ML16343A234
| ML16343A234 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 06/30/1994 |
| From: | Westerman T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML16342C611 | List: |
| References | |
| 50-275-94-17, 50-323-94-17, GL-89-10, NUDOCS 9407110063 | |
| Download: ML16343A234 (28) | |
See also: IR 05000275/1994017
Text
APPENDIX
U.S.
NUCLEAR REGULATORY COHHISSION
REGION IV
Inspection Report:
50-275/94-17
50-323/94-17
Licenses:
DPR-82
Licensee:
Pacific
Gas
and Electric Company
77 Beale Street,
Room
1451
P.O.
Box 770000
San Francisco,
Facility Name:
Diablo Canyon Nuclear Generating Station, Units
1 and
2
Inspection At:
Diablo Canyon Site,
San Luis Obispo Coun.y, California
Inspection
Conducted:
June 6-10,
1994
Inspectors:
C. Hyers,
Reactor Inspector,
Engineering
Branch
Division of Reactor Safety
H. Runyan,
Reactor
Inspector,
Engineering
Branch
Division of Reactor Safety
Approved:
esterman,
ie
,
ngineenng
rane
Division of Reactor Safety
ate
Ins ection
Summar
Areas
Ins ected
Units
1 and
2
Routine,
announced
inspection of open
items
from previous
NRC inspections of the licensee's
motor-operated
valve program
in response
Temporary Instruction 2515/109
and
Inspection
Hodules
93903
and
40500 were used
as guidance during the
inspection.
Results
Units
1 and
2
The licensee
has
shown good progress
toward the closure of their Generic Letter 89-10 program.
Host of the previously open inspection
issues
from NRC
Inspection
Report 50-275; 323/93-19
were closed
and
one of two open inspection
issues
from NRC Inspection
Report 50-275; 323/91-39.
These include the
following issues:
~
Hotor-operated
valve trending (Section 1.2),
94071l00b3
940705
ADOCN 05000275
8
0
~
Corrective action for Motor-Operated
Valve 2-LCV-109 (Section 1.3),
I
~
Evaluation of diagnostic signature for Motor-Operated
Valve 2-FCV-37
(Section 1.5),
Use of valve specific values in calculations
(Section 1.6),
Extrapol.ation of peak opening thrust for rising stem valves
(Section 1.7),
and
~
Clarification of the torque switch setpoints for limit-seated motor
control logic (Section 1.8).
Issues
remaining open,
which will be addressed
during the
NRC closure review
of Generic Letter 89-10 actions,, include the following:
~
Periodic verification of motor-operated
valve design. basis capability
(Section 1.1),
and
Pressure
locking and thermal
binding (Section 1.4);
Two observations
from the inspection include:
Licensee self assessments
of the motor-operated
valve program were
observed to be comprehensive
and technically oriented
(Section 2).
~
Two instances
were identified where the troubleshooting
performed did
not identify .the root cause of the operational
failures of
motor-operated
valves prior to returning the valves to operation
(Section 3).
Summar
of Ins ection Findin s:
Inspection
Followup Item 275;323/91-39-02 will remain
open
(Section 1.1).
Inspection
Followup Item 275;323/91-39-03
was closed
(Section 1.2).
Unresolved
Item 323/93-19-01
was closed
(Section 1.3).
Inspection
Followup Item 275;323/93-19-02 will remain
open
(Section 1.4).
Inspection
Followup Item 275;323/9319-03
was closed
(Section 1.5).
Inspection
Followup Item 275;323/9319-04
was closed
(Section 1.6).
Inspection
Followup Item 275;323/9319-05
was closed
(Section 1.7).
P
~
Inspection
Followup Item 275;323/9319-06
was closed
(Section 1.8).
t
Attachment:
~
Attachment
Persons
Contacted
and Exit Meeting
Detail s
1
FOLLOWUP OF
OPEN
ITEMS FROM PREVIOUS
NRC INSPECTIONS (93903)
1.1
0 en
Ins ection Followu
Item 275 323 9139-02:
eriodic Verification
During previous
NRC inspections,
the inspectors
had found that the licensee-
had not developed
a program for periodic verification of motor-operated
(HOV)
design basis capability.
