance testing while Train 8 was inoperable.
The longest period
when both trains were inoperable
was approximately
50 minutes,
occurring
on June
6; however, diverse methods of tripping the turbine were
available while reactor trip permissive
The methods
included operator initiated trips and automatic trips as
a result of a
safety injection signal, electrical faults, or overspeed.
The licensee
performed simulator scenarios
and confirmed that the turbine would trip
approximately
15 seconds
following a reactor trip if P-4 was disabled
and no operator actions
were performed.
The inspector confirmed that
licensee
procedures
di rect operator s to verify or initiate a turbine
trip following a reactor trip.
The inspector
noted that
an October 27,
1987.
evaluation
concluded that failure of the turbine to
trip 'following a reactor trip was not likely, gi ven the diverse
means of
tripping the turbine.
Conclusions
The safety impact of this event
was not significant.
The licensee
identified and corrected the violation and. therefore, this occurrence
is being treated
as
a noncited violation, consistent with
Section VII.B.1 of the
(NCV 50-323/96016-04).
Surveillance Observations
Ins ection Sco
e
61726
Selected
surveillance tests
required to be performed
by the
TS were
reviewed
on
a sampling basis to verify that:
(1) the surveillance tests
were correctly included on the facility schedule.
(2)
a technically
adequate
procedure existed for the performance of the surveillance
tests,
(3) the surveillance tests
had been performed at
a frequency
l
0
0
-8-
M6
N6.1
specified in the TS,
and (4) test results satisfied
acceptance criteria
or were properly dispositioned.
The inspectors
observed all or portions of the following surveillance:
~
STP R-2B1. Revision 6.
"PPC Operator
Heat Balance," Unit 2
~
STP R-1B, Revision 16,
"Rod Drop Measurement,"
Unit I
~
STP I-38-B.l. Revision 2,
"SSPS Train
B Actuation Logic test in
Modes l. 2, 3, or 4," Unit 1
Observations
and Findin s
The inspectors
found that the survei llances
reviewed and/or observed
'were being scheduled
and performed at the requi red frequency.
The
procedures
governing the surveillance tests
were technically adequate
and personnel
performing the surveillance
demonstrated
an adequate
level
of knowledge.
The inspectors
also noted that test results
were
appropriately dispositioned.
Maintenance Organization
and Administration
Mana ement of Maintenance
Backlo
Ins ection Sco
e
62707
The inspectors
reviewed the licensee's
maintenance
backlog reports,
nonoutage priority 4 action requests
(ARs) and Interdepartmental
Administrative Procedure
OM7. ID1. Revision 6,
"Problem Identification
and Resolution."
Observations
and Findin s
The maintenance
backlog reports indicated
an overall decreasing
number
of maintenance
backlog items at Diablo Canyon during the past year.
Although there
was no formal procedure for reporting or monitoring
backlog,
maintenance
managers
routinely received reports that indicated
the backlog of corrective maintenance
in their respective
areas.
The
reports
were focused primari ly on the higher priority corrective
maintenance
items;
however,
lower priority (priority 4) issues
were also
being separately
tracked.
The work loads at the work. planner
and
maintenance
group assigned to perform the work appeared
to be the
largest contributors to the backlog.
The licensee
was evaluating what
actions were necessary
to reduce the backlog at the work planning stage.
These
items were noted to be in large part lower priority work items.
Action requests
are prioritized by assigning
a priority level of 1
through 4,
4 being the lowest priority maintenance
issue.
Until
recently,
the
AR initiator determined its priority.
The licensee
has
0
I
I
<I
-9-
M8
M8.1
changed the process
by establishing
a review panel that meets
on
a daily
basis,
in part, to ensure that
ARs are assigned
the correct priority.
The licensee also groups
ARs into nonoutage
and outage categories.
The
focus of the licensee
maintenance
organization
was to reduce the
nonoutage
maintenance
backlog.
As
a result, the total backlog
number
was not readily available but was later determined to be approximately
1400 items.
Several initiatives had recently been started in an effort
to reduce the backlog.
The most significant of these efforts was the
establishment
of a "fixit now" team to perf'orm minor maintenance.
Control
room deficiencies
were not being tracked separately.
Although
the licensee
had previously established
a program for tracking control
room deficiencies,
the decision
was
made that the deficiencies
had been
reduced to the number that was easily manageable.
The licensee is
determining the number of outstanding control
room deficiencies
and is
r'eevaluating the need to track control
room deficiencies
separately
from
other corrective maintenance
items.
Discussions with the Maintenance
Hanager indicated that there were no
organizational
goals for the reduction of the backlog.
In addition,
a
number of priority 4 ARs had been written greater
than
1 year ago,
and
several that had been written close to 3 years
ago without having been
corrected.
Concerns
were raised during the inspection regarding the
potential
worsening of equipment conditions that when first noted were
categorized
as priority 4.
The licensee
had not conducted
reviews of
priority 4 ARs that had been in existence for long periods of time
(e.g.
minor oil leaks,
seal
leaks
and valve packing leaks) to determine
if degradation
from the initially reported condition had occurred which
in turn warranted raising the priority of the maintenance
item.
Conclusion
The licensee
appeared to be effectively tracking and managing the
backlog of corrective maintenance.
Several
areas of potential
improvement were noted including:
establishing
goals for the reduction
of the backlog, periodic reviews
and reassessment
of priorties of
equipment deficiencies that have existed for a period of time without
being corrected,
and the separate
tracking of control
room deficiencies.
Miscellaneous
Maintenance
Issues
(92902)
Closed
Licensee
Event
Re ort
LER
50-275/95-005-00.:
TS 3.4.6. 1 not
met with the reactor coolant leakage detection
systems
due to
personnel
error.
The
LER was issued
due to the TS requirements
for
containment
atmosphere
radiation monitoring systems
not being met.
TS
Re uirements f'r Containment
Atmos here Radiation Monitor in
S stems
TS 3.4.6. 1 requires that three different reactor coolant system
(RCS)
leakage detection
systems
be operable in Modes
1, 2,
3 and 4. If only
two systems
are operable.
operation
can continue for up to 30 days
-10-
provided grab samples of containment
atmosphere
are obtained
and
analyzed
once per 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
The
TS requires if less than two of the
leakage detection
systems
are operable that the unit be in HOT STANDBY
within the next
6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />,
and in COLD SHUTDOWN within the following 30
hours.
The licensee failed to meet
TS 3.4.6.1 in that only one of the
three
RCS leakage detection
systems
was operable for an estimated
time
period of approximately
1 day.
Upon discovery operators
placed the
containment
fan cooler collection monitoring system in service
and
entered the .TS action statement.