Through discussions
with licensee
personnel
during this inspection,
the
inspectors
found that the licensee's
program for periodic verification was
still under development.
The licensee's
program for periodic verification was
scheduled
to be implemented
by December
1994, consistent
with their
commitments
in response
The licensee currently
planned to verify the design basis capability of each
HOV in their
program every three refueling outages
by static diagnostic testing.
The
licensee
stated that probablistic risk analysi~
wo~ld be used to prioritize
valves for testing
based
on safety significance.
The inspectors
discussed
with the licensee
the weaknesses
in the use of static testing only to verify
design basis capability.
The licensee
acknowledged
the inspectors'oncern
and stated that they were monitoring industry efforts in this area.
This 'item will remain
open pending the development
and implementation of the
licensee's
periodic verification program in December
1994.
Closing of this
item will be included
as part of the
NRC review for closure of GL 89-10.
NRC
acceptance
of static diagnostic testing for periodic verification of HOV
design basis capability is dependent
on the licensee
developing
an adequate
technical
basis.
1.2
Closed
Ins ection Followu
Item 275 323 9139-03:
HOV Trendin
During
a previous
NRC inspection,
the inspectors
noted that the licensee
had
not yet implemented a'rogram for HOV trending.
During this inspection,
the inspectors
found that Procedure
HA1. ID4, "Control
and Trending of Hotor-Operated
Valve Diagnostic Information," Revision 0,
had
been
issued
on September
28,
1993.
This procedure
provided guidance for
evaluating diagnostic test results
and identifying trends that
may indicate
impending problems.
The procedure
also provided
a method for trending
HOV
failures.
The inspectors
found that the licensee
had implemented
a computerized
trending
program.
The inspectors
observed
a demonstration
of the'program's
capabilities.
The trending information was used primarily as
a qualitative
indicator of degrading conditions.
According to the licensee,
quantitative
limits on allowable deviation
and degradation will be developed
as
more data
is accumulated.
HOV diagnostic test data
was=- loaded into the trending program
and evaluated within approximately
2 to
6 months following the test date.
The
I)
Ii[
r'
trended
parameters
included running load thrust,
unseating thrust, thrust at
torque switch trip, maximum thrust;- elapsed
time between
motor start
and
hammerblow,
current at torque switch trip,
and limit switch actuation points.
The inspectors
noted that valve factor,
stem friction coefficient, rate-of-
loading,(ROL),
and springpack
compression
were not trended.
The licensee
stated that adverse
trends related to these
parameters
could
be detected
indirectly by analyzing the other trended
parameters.
The inspectors
determined
that the trending
program
was acceptably
formulated
and sufficiently implemented to'permit closure of this part of the licensee's
commitment to
The fact that the trending
program was fully
implemented well in advance of the licensee's
GL 89-10 closure
commitment, date
(December
1994)
was considered
a program strength.
1.3
Closed
Unresolved
Item 323 9319-01:
Corrective Actions for
MOV 2-LCV-109
During
a previous
NRC inspection,
the inspectors
found that the licensee's
roubleshooting
and in'tial corrective actions for ar operational
failure of
MOV 2-LCV-109 in July
1993 did not appear
to be adequate
to identify and
correct the root cause of the failure prior to return of the
MOV to service.
The failure mode
was not reproducible
and the cause of the failure was not
evident during initial maintenance
troubleshooting.
The licensee
had
originally diagnosed
the cause of the failure to be
a faulty torque switch.
However, the torque switch was
found to be performing correctly during
troubleshooting.
Without finding any other obvious
component
problems,
the
licensee
continued to focus
on the torque switch
as being the cause
of the
failure and rationalized that the Failure
mode of the torque switch was
intermittent.
The licensee
replaced
the torque switch
as corrective action.
The inspectors
had con'sidered
the. adequacy
of the licensee's
corrective
actions to be
an unresolved
item pending review of open
Nonconformance
Report
(NCR)
No. DC2-93-EM-N037, which the licensee
had initiated to further
investigate
the root cause of the failure.
During this inspection,
the inspectors
found that the licensee
had originally
documented
the problem
and the initial corrective maintenance
performed in .a
quality evaluation
(gE) at the time the
MOV was returned to service.
After
technical
review of the
gE by the plant review board
(PRB), the licensee
had
determined that additional electrical
troubleshooting
was warranted
to rule
out potential
causes
other than the torque switch.
NCR No.
DC2-93-EM-N037
was written for the additional troubleshooting.
The licensee
had initiated
the
NCR within 5 days after return to service
in accordance
with their gE
review procedure.
The
MOV was
removed
from service for additional electrical
troubleshooting,
which identified
an intermittent high contact resistance
in
an auxiliary contact
in the motor starter.
The licensee
replaced
the
auxiliary contacts
and performed diagnostic testing prior to returning the
valve to service..
In addition,
increased
frequency surveillance testing with
diagnostic
instrumentation
was conducted for approximately
3 months without
any repeat of the problem.
. The inspectors
found the licensee's
corrective actions
'under
No. DC-2-93-EM-N037 to be adequate.
The inspectors
found that the licensee's
initial maintenance
troubleshooting activities
had
been
narrow and prejudiced
toward the preconceived
cause
at the time.
The troubleshooting
had not been
expanded
in
a controlled
and systematic
fashion
when the original diagnosis
failed to be confirmed.
The inspectors
observed
that the
MOV appeared
to have
been initially returned to service
on the basis that the maintenance activity
of replacing the torque switch
had
been
performed adequately
rather than
on
the basis that the root cause of the failure had
been corrected.
The
inspectors
considered this to be
an example of a weakness
in the control
and
technical
depth of troubleshooting activities
and the acceptance
criteria for
returning
MOVs to service following troubleshooting.
Another example of this
weakness
is discussed
in Section
3.
Based
on the adequacy of the root cause
evaluation
conducted
under
No. DC-2-93-EM-N037, which the licensee
had initiated
as
a result of their
gE
evaluation
and the licensee's
commitment
as discussed
in Section
3, this item
is closed.
1.4
0 en
Ins ection Followu
Item 275 323 9319-02:
Pressure
Lockin
During previous
NRC inspections,
the inspectors
had found that corrective
action's for six MOVs were planned to be
implemented at the next available
outage.
The licensee
had determined that the six MOVs were susceptible
to
pressure
locking.
During this inspection,
the inspectors
found that the licensee
had modified
MOV 1-8703 during the
1R6 outage=by drilling a hole in the valve disk to
preclude
pressure
locking.
The remaining five MOVs were Unit 2
MOVs which
were planned to be modified in an identical
manner during the
2R6 outage
in
. September
1994.
i
This item will remain
open pending
completi'on of the modifications of the
Unit 2
MOVs in September
1994.
1.5
Closed
Iris ection Followu
Item 275 323 9319-03:
Evaluation of
Dia nostic Si nature for MOV 2-FCV-37
During
a previous
NRC inspection,
the inspectors
had questioned
the accuracy
of the licensee's
diagnostic thrust signature
analysis for MOV 2-FCV-37.
The
licensee
had performed the analysis of thrust data obtained
from differential
pressure
testing to demonstrate
the design basis capability of the
MOV under
their
GL 89-10 program.
Specifically, the inspectors
had questioned
an apparent
non-conservative
'dentification of the zero-thrust point on the diagnostic thrust signature.
Proper, identification of the zero-thrust
event
was important in the analysis
of the thrust signature
in order to accurately
determine
the magnitude of
thrust during key events
in the operating cycle of the
MOV.
Typically the
zero-thrust point was identified using the closing thrust trace.
If not
evident in the closing trace,
an alternate identification was possible
using
the opening thrust trace.
The inspectors
had reviewed the thrust signature
for NOV 2-FCV-37
and noted that the licensee
had marked the zero point using
the opening thrust trace.
However,
the inspectors
had noted
a prominent
zero-thrust
plateau
evident in the closing thrust trace which appeared
to be
a
more conservative
determination of the zero-thrust point.