Licensee
Investi ative and Corrective Actions:
The licensee's
investigation of the incident revealed that
on June 22,
1995,
maintenance
was performed which removed the containment
atmospheric
particulate monitoring system
(RH-11) and the containment
atmosphere
gaseous
radioactivity monitoring system
(RH-12) from service.
During the
maintenance
on RH-11 and RH-12,
requirements
were satisfied
by having two reactor
coolant leakage detection
systems
available
and by
entering the action statement,
which required the analysis of grab
samples
every 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
The licensee
determined that after completion of the maintenance,
contrary to procedure
guidance,
the technicians incorrectly left the
RH-ll and
RH-12 sample selector switch in the purge position which
prevented
the sampling of containment
atmosphere.
Both RH-11 and
RH-12
use
a
common sample path.
The improper alignment
was noted
by control
room operators
during routine shift checks.
After discovering the
problem operators
placed the
CFCU collection monitoring system into
service
and entered the
TS 3.4.6.1 action statement.
The licensee
noted
in the
LER that during the time period when only one reactor coolant
leakage detection
system
was in service
a grab sample of containment
atmosphere
was taken
and,RCS
1'eak rate calculations
were being performed
as required.
Corrective actions initiated in response to this event included:
revision of the surveillance
procedure to verify that the sample path
alignment is correct prior to placing
RH-11 and
RH-12 back into service,
review of other procedures
for similar concerns
and training of
operators
and
18C technicians
on the importance of self-verification.
These actions
appeared to address
the problem appropriately.
Conclusion
The fai lure to maintain the required
number of reactor
coolant leakage
detection
systems
in service is
a violation of TS 3.4.6. 1.
Based
upon
the details noted above, this failure constitutes
a violation of minor
safety signif'icance
and is being treated
as
a noncited violation
consistent with Section
IV of the
NRC
ENFORCEHENT POLICY
(NCV 50-275/96016-05)
-11-
M8.2
k
El
El. 1
Closed
LER 50-275/95-008-00
01
and 02:
control
room ventilation
system
(CRVS) outside design basis
due to a programmatic deficiency in
the operation
and maintenance of the HEPA filter and charcoal
absorber
system.
On August 11,
1995, the licensee
determined that the
CRVS had
been outside its design basis
on four occasions
when charcoal
absorbers
were replaced,
and isolation was not provided to separate
the
ventilation trains.
The licensee
determined that
a design basis
accident would have introduced unfiltered air into the control
room.
and
initial calculations
showed that postaccident
exposure limits would have
been
exceeded for control
room operators.
The licensee
subsequently
performed detailed calculations
which showed
that exposure limits would not be exceeded if the open ventilation train
was isolated within 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> after
a design basis accident.
The licensee
concluded that operator action after receiving control
room radiation
alarms would result in train isolation within approximately
1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> and
20 minutes after-the event.
The licensee ultimately concluded that
a
condition outside of the design basis
had not existed.
Additionally.,TS
had not been
exceeded
because
the charcoal filter replacement
took
approximately
24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> out of an allowed outage time of 7 days.
As corrective actions,
the licensee
placed signs
on the ventilation
filter units to warn personnel
to not perform work unless the units were
isolated.
The licensee
also changed
maintenance
and surveillance
procedures
to install
a blank filter isolation when work was performed.
The inspector
reviewed the licensee's
dose calculations for this issue
and the changes
made to licensee
procedures.
The inspector
concluded
that the licensee's
corrective actions
were adequate.
III. En ineerin
Conduct of Engineering
EDG Volta e
Re ulator Circuit Board Deficiencies
Ins ection Sco
e
37551
The inspectors
reviewed the licensee's
actions taken in response to the
identification of concerns
associated
with the
EDG voltage regulator
ci rcuit cards identified in ARs A0399437,
A0399596,
and A0401207
~
and
Quality Evaluation
(QE) Q0011867.
The inspectors
also interviewed
engineers
associated
with the licensee's
investigation.
Observations
and Findin s
The inspectors
developed
a time line for the licensee's
actions after
identification of six concerns
documented
on AR A0399437
on April 17,
1996.
The inspectors
noted that the
AR identified
a potential
concern
that printed circuit cards in the voltage regulators for the six EDGs
were not in accordance
with design drawings.
The inspectors
noted that
-12-
the licensee initiated
a second
AR (A0399596)
on July 18 to address
only
the first two issues identified on AR A0399437.
On July 19, the
licensee
determined that the issues
were such that validation was needed
to determine the impact on operability.
The inspectors
noted that
Inter-Departmental
Administrative Procedure
OH7. ID8,
"Oper ability
Evaluation," Revision 2, allowed the licensee
30 days to determine if a
degraded
condition existed;
however,
a
POA was requi red to be completed
within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of determining that
a degraded
condition existed.
A
degraded
condition, in accordance
with Procedure
ON7. ID8, is any
"[p]hysical or documentation
evidence
(including design,
analysis,
licensing
or qualification) of deg'radation of the ability of a
structure,
system,
and component to perform its Specified Safety
Function(s)."
On April 22, the system engineer
documented that the printed circuit
boards in EDGs 1-1. 1-3,
and 2-1,
had undergone
a number of tests,
including the
18 month surveillance tests,
without experiencing
any
problems related to not having
a capacitor
in the field voltage feedback
circuit.
The inspectors
noted that Procedure
OH7.108 stated that ".
.
.
performance of surveillance
requirements
alone
may not be sufficient to
determine Operability
.
. ."
The inspectors
observed that the system
engineer
provided
a replacement
schedule for the affected circuit
boards.
The inspectors
found that the conclusion of the system engineer
that the
on the basis of their passing the tests
was
not substantiated
by additional evaluation,
which could have
analytically demonstrated
that the testing accounted for the degraded
condition.
On April 22. the supervisor for procurement
engineering
rejected the
initial AR (A0399437)
and transferred all actions, to the second
AR (A0399596).
The licensee
determined that the sixth issue identified
on the initial AR was not valid.
The sixth issue alleged that the
printed ci rcuit boards
had been
upgraded for safety-related
use.
The
inspectors
agreed with the evaluation that the subject printed circuit
boards
had not been
upgraded,
but had been
purchased
as safety-related
in 1978 in accordance
with the regulations
in effect at that time.
The
licensee
determined that the root cause of the configuration control
problem was that the incorrect circuit boards
had been supplied
by the
vendor.
At the time the circuit boards
were procured the vendor did not
identify any differences
between the typical schematic of the voltage
regulator
and the printed circuit boards that were supplied to the
licensee.
On April 24, the licensee
issued
Work Order C0144207 to replace the
affected voltage regulator printed ci rcuit board in
EDG 2-2.