This thrust plateau
was offset 1600 pounds-force (lbf) in the compressive direction from the zero
point selected
by the licensee.
Using the more conservative
apparent
zero
point from the closing trace,
the inspectors
had been
concerned
that the
may not have demonstrated
adequate
capability because
the resulting
indicated
thrust at torque switch trip was approximately
200 lbf less
than the
calculated
minimum thrust required to close the valve under design
basis
conditions.
In response
to the inspectors'oncern,
the licensee
contacted
Liberty
Technologies,
the supplier of the
VOTES diagnostic
system.
The licensee
stated that Liberty Technologies
concurred with their original selection of
the zero-thrust point.
During this inspection,
the inspectors
examined
the thrust trace for the
identical Unit
1
NOV I-FCV-37.
The inspectors
noted the
same general
shape of
a thrust plateau
evident in the closing trace
as
had
been
noted in
HOV 2-FCV-37.
The licensee
had also
used the opening trace for HOV I-FCV-37
to identify the zero-thrust point as they had done in the case of
HOV 2-FCV-37.
The inspectors
further reviewed both valve signatures for other confirmatory
indications of the appropriate
zero point.
The inspectors
noted that use of
the more conservative
zero-thrust point from the closing trace would result in
indicated running loads inconsistent with the effects of stem ejection forces.
Further,
the inspectors
found that use of the licensee's
zero point was
consistent
with the assumption
that stem ejection loads
exceeded
packing
loads.
Based
on these
observations
and the vendor concurrence,
the inspectors
determined that the licensee's
selection of the zero point was acceptably
justified.
The inspectors
noted that the unexplained thrust plateaus
in the closing
signatures
of both valves
were not documented
as anomalies
for. the purposes
of
evaluation
and trending.
The licensee
stated that they had not originally
considered
the trace characteristics
to be anomalies.
The licensee
stated
that they would review their documentation
of unexpectec
trace characteristics
to assure
that their signature
analysis
was adequately justified and
documented,
and that anomalous
conditions
were appropriately identified.
The
inspectors
did not consider the anomalous
thrust plateaus
in the closing
traces for HOV I/2-FCV-37 to constitute
an immediate operability concern.
0
h
1.6
Closed
Ins ection Followu
Item 275 323 9319-04:
Use of Valve
S ecific
Values in Calculations
4
During
a previous
NRC inspection,
the inspectors
had noted that the licensee's
program did not provide for feedback of test results to validate the value for
rate-of-loading
(ROL) assumed
in the licensee's
HOV setpoint calculations.
ROL (also
known as load-sensitive
behavior)
was
a dynamic operating
characteristic
displayed
by some
HOVs in which the torque switch actuates
prematurely during closure
under differential pressure
conditions.
A ROL
allowance
was incorporated
in the licensee's
setpoint calculations to account
for this difference in HOV performance
under differential pressure
conditions
compared
to the static conditions
under which the torque switch was adjusted.
ROL was defined
as the difference in thrust at torque
swit'ch trip measured
under static testing
and dyna'mic testing,
expressed
as
a percentage
of static
thrust.
Licensee
Procedure
ICE-12,
"ILC Engineering
Procedure for Preparation
of
Hotor-Operated
Valve Sizing
and Switch Setpoint Calculations,"
Revision 4,
as
umed
a value of 15 percent for ROL in establishing
an allowable setpoint
range for the torque switch setting.
While the licensee
found this value to
be conservative
in most cases,
some
HOVs displayed
ROL in excess
15 percent
during testing.
For these
MOVs, the minimum thrust setpoint would not have
been revised to account for actual
valve performance
which exceeded
assumptions.
Even though testing demonstrated
the adequacy of the existing torque switch
setting,
the inspectors
had
been
concerned
that torque switch settings
could
be lowered after differential pressure
testing to the minimum thrust setpoint.
Without appropriate
revision of the minimum thrust setpoint for higher than
expected
ROL, the lower range of allowable torque switch settings
would not be
adequate
to assure
design basis capability.