The
inspectors
considered that this action confirmed that the licensee
had
determined that
EDG 2-2 was in a degraded
condition:
however,
no
POA was
documented,
as required
by Procedure
OM7. ID8.
0
-13-
On April 27,
a vendor representative
informed the licensee that the
proper configuration of the printed circuit boards
requi red two
resistors
in parallel in a filter circuit to minimize the effects of the
firing of the silicon-controlled recti fiers on the field voltage
circuit.
Also, the vendor representative
stated that if only one
resistor
was in the circuit. the single resistor would overheat
and
cause
scorching.
In addition to the two resistors.
the vendor
representative
stated that there were differences
between the installed
circuit b'oards
and the design in the feedback ci rcuit.
These
differences affected
a capacitor
and two additional resistors.
The
licensee
determined that the subject printed circuit boards in the
voltage regulators for EDGs 1-2, 1-3, 2-2,
and 2-3 did not conform to
the design drawing and should have
been replaced.
The inspectors
found
that this information was sufficient for the licensee to determine that
.a degraded condition existed;
however, the licensee failed to promptly
document
as required
by Procedure
ON7. IDB.
On Nay 2, the system engineer,
along with the senior
resident inspector.
found that the single resistor
on the subject circuit board in the
voltage regulator for EDG 1-3 had visible signs of scorching.
The
system engineer
concluded that the
because
".
.
. there
was
no immediate danger that the board would fail
.
. ."; however, the
system engineer
recommended
that the board
be replaced
as
soon
as
possible.
The inspectors
found that the scorched resistor
was yet
another indication of a degraded condition;
however, the licensee failed
to document
a
POA as required
by Procedure
ON7. IDB.
On May 7.
a system engineering
supervisor
documented that on the basis
of "previous successful
surveillance testing,
discussions
and
observations
described [in AR A0401207] and engineering
judgement, it
[was] expected that
EDG 1-3 voltage regulator operate correctly and i s
in no immediate
danger of fai ling."
Also on May 7, licensee
management
determined that
a
POA was required.
The inspectors
noted that the
system engineering
supervisor
documented
the
POA, in accordance
with
Procedure
ON7. IDB, later
on May 7,
and that the Director of
Instrumentation
and Controls Engineering concurred.
The inspectors
found that the licensee
had at least
four opportunities
to appropriately declare
a degraded
condition for four of the six EDGs.
At each opportunity, the licensee
was aware that. either the installed
configuration did not match the design
(physical
and documentation
evidence)
or the resistors
were scorched
(physical evidence).
In
addition. the vendor representative
informed the licensee that the
installed configuration did not match the design.
The inspectors
determined that the licensee failed to appropriately identify a degraded
condition and failed to properly perform
a
POA in accordance
with
Procedure
ON7. ID8, Revision 2.
The failure to comply with
Procedure
OM7. IDB was
a violation of Criterion
V of Appendix
B to 10 CFR Part 50.
This was identified as violation (VIO 50-275/96016-96:
50-323/96016-06).
0
-14-
E1.2
Conclusions
The fai lure to promptly identify a degraded
condition and perform
a
POA
in accordance
with procedural
requirements
was identified as
a violation
of Criterion
V of Appendix
B to 10 CFR Part 50.
Ins ection of Premature
MSSV Lifts
Ins ection Sco
e
92903
On August 10.
1996, following reactor trips on Diablo Canyon Units
1 and
2.
one
HSSV on Unit 1 and two MSSVs on Unit 2 lifted at pressures
below
their requi red set pressure.
The inspectors
reviewed the licensee's
actions in response
to this event.
Observations
and Findin s
Following a dual unit trip, one
HSSV (RV-7) on Unit
1 and two HSSVs
(RV-4 and RV-8) on Unit 2 lifted prematurely.
Prior to the trip, on
August 7-8,
1996, the licensee
had tested the Unit 2 valves in place
using the AVK hydraulic assist
device.
The licensee
performed the
testing
as part of their augmented test program to determine the reason
for the
MSSV high initial lifts and set pressure drift previously
experienced.
Both Unit 2 valves which lifted prematurely
had their set
pressure
adjusted during this test series.
The valve set pressure
is
adjusted
by turning the adjusting bolt and either compressing
or
relaxing the spring.
The number of flats turned
on the adjusting bolt
indicates
how much
a valve was required to be adjusted.
Each flat on
the adjusting bolt typically corresponds
to a change in the set pressure
of approximately
5 to 20 psi.
The Unit
1
HSSV had its set pressure
adjusted during testing
on
April ll, 1996.
The inspectors
reviewed the licensee's
test data
from
April ll, 1996,
and found that .the Unit 1 Valve,
RV-7, had
an "as-left"
set pressure of 1065 psig.
During online testing the
MSSV adjusting
bolt had been adjusted
a total of -4.5 flats from the valve setting
on
live steam at the Westinghouse test facility.
The -4.5 flats adjustment
caused
a decrease
in the MSSV's set pressure.
During the plant trip,
the
MSSV lifted at 1014 psig.
A review of the Unit 2 test results of
August 7-8,
1996 found that
RV-4 had
an
AVK "as-left" set pressure of
1086 psig with a total adjustment of'5.0 flats of the adjusting bolt
from the valve setting
on live steam at the test facility.
Valve RV-8
had
an "as-left" set pressure of 1083 psig with a total adjustment of
the set pressure
adjusting bolt of -4.0 flats from the valve setting
on
live steam at the test facility.
During the plant trip, RV-4 had opened
at 1020 psig and RV-8 at 1014 psig.
After the three valves
had lifted low. the licensee
retested
the valves
using the AVK device.
AVK test results indicated that the set pressures
were approximately the
same
as the "as-left" values prior to the plant
0
0
trips.
The AVK test results did not reflect the low set points recorded
by the plant process
computer
.
Following the
AVK tests,
the licensee
performed additional testing using Trevitest equipment
on eight of the
forty MSSVs and establish
set pressures
independent of AVK testing
and
the plant process
computer.
The inspectors
reviewed the Trevitest test
data
and found that the "as-found" set pressures
were
1035 psig for
RV-7,
1055 psig for RV-4, and
1032 psig for RV-8.
The Trevitest results
correlated closer to the results
recorded
by the plant process
computer
than the
AVK results did.
After the licensee
reviewed the test data for
the eight valves,
they proceeded
to test the remaining valves
on both
units with the Trevitest device
arid reset
them according to the
Trevitest results.
During the last outages for Units
1 and 2, the licensee
developed
individual valve correction factors to be used
when testing the valves
in place using the AVK test apparatus.
These individual valve
correction factors were developed
by comparing
AVK test results with
steam test results.
The licensee
applied correction factors
by changing
the valve seat
area utilized to calculate lift pressure.
These
correction factors resulted in the effective seat
area of individual
valves ranging from 20.18 square
inches to 24.27 square
inches.
The
manufacturer of the AVK device
recommended
using
a 22.46 square
inch
mean seat
area.
The manufacture
of the Trevitest device
recommended
using
a mean seat
area of 23.046 square
inches.
When
a hydraulic lift
assist
device is used,
the valve set pressure
is determined
by adding
the valve inlet pressure to the load cell force divided by the mean seat
area.'
low mean seat
area would cause the hydraulic lift device
contribution to the set pressure to be proportionately too high.
Therefore,
the valve would be set too low resulting in the valve lifting
at
a lower than expected
pressure.
The three valves that opened at low
pressures
had been adjusted
using low mean seat
areas.
A review of the
AVK and Trevitest test data
found that when tested with
the
AVK device,
the valves which had a, low mean seat
area
had the
largest
number of flats adjusted to decrease
the set pressure after
having been set
on steam at the Westinghouse
Test Facility. It was also
noted that,
when tested with the Trevitest device,
the
same valves with
the low AVK mean seat
areas
were found to have low set points
and had to
be adjusted
by increasing the set pressure.
These adjustments
were
approximately equal
but opposite to the adjustments
made when tested,,
with the
AVK device.
The inspectors
found that the Trevitest results
correlated
more closely with the steam test results
than the AVK results
with valves with low mean seat
areas.
The inspectors
reviewed Operability Evaluation. 96-06, "Operability of
Main Steam Safety Valves," Revision 0.
The purpose of the operability
evaluation
was to document the operability of the
MSSVs for Units
1 and
2.
The li'censee
concluded that the TS required
number of MSSVs were
operable while the plant was in Mode 3 and that Unit 2 could increase
power up to 87 percent before the remaining
MSSVs had to be tested.
The
0
-16-
licensee
concluded that the incorrect correction factors
(mean seat
areas)
could have contributed to the
MSSVs lifting low and the cause of
the valves lifting low was primarily due to the low set points
established
using AVK.
In addition, the licensee stated that the
continuing formal augmented test program for the
HSSVs to determine the
reason for the high initial lifts and set pressure drift would be
revised to reflect the recent testing experience.
The licensee stated
that the program would be ready for Plant Staff Review Committee
approval
by August 30.
1996.
On August 10,
1996, the licensee
was performing surveillance tests
on
the Unit
1 Lead
1
MSSVs using the
AVK device prior to the plant trip.
During the surveillance tests
the licensee incorrectly used the Unit 2
1 individual valve mean seat
areas
on the Unit
1 valves.
AR A410913410913
dated August 10.
1996,
was written to document the failure
to use the correct
mean seat
areas
and to document the fact that three
of the five HSSVs were outside of the
TS as-left set point +1 percent
tolerance but within the +3 percent tolerance.
As discussed
above,
the
valves were retested
and adjusted,
as necessary.
using the Trevitest
lift assist
device.
The testing
was performed prior to Unit 1
reentering
Hode 2.
The failure to use the correct
mean seat
areas
was
identi fied as
a violation of Procedure
HP M-4. 18, Revision 14,
"Verification of Lift Point Using Ultra Star Assist Device for the Main
Steam Safety Valves."
Procedure
No. OPl.DC17, Revision 2A, "Control of Equipment Required
by
the Plant TS."
requires that whenever
TS equipment is declared
a
TS sheet
should
be initiated.
The licensee
determined
that three of the Unit 1
HSSVs were outside of their "as-left" set point
tolerance
on August 10.
1996,
when they identified that the Unit 2 mean
seat
areas
had been
used
on the Unit 1 valves.
The licensee did not
'nitiate the
TS sheets until August 12,
1996.
Although the TS action
statement" requirements
were followed and. therefore.
no violation
occurred.
the inspectors
considered this to be
a weakness.
Conclusions
E2
During surveillance testing of Unit
1 Lead
1 MSSVs, the licensee
used
incorrect
mean seat
areas
which caused three of the five valves to be
outside of the "as-left" set pressure
tolerance of +1 percent.
This
issue is being considered
as
an unresolved
item since further inspection
is requi red to determine whether the licensee
complied with the
requirements of the maintenance
rule (URI 50-275/96016-07).
Engineering
Support of Facilities and Equipment
Review of U dated Final Safet
Anal sis
Commitments
A recent discovery of a licensee operating their facility in a manner
contrary to the
UFSAR description highlighted the need for a special
0
-17-
focused review that compares
plant practices,
procedures.
and/or
parameters
to the
UFSAR description.
During the inspection period, the
inspectors
reviewed the applicable sections of the
UFSAR that related to
the inspection
areas
discussed
in this report.
There were no
inconsistencies
noted between the wording of the
UFSAR and the plant
practices,
procedures,
and/or parameters
observed
by the inspectors.
E8
Hiscellaneous
Engineering
Issues
l
f8.1
Closed
LER 50-275/95-011-00:
Unit 1 Hain Steam
System outside of
design basis
due to high initial HSSV lift points.
Testing to identify
the cause
and to resolve problems associated
with high initial MSSV lift
points is ongoing.
LER 1-96-003
has
been
opened to report the results
of continued testing,
investigation,
and corrective actions for the
HSSVs in Units
1 and 2.
No further inspection
on
LER 50-275/95-011,
Revision
00 is necessary,
since the issue is being tracked
under
LER 1-96-003.
X1
Exit Meeting Summary
V. Mana ement Meetin s
The inspectors
presented
the inspection results to members of licensee
management
at the conclusion of the inspection
on August 22.
1996.
The
licensee
acknowledged
the findings presented.
The .inspectors
asked the licensee
whether
any materials
examined during the
inspection should
be considered
proprietary.
No proprietary information was
identified.
X2
Predecisional
Enforcement
Conference
Summary
On July 1,
1996,
a predecisional
enforcement. conference,
open to public
observation,
was conducted at the Region
IV office in Arlington, Texas.
The
purpose of this meeting
was to discuss
the April 1996 augmented testing of the
Diablo Canyon Unit
1 HSSVs
and subsequent
actions
as described in NRC Special
Inspection Report 50-275/96-12;
50-323/96-12.
Meeting attendees
are listed in
Attachment 2.
The apparent violations as originally proposed
and discussed
at
the meeting are presented
in Attachment 3,
and the licensee's
presentation
is
provided in Attachment 4.
The action taken
by the
NRC following the
inspection
and predecisional
enforcement
conference
was documented
in a letter
dated July 10,
1996.
0
ATTACHMENT 1
Licensee
PARTIAL LIST OF
PERSONS
CONTACTED
M. J.
Angus,
Manager,
Regulatory
and Design Services
J.