During the original inspection,
the licensee
was in the process
of revising
Procedure
ICE-12 to incorporate test results.
The licensee
committed to
review existing test data to justify their use of a
15 per'cent
margin for ROL
and ensure that current torque switch settings
based
on that assumption
were
valid.
During this inspection,
the inspectors
found that the licensee
had completed
the review of existing test data
and
had verified that all existing torque
switch settings
were valid.
The thrust setpoint
windows for several
HOVs had
been revised to account for ROL greater
than
15 percent.
Until the revised
setpoints
were fully implemented,
the licensee
had instituted administrative
restrictions to prevent
any adjustment of the torque switch setting of
HOVs without engineering
approval.
The inspectors
reviewed
Procedure
ICE-12, Revision 7,
and verified that the
licensee
had incorporated
valve specific
ROL in the thrust setpoint
window if
the observed
ROL exceeded
the
assumed
value of 15 percent.
In addition,
the
minimum required setpoint
included
a nominal margin of 20 percent to account
for torque switch repeatability
and
stem lubrication degradation.
To justify their generic
assumption of 15 percent for ROL, the licensee
performed
a review of the existing test data.
The licensee
documented this
review in a letter dated
September
3,
1993.
The licensee's
review indicated
that all but one gate valve had
an
ROL less than.15 percent.
The inspectors
reviewed the licensee's justification and found it to be adequate for gate
valves.
However, the licensee
determined that
6 of 18 globe valves displayed
a
ROL in
excess of 15 percent.
The licensee
had
used yoke-mounted
sensors
during
dynamic testing of these valves.
The licensee
performed
seven repeat
diagnostic tests of globe valves using stem-mounted
sensors
and found the
value: to differ significan'
from those determined
using yoke-mounted
sensors.
The
ROL determined
from these
repeat tests
ranged
from -4.75 percent
to 7.54 percent.
In one case for NOV 1-8105,
ROL wa's determined
to be
51 percent
using the yoke-mounted
sensor
but only 4.84 percent
using the
stom-mounted
sensor.
Likewise, the
ROL for HOV 1-8974B appeared
to change
from 55 to 5.37 percent.
The lower
ROL found was attributed to the higher
indicated thrust at torque switch trip measured
under dynamic conditions using
stem-mounted
sensors.
The licensee
had preliminarily determined that the high
ROL observed with the yoke-mounted
sensors
was erroneous,
resulting from an
interaction
between
the sensor
and the anti-rotation device installed
on these
valves.
The licensee
concluded that the actual
load sensitive
behavior of the
globe valves determined
from stem-mounted
sensors
was consistent with the
15 percent
assumption.
The inspectors
reviewed this material
and concluded
that the licensee
was justified in maintaining
a generic
15 percent
assumption for globe valves.
According to the licensee,
the anomalous
behavior of the
VOTES sensors
would
be pursued with the vendor (Liberty Technologies)
to determine
the extent of
the generic
problem in applications
involving similar anti-rotation devices.
The inspectors
concluded that the licensee
had satisfactorily addressed
the
concerns related to
ROL.
1.7
Closed
Ins ection Followu
Item 275 323 9319-05:
Review of Previous
Test Data
During
a previous
NRC inspection,
the inspectors
had noted that licensee
Procedure
ICE-12, Figure 8, "Rising Stem
HOV Test Evaluation Sheet,"
Revision 3, did not extrapolate
the peak opening thrust after pullout to
assure
adequate
NOV capability under design basis conditions.
A later
Revision
4 to the evaluation
sheet
had incorporated this extrapolation.
The
licensee
committed to review previous test data using the Revision
4 criteria
by October 30,
1993.
The inspectors verified that extrapolation of peak opening thrust was required
through the current Revision
7 to Procedure
ICE-12.
The licensee
had
-10-
completed
the review. of previous test data
and
had not identified any
resulting operability concerns.
The inspectors
concluded that the licensee
had acceptably
addressed
this issue.
1.8
Closed
Ins ection Followu
Item 275 323 9319-06:
Review of Limit
Seated
MOV Tor ue Switch Settin
During
a previous
NRC inspection,
the inspectors
were concerned that the
licensee
did not include
a margin for ROL for limit-closed HOVs in their
determination of the minimum required thrust setpoint.