R. Becker, Director, Operations
J.
E. Bonkonsky,
General
Foreman,
Technical
Maintenance
D. F. Brosnan,
Director, Regulatory Services
W.
G. Crockett,
Manager,
Nuclear Quality Services
S.
R. Fridley, Manager of Outage Services
C.
R. Groff, Director, Nuclear Secondary
Systems
J .
R. Hinds, Director, Quality Control. Nuclear Quality Services
T. L. McKnight, Engineer,
Regulatory Services
D.
B. Miklush, Manager,
Engineering Services
E.
P. Nelson, Materials Supervisor.
Procurement
Engineering
D. A. Vosburg, Director, Nuclear Steam Supply Systems
Engineering
l,
-2-
IP 37551:
IP 61726:
IP 62707:
IP 71707:
IP 92700:
IP 92902:
IP 92903:
INSPECTION
PROCEDURES
USED
Onsite Engineering
Surveillance Observations
Maintenance
Observations
Plant Operations
Onsite
LER Review
Followup - Maintenance
Followup - Engineering
~oened
50-275/96016-01
50-323/96016-01
50-323/96016-02
50-275/96016-03
50-323/96016-03
50-323/96016-04
50-275/96016-05
50-275/96016-06
50-323/96016-06
50-275/96016-07
Closed
50-275/96016-01
50-323/96016-01
50-323/96016-02
50-323/96016-04
50-275/96016-05
50-275/95-005-00
50-275/95-011-00
ITEMS OPENED,
CLOSED,
AND DISCUSSED
Failure to follow 10 CFR 50.72 Reporting
Requirements
Failure to follow CFCU operating
procedure
IFI
Installation of Improper
ASW Pump Shaft
Key
Inoperable Reactor Trip Permissive
P-4
Violation of TS 3.4.6. 1
Failure to take prompt and adequate
corrective actions following identification of
a degraded
condition
Unit
1 Lead
1 MSSVs'Set~ Using Unit 2 Lead
1
. Correction Factors
Failure to follow 10 CFR 50.72 Reporting
Requirements
Failure to follow CFCU operating
procedure
Reactor Trip Permissive
P-4
Violation'of TS 3.4.6. 1
LER
Technical Specification 3.4.6.1
Not Met with the
Leakage Detection Systems
Due, to Personnel
Error
LER
System Outside of Design Basis
Due to
High Initial MSSV Lift Points
I4
-3-
50-275/95008-00.
LER
Control
Room Ventilation System Outside Design
Basis
Due to
a 01,
and
02 Programmatic
Deficiency in the Operation
and Maintenance of
the
HEPA Filter and Charcoal
Absorber System
CFCU
kv
LER
OP
POA
s~co
TS
LIST OF ACRONYMS USED
action request
auxiliary saltwater
containment
fan cooler unit
control
room ventilation system
emergency diesel
generator
kilo-volt
licensee
event report
operating
procedure
prompt operability assessment
public document
room
pump
radiation monitor
senior control operator
shift supervisor
Technical Specification
Updated Final Safety Analysis Report
ATTACHMENT 2
LIST OF ATTENDEES
PREDEC ISIONAL ENFORCEMENT CONFERENCE
Licensee/Facility:
Pacific Gas
and Electric Company/
Diablo Canyon Unit 1
Date/Time:
SUBJECT:
July 1,
1996,
2:30 p.m.
APRIL 1996
AUGMENTED TESTING OF DIABLO CANYON UNIT 1 MAIN
STEAM SAFETY VALVES AND SUBSEQUENT ACTIONS AS DESCRIBED IN
NRC SPECIAL INSPECTION
REPORT
50-275/96-12;
50-323/96-12
Licensee Attendees
S. Allen, Senior
Engineer,
Valve Engineering
J. Alviso. Engineer Assistant,
NRC Interface
M. Angus.
Manager.
Regulatory
and Design Services
P.
Beckham.
Senior Consultant
D. Brosnan.
Director, Regulatory Services
B. Coley, Engineer.
Regulatory Services
C. Groff, Director, Engineering Services
R. Locke, Attorney
D. Miklush, Manager,
Engineering Services
G.
Rueger
~ Senior Vice President,
Nuclear
Power Generation
L. Womack,
Vice Presidents
Nuclear Technical Services
NRC Attendees
W. Bateman,
Project Director
S.
Bloom. Project
Manager
S.
Boynton, Resident
Inspector
K. Brockman,
Deputy Director, Division of Reactor Safety
S. Collins, Deputy Regional Administrator
D. Corporandy,
Project Engineer
P. Goldberg,
Reactor inspector
B. Henderson'ublic Affairs Officer
K. Perkins,
Director. Walnut Creek Field Office
G. Sanborn,
Enforcement Officer
M. Tschi ltz, Senior Resident
Inspector*
H.
Wong, Chief, Reactor Project
Branch
E
- By telephone
ATTACHMENT 3
APPARENT VIOLATIONS AS ORIGINALLY PROPOSED
DIABLO CANYON
APPARENT VIOLATION j.
10 CFR Part 50, Appendix B, Criterion XVI, "Corrective Action," requires that
measures
shall
be established
to assure that conditions adverse to quality,
such
as failures, deficiencies,
and deviations,
are promptly identified and
corrected.
Contrary to the above.
between April 2,
1996,
and April 14,
1996,
a condition
adverse to quality (out of tolerance
(MSSVs)) existed
that was not promptly identified or corrected.
Specifically,
on April 2,
1996
~
the licensee identified by testing that the pressure liftsetpoints
on three of
five MSSVs on Unit
1 exceeded
allowable Technical
Specification tolerances.
The licensee did not take prompt corrective action
to identify. and correct the out-of-tolerance deficiencies
on the remaining
The
MSSVs on main steam
lead were not tested until April 11,
1996,
when
three of the five MSSVs were found to be out-'of-tolerance.
The
MSSVs on main
steam
3 and 4 were not tested until April 14,
1996,
when six of ten
were found to be out-of-tolerance.
This apparent violation is subject to change
based
upon discussion.
0
-2-
DIABLO CANYON
APPARENT VIOLATION 2
CFR Part 50, Appendix 8, Criterion V, states,
in part, that activities
affecting quality shall
be prescribed
by documented
instructions,
procedures.
or drawings
.
.
.
and shall
be accomplished
in accordance
with these
instructions,
procedures,
or drawings.
Diablo Canyon Procedure
ON7. ID8, Revision 2, Subsection
2.2.3,
requires
that:
"For Degraded Conditions impacting Structure.
System
and Component
operability identified by physical
evidence at
DCPP. the
POA (Prompt
should
be completed
and documented
during the
operating shift in which the physical
evidence
was identified.