The licensee's
limit-closed motor control logic used the closed limit switch for closure
control.
The torque switch was set at the maximum achievable setting
permitted
by the limiter plate to act
as
a backup for the limit switch.
Since
the torque switch was in series with the limit switch and enabled for the
entire closing stroke,
the insnectors
were concerned
that the torque switch
setting
may not
be adequate
to assure that it did not inadvertently actuate
under design basis conditions
due to
ROL and prematurely interrupt the valve
closure.
During this inspection
the licensee clarified that
a l5 percent
margin was
incorporated
in the minimum required setpoint for ROL in limit-seated
HOVs but
not for bypass-seated
HOVs.
The licensee
defined bypass-seating
as
a
variation of limit-seating motor control logic in which the torque switch was
bypassed
during closure until the point of valve seat contact.
The inspectors
reviewed
Procedure
ICE-12, Revision 7,
and verified the
licensee's
clarification of the torque switch setpoint determination for MOVs
using limit-seated motor control logic.
No additional
concerns
were
identified.
2
QUALITY ASSURANCE INVOLVEMENT (40500)
The inspectors
reviewed the following assessments
of HOV related activities
performed
by the site quality assurance
organization:
Electrical Maintenance
Surveillance
Report
No. SQA-94-0006,
Maintenance Quality Audit No. 94-0101,
and
GL 89-10 Management
Prerogative Audit No. 94-0161.
Although Audit
No. 94-0161
was still in progress,
the inspectors
discussed
the extent. and
preliminary findings of this effort with the audit team leader.
The inspectors
found that Audit No. 94-0161
was
a self-assessment
of the
licensee's
GL 89-10 program.
The audit had
been started
in February
1994
and
included approximately
5 man-months of inspection.
The audit reviewed the
licensee's
program for preparation of MOV sizing
and switch setpoint
calculations,
specifically focusing
on the adequacy of design basis reviews,
MOV calculations,
as-left torque switch settings,.flow test evaluations,
and
corrective actions
taken in response
to the previous
NRC
MOV inspection.
Each
of the auditors
attended
a 5-day team training course
provided
by a contractor
(General
Physics) prior to the start of the audit.
0
I
The inspectors
reviewed several
documents detailing the methodology
used in
assessing
MOV sizing
and torque switch settings
and design basis calculations,
including the determination of maximum expected differential pressure,
maximum
expected
flow rate,
and minimum anticipated
voltage.
Based
on this review and
discussions
with the audit team leader,
the inspectors
concluded that the
self-assessment
was
a comprehensive,
technically-oriented effort.
At the exit
the inspectors
encouraged
the licensee
to be responsive
to the quality
assurance
audit findings to assure
adequate
resolution of any deficiencies
within the committed program schedule.
3
-FAILURE OF
NOV 2-FCV-37
During this inspection,
the inspectors
reviewed
an operational
failure of
HOV 2-FCV-37, which occurred
in January
1993, prior to
GL 89-10 differential
pressure
testing in March 1993.
During
a routine
pump surveillance test
on January
31,
1993,
HOV 2-FCV-37,
a
turbine-driven auxiliary feedwater
steam supply isolation valve, failed to
close.
The torque switch tripped prematurely
at the start of the closing
stroke.
Valve motion stopped
at
a valve position approximately
95 percent
fully open
when the limit switch enabled
the torque switch in the motor
control circuit.
During initial troubleshooting activities performed
by the licensee's
maintenance
personnel,
the valve was stroked manually
and
was noted to be
difficult to operate
over
a portion of the closing stroke.
The valve was then
stroked electrically, with the observation that the torque switch was
"bouncing."
On the third electrical 'stroke,
the failure was duplicated with
the valve stopping after closing only five percent of the full stroke.
Inspection of the
stem revealed
marginal lubrication.
After the
stem
had
been
relubricated,
four successful
electrical
strokes
were conducted,
and torque
switch bouncing
was
no longer evident.