In all
cases'he
POA shall
be completed
and documented within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />
following identification of a Degraded Condition."
Contrary to the above.
as of May 14,
1996,
NRC inspector s identified that
the licensee
had not documented
a
POA of a degraded
condition
(n'amely
three of five HSSVs out-of-tolerance).
On April 2,
1996, licensee test
engineers
identified three of five HSSVs on steam
1 ot Unit
1 to be
5
out-of'-tolerance
(high).
b.
Diablo Canyon Procedure
ON7. 108, Revision 2, Subsection
4.1,
requi res
that "The individual and his/her group supervisor
identifying a Degraded
Condition or an Issue
Needing Validation is responsible for:
Immediately
0
-3-
noti fying the Shi ft Foreman,
if the condition is an observed
physical
Degraded Condition at the plant that could adversely affect the
Operability of a Structure,
System,
and Component."
Contrary to the above,
on April 2,
1996,
licensee test engineers
identified three of five IISSVs on steam
1 of Unit
1 to be out-of-
tolerance
(high), but did not notify the Shift Foreman.-
This apparent violation is subject to change
based
upon discussion.
0
ATTACHMENT 4
LICENSEE PRESENTATION SLIDES
FOR PREDECISIONAL ENFORCEMENT
CONFERENCE
0
LICENSEE'
PRESENTATION
~ DCPP Main Steam Safety Valves (SV)
Background
~ Inspection Report Issues
Unit 1 Correction Factors
Formality and Conduct of Augmented Test
Program
SV Operability Evaluation and Timeliness
~ Safety Significance
~ Concluding Remarks
Ã!
OCF'F'NAIN STEAM SAFETY
VAL.VE BACKGROUNQ
~ DCPP SV Type, Purpose, and Design Basis
~ SV Testing History
~ Program to Resolve Issues
~ Findings To-Date
~ Future Program Activities
Pageos
0
hi
SAFETY VALVETYPE
~ Dresser Model 3707R
~ 20 installed on each unit and 3 spares
~ Setpoints 1065, 1078, 1090, 1103, and 1115 psig
~ Common SV, used at other plants
NIfll
SV PURPQSE 8 DESIGN BASIS
~Ori inst
Present
/-1%
+/-3% (1 065 +3/-2%)
/-1%
/-1%
- As-left
- Test frequency each refueling outage.
(ASME Code would
allow 5 years for 20 valves)
t
Prevents Steam Generator (SG) pressure from rising above
110% of design pressure.
Limiting event is:
Turbine trip without a reactor trip (MFW maintained)
Maintain minimum AFW flow, 410 gpm (upper setpoint limits)
Prevent SG overfill (lower setpoint limits)
Technical Specifications
As-found
'"5 3
Page 2
SV TESTING HISTORY
~ Trevitest instrumentation historically used at DCPP and
much of the industry for SV surveillance testing and
setpoint adjustment
~ Maintaining SVs within+/-1% of setpoint has been a
problem since QCPP startup
~ SV out-of-tolerance (OOT) performance an industry
issue per NRC document AEOD 4/92
~ Trevitest validity brought into question (NRC IN 94-56)
~ 1994: To resolve SV issues, PGKE took the initiative to
pursue a comprehensive research and corrective
actions program
"63
PROGRAM TQ RESOLVE ISSUES
~ Obtain improved test instrumentation
~ Reanalyze basis for+/-1% TS setpoint tolerance to
support a wider band of acceptability
~ Develop a comprehensive understanding of SV
operating, testing, and maintenance characteristics
~ Ifresearch findings dictate, redesign/refurbish SVs to
resolve setpoint problems
Pll
Page 3
FINQiNGS TO DATE
~ AVKtest instrumentation selected and purchased
AVKresults repeatable within an acceptable
distribution
AVKand PG8E test data substantiates
AVK
mean seat area (MSA) adequate for on-line
SV setting
~ Re-analysis allows change in TS limits
Design basis supported with +/- 3% {+3/-2%
for 1065 SVs)
LAR approved on 10/1/95 with augmented
test program on Unit 1
Pl
AL.TERNATIVE TES T/NG
METHODS COlNPARISON
Page 4
0
FINDiNGS TO DATE (cont'd)
~ SV setpoint distribution considered during setpoint
adjustments
~ SVs exhibit initial high lifts
Believed to be due to thermal bonding which arises
from seating surface galling during thermal cycle
differential expansion
Thermal bonding not observed at beginning of cycle
Thermal bonding present 120+ days into cycle
'inimal
thermal bonding once SVs lifted "hot"
Seat refurbishment and materials appear to increase
" likelihood of bonding
~ SV liftvariation
Torque differences, installation vs test stand (Theory)
SVs spring hysteresis (Theory)
I4
TRENDS!N INITIALLIFTS
8FI
Page 5
0
FUTURE PROGRAM AC T!VlTIES
~ AVKtest methodology validation
Complete uncertainty analysis
~ Initial high lifts
Complete seating materials replacement
evaluations
'
Test all SVs on return to power from Mode 5
outages
3% accumulation on "sticking" SVs can be
eliminated for operability and safety analysis
~ Setpoint variation
SV-specific performance evaluation
Potential SV parts improvement program
~ Monitor industry SV experience
Eall
!NSPEC TION REPORT ISSUES
~ Unit 1 Correction Factors
~ Formality and Conduct of Augmented Test
Program
~ SV Operability Evaluation and Timeline'ss
~ For each issue PG8 E willcover:
Events Chronology
Key Conclusions
Corrective Actions
twirl
Page 6
0
CORRFCT/ON FACTQRS
CHRONOLOGY
~ 9/23/95, U2 Trip, 3 SVs lifted low, AVKvs steam
correction factor evaluation considered necessary
~ 9/30/95, LAR 95-06 submitted requesting setpoint
tolerance relaxation
~ 10/1/95, NRC LA 108 and 107 issued, requiring
augmented test program
~ 10/8 - 18/95, Unit 1 SVs sent to Westinghouse
Western Service Center (WSC) for testing/adjustment
on steam
Set all SVs on steam for normal TS testing
Collected AVKCF data for all SVs
o
Data for 10 SVs subsequently determined to
be invalid
o
Data*for 1 SV lost
/~II
CORRECTION FACTORS
CHROWOLOGV conm
~ 11/1/95, DCL-95-241 described formal augmented test
program and indicated that SV-specific CF testing would be
obtained during 1R7 and 2R7. Regulatory Services failed
to document Unit 1 CF commitment on t'racking AR.
~
11/1 5/95, Engineering verbally informs NRC Resident of
installation of 11 SVs without CFs and agreed this is not a
restart issue.
Regulatory Services notinformed by
Engineering.
~
11/15/95, Action Request (AR}written outlining an
alternative process to obtain AVKCF data on-line.
Subsequent
PSRC discussion questions validity of this
approach.
As a result, further testing was not pursued
~ 11/27/95, Unit 1 restart.
Commitment tracking omission
results in a missed opportunity to identifyneed forwritten
NRC notification
Page 7
0
CORRFC TION FACTORS
CHRONOLOGY conf'd
~ 4/1 5-26/96, Unit 2 SVs sent to WSC for testing and
adjustment on steam.
Set all SVs on steam for normal TS testing
Collected AVKCF data for all SVs
Unit 2 results substantiate both Unit 1 data and
AVKMSA
~ 5/3/96, OE 94-02, Rev 5 updated to include Unit 1
Augmented Testing experience.
~ 6/26/96, OE 94-02, Rev 6 concluded that CFs of the
remaining SVs were not required since the
overpressure
and AFW design basis was met with
worst case as-left setting using Vendor supplied MSA
"53
CF CONCLUSIONS AND
LESSONS LEARNED
~ AVKand PG8 E test data substantiates
that use of
AVKMSA is adequate for on-line SV setting of all
valves
~ Communications were inadequate between
Regulatory Services and the responsible Engineering
organization doing the testing
~ Commitment tracking AR not written for Unit 1 CFs.
Cognizant personnel failed to promptly inform NRC
and PG8 E management
~ PG8 E should have communicated with NRC in a
more complete and formal manner regarding the
decision to restart without all Unit 1 CFs.
NB
li'r5 3
Page 8
0
CF CQRRECTlVE ACTIONS
~ Updated OE 94-02 to address CFs (ref Rev 6)
~ Commitment tracking process reviewed relative
to observed problems.
Existing process deemed
adequate
but ineffective in this circumstance due
to personnel error
~ Management expectations regarding licensing
commitment implementation and communications
(both internal PG8 E and NRC) re-emphasized
with responsible personnel
~ Technical Staff training to review appropriate
licensing commitment implementation and NRC
communication requirements
s agr
~ NRC required an augmented testing program as
part of TS limitchanges
~ PG8 E letter of 11/1/95, detailed planned program
PGB E viewed program as an augmentation,
not a TS test for sample expansion
This understanding
is acknowledged
in NRC
Safety Evaluation Report of 12/26/95
Augmented testing for Unit 1 only
~ Initiated to evaluate the potential time
dependency of seat bonding
~ Finding high initial lifts during the augmented test
program was expected
Fl
Page 9
AUGMENTED TES T PROGRAM
PLAN
~ Testing requested
on an Action Request (AR);
documented on Work Order; performed under
Maintenance Procedure MP M-4.18
~ Originally, augmented testing program to be
performed on Unit 1 SVs only
~ Testing performed on one header on a staggered
basis to preserve time dependency
~ Valves outside +/-1% will be reset
~ If a liftoutside +/-3% occurs, test expansion to be
discussed with NRC per augmented test program
agreement
~ TS testing to occur as normally scheduled prior to
2R7 and 1R8
Pl
AUGMENTED TEST PROGRAM
RATIONALE
~ TS provided operability requirements
~ OE 94-02, Rev 4 provided operability basis for high
initiallifts
~ DCPP practice, even under TS surveillance, was not to
project test results from one SV to another while in
testing activities unless an equipment concern existed
regarding the adequacy of last setpoints established
~ Regular TS surveillance requires automatic expansion
for unsatisfactory results
~ Augmented testing program expansion was not
automatic; consultation with the NRC was required to
determine the extent of expansion
Page 10
0
AUGMENTED TES T PROGRAM
LOOP 7 CHRONOLOGY
~ 4/2/96,
Loop
1 SVs tested (3 of 5 OOT high) and
returned to +/-1%
The degree of OOT not unusual
Operations aware of testing and that all SVs
returned to within TS requirements
~ 4/3/96, Shift Foreman informed of Loop 1 test
r suits (approximately 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after tests)
'5$
AUGMENTED TES T PROGRAM
LOOP I CHRONOLOGY cont'd)
~ 4/3/96, Reviewed data to assess
operability
Existing OE 94-02, Rev 4 bounded test
results
OE considered AFWflow, although could
have been more clear
~ 4/4/96, Analysis results received for Loop 1 test
results
~ 4/8/96, PG&E/NRC conference call
Loop 1 testing results discussed
Potential expansion explored
PGRE and NRC agreed to test 5 more SVs
under the augmented test program'!
Page
11
AUGMENTED TEST PROGRAM
LOOP 2,3,4 CHRONQLOGY
~ 4/10-11/96, Loop 2 SVs tested (3 of 5 OOT high) and
returned to +/-1% except RV-223
The degree of OOT raised concern
Work orders requested to test Loops 3 and 4
RV-223 left at+1.4% since no CF was available and
steam setting was considered more accurate
~ 4/12/96, PG&E/NRC call at 1000 PDT.
PG&E informed NRC that Loops 3 and 4 would be
tested on 4/14/96
SVs < 1.5% without CFs would not be reset since live
steam was considered more accurate
Operable since OOT SVs reset and existing OE
applicable
Expeditious testing of Loops 3 and 4 planned
'53
AUGMENTED TEST PROGRAM
LOOP 2,3,4 CHRONOLOGY(conf'd)
~ 4/12/96, Operations made 10 CFR 50.72 report to NRC at
approximately 1730 PDT on SG 1-2 outside design limits
~ 4/12/96, 2130 PDT, NRC called requesting clarification on
1-hour report and conditions of Loops 3 and 4 ifprojection
of Loop 2 results was made to Loops 3 and 4
PG&E does not believe projection is appropriate, but
responded that Loops 3 and 4 would be operable
PG8 E's conclusion was based on incorrect
understanding of the analysis model
'3
Page 12
0
AUGMENTED TEST PROGRAM
LOOP 2,3,4 CHRONOLOGY(cont'd)
~ 4/13/96, PG8 E reviewed the validity of previous evening
response to NRC
Original basis found to be incorrect
Loops 3 8 4 were still considered operable based on
availability of 10% ADVs
NRC not immediately notified of change in reason
behind operability statement
~ 4/14/96, Loops 3 (5 of 5OOT) and 4 (1 of 5OOT) tested
and reset to +/-1% except RV-14
RV-14 left at+1.2% since no CF was available and
steam setting was considered more accurate
~ 4/16/96, PG&E/ NRC call on Loops 3 and 4 test results
and proposed Unit 2 testing.
NRC informed of PG8 E
miscommunication on 4/12 regarding basis of operability
statement
AUGMENTED TEST PROGRAM
LOOP 2,3,4 CHRONOLOGY (cont'd)
~ 4/19/96, PG8 E/NRC call
Agreed to test Unit 2 SVs in Mode 3
Expand the augmented test program to Unit 2 for
cycle 8
NRC stated that Unit 2 SVs must be returned to+/-1%
~ 4/21/96, PG&E initiated actions to ensure all Unit 1 SVs
returned to +/-1%
RV-14 was adjusted
RV-223 was tested and required no adjustment
~ 5/3/96, OE 94-02, Rev 5 approved to encompass
recent
test results
~ as
~
Page 13
0
CONDUCT OF TEST PROGRAM
CONCLUSIONS
~ Formality of the test program was adequate
Work was performed under an approved Work
Order and in accordance with MP 4.18
PG8 E decided not to use STP M-77. This
decision was consistent with our 11/1/95 letter
and the NRC's understanding noted in the SER
dated 12/26/95
The test program was carried out as planned.
Automatic expansion was not called for prior to
discussion of results with the NRC
PG8 E agrees that the Shift Foreman should
have been informed of results earlier, but it is
believed that doing so would not have resulted
in accelerated
testing of SVs
'3
CONDUCT OF TEST PROGRAM
CONCLUSIONS cont'6
~ Formality of the test results evaluation was not
adequate
PGKE should have anticipated and planned
appropriately given the range of results
obtained in prior SV testing
Information exchange requirements (type and
timeliness) with the NRC should have been
specified in the testing plan
Once it was known that valve specific CFs
were not available for 11 SVs, the effect on
results evaluation and SV reset action should
have been clearly analyzed and incorporated
in the testing plan
5FI
Page 14
CONDUCT QF TEST PROGRAM
CORRECTIVE ACTIONS
~ Immediately reset all Unit 1 SVs initiallyleft outside
+I- 1% tolerance
~ Tested SVs on Unit 2 in Mode 3 and at full power (to
check for thermal bonding)
~ Included Unit 2 in augmented testing program
~ SV augmented test plan has been revised to include
test scope expansion and reset criteria
~ Technical Staff training willreview this event focusing
on technical data evaluation, formal communication
and compliance with commitments
~ Engineering personnel involved with the evaluation of
test data have been counseled regarding the need to
provide formal documentation
P'll
SV OF'ERABILITYEVALUATION4
TIMELINESS CONCLUSIONS
~ Based on analysis of Loop 1 results, there was no
immediate operability issue, affording time to
communicate and plan additional testing
~ Given the nature of the Loop 1 results, PG8 E's actions
not to project results for operability determinations was
believed appropriate and consistent with past practice
All Unit 1 SVs were set on steam during 1R7, thus
pre 1R7 results should not be "projected" to post
1R7 testing
~ OE 94-02, Rev 4, was judged to bound Loop 1 test
results and therefore a POA was not required
Pll
Page 15
SV OPERABILITYEVALUATION8
TIMELINESS CONCLUSIONS (cont'd)
~ Though judged adequate
at the time, OE needed
clarification forAFW(revised 5/3/96)
~ The reasoning behind the adequacy of OE
should have been documented
~ PGB E initial communication with the NRC after
Loop 1 testing should have been more timely and
more complete
~ The implications of not resetting SVs steam set
without specific CFs and found less than 1.5%
OOT, though discussed with the NRC, should
have been communicated more fully and formally
SV OPERABILITYEVALUAT!ON4
TIMELINESS CONCLUSIONS (cont'd)
~ PG8 E should have notified the NRC on April 13,
1996 of the discovery that the "Operability under
projection" statement on April 12, 1996 was
based on erroneous assumptions
~ OE 94-02, Rev 5 and 6, should have been
updated in a more timely fashion
Page 16
SV OF'ERABILITYEVALUATIONE
TIMELINESS CORRECTIVE ACTIONS
~ Management expectations re-emphasized
with
'esponsible
Engineering personnel regarding:
Formal documentation of operability
Formality of special test programs
~ Technical staff training on operability
determination procedures (on-going)
~ Lessons Learned to be communicated at
technical staff sessions (future)
Emphasis on established schedules for
Operability Evaluation updates
Pl
SAFETY SIGNIFICANCE,
~ Though the analysis for Loops 2 and 3 shows that the
design basis pressure for the SG would have been
exceeded,
the system remained operable and would have
performed its event function because:
40% steam dump valves available
Atmospheric steam dump valves available
Reactor trip from turbine trip functional
~ SGs were hydro tested to 125% and designed for 5 hydro
tests at this pressure
~ AFW flows exceeded minimum 410 gpm requirement
T
~ AFW flownot affected by high initial lifts since credit not
assumed until 60 seconds following event initiation and
subsequent
lifts tend to return to nominal setpoint range
I I g
~
Page 17
SAFETY SIGNIFICANCE (conf'd)
~ Conclusions
No increased risk to public health
and safety
Safety significance is low
O'I
Page 18
&
0
DIABLOCANYON UNIT 1
TREVITEST
AVK
Range
(%)
<3%
3% <<5%
> 5%
1R1
- 2.2 to+
5.2
18
1R2
-4.1 to
+ 9.2
15
1R3
-2.6 to
+ 10.2
15
1R4
-1.5 to
+ 3.1
19
1R5
-1.6 to
+ &.7
16
1R6
- 1.7 to
+ 9.1
13
1R7
+ 0.4 to
+ 9.3
Mode 1
(Apr-96)
- 1.9 to
+ 9.3
10
TREVITEST
DIABLOCANYON UNIT 2
STEAM
MISC
AVK
2R1
2R2
2R3
2R4
2R5
2R6
Mid-
Cycle
2R7
Mode 3
May-96
Mode 1
Jttn-96'ange(%)
s3%
3% -5%
) 5%
- 0.5 to
+ 5.5
16
3v'
1.8 to
+ 3.4
-1.8 to
+ 2.7
20
-0.9 to
+ 8.3
17
-1.4 to
+ 8.8
- 1.7 to
+ 3.6
18
-0.9 to
+ 8.7
13
- 1.3 to
+ 4.5
18
-1.5 to
+ 5.0
17
- 1.2 to
+ 4.5
Qnty '10 Valves were tested
10
0
ALTERNAT/VE TES T/MG
ME7 HQQS CQMPAR/8QM
Test Method
LiftSensor
TREVlTEST
(In-Situ)
H. E."
Processor
Load Cell
None
Hydraulic with
Gripper (Man.)
Line Pressure
Heise Gauge
AVK
(In-Situ)
Acoustic
Transducer
Micro
Hydraulic with
Gripper (Auto.)
STEAM
(Test Stand)
H.
E.'eise Gauge
None
None
Correction
Factors
MSA/CF
- H. E.- Human Ear
Y"
r
~fl
0