The immediate root cause
was
determined
to be
a sticking valve stem resulting
from degraded
lubrication.
The valve was returned to service
on February
1,
1993.
Diagnostic testing
was
not performed at this time.
On February
5,
1993,
gE 90010397
was initiated to address
the quality concerns
related to the event.
Over the next month,
VOTES diagnostic tests
were
conducted
under static conditions with satisfactory results.
'During subsequent
actuator
overhaul
on March 12,
1993,
in preparation for
GL 89-10 testing,
an internal
inspection
revealed significant corrosion of the
upper thrust bearing
and extensive lubrication contamination
and degradation.
The corrosion
was apparently
caused
by moisture intrusion that resulted
from
an earlier improper maintenance
activity in April 1990.
At that time,
an
environmental
seal'(quad
ring) was not installed
and another
was replaced with
a quad ring of improper fit.
NCR DC2-93-EH-N014
was issued.
The licensee
determined that the previous failure in January
1993
was
due to excessive
internal operating
torque in the actuator
which reduced
the output thrust
capability.
The licensee
determined
the root cause to be
an inadequate
0
0
-12-
maintenance
procedure that did not have sufficient detail to ensure that the
seals
were properly .rei'nstalled following operator disassembly.
Procedure
HP E-53. 10M, "Limitorque SHB-00 and SB-'0 Valve Operator
Maintenance,"
was revised to contain detailed
steps
and
a composite
assembly
drawing for reassembly
of the operators.
The inspectors
reviewed
NCR DC2-93-EM-N014
and discussed
the event with the
licensee.
The inspectors
found that once the actual
cause of the valve
failure was determined
during overhaul,
the licensee
took effective correction
actions.
However,
the inspectors
were concerned
that the immediate actions
taken
by the licensee
before the valve was returned to service in February
1993 appeared
to treat only the
symptoms of the problem
and not the root
cause.
Based
on the information available at the time, the inspectors
concluded that the licensee's
troubleshooting
appeared
to focus more
on
correcting observable
deficiencies rather than identifying the cause of the
failure in a controlled
and systematic
manner.
Without an obvious
component
failure, proper diagnosis of operational
HOV failures requires
consideration
of design basis capability.
However,
in this case,
the determination of
design
basis capability using diagnostic
i;est'...y
~:as not used prior to
returning the
MOV to service.
Although the lubrication of the stem appeared
marginal, this condition alone
could not have
caused
the torque switch to actuate
during the running portion
of the strok'e.
The licensee later determined that the effective stem friction
coefficient would have
had to be 0.7 for the
stem lubrication to have
caused
the torque switch trip.
Poor stem lubrication typically r'esults
in a friction
coefficient of 0.2 or slightly higher.
The inspectors
were concerned
that the
licensee failed. to verify that the immediate
proposed
root cause
was
consistent
with the available information.
In this case,
the inspectors
concluded that the mismatch
between
the failure mode
and the postulated
cause
should
have alerted the licensee that the actual root cause of the event
had
not been identified.
As
a result of superficial 'troubleshooting,.
MOV 2-FCV-37 was returned to
service in
a degraded
condition.
The inspectors
identified this
and
a similar
incident with HOV 1-LCV-109 (paragraph
1.3)
as two examples of a weakness
in
the licensee's
program for evaluating
nonconforming conditions prior to
returning equipment to service.
Within the review of the failures of MOVs 1-LCV-109 and 2-FCV-37, the
inspectors
observed
a tendency
on the part of the licensee
to curtail
troubleshooting efforts once the first discrepancy
was identified.
At this
point, the licensee
did not search for the presence
of other problems
nor
verify that the identified discrepancy,
by itself, could have
caused
the
observed failure.
The inspectors
were concerned
that despite
the licensee's
previous experience
with the failures of MOVs 2-FCV-37 and
2-LCV 109, there
had
been
no
programmatic'hanges
to incorporate
lessons
learned
in the control of
0
'
-13-
troubleshooting.
The licensee
acknowledged that the failures'ould probably
be handled
in
a similar manner if they were to occur today.
At the exit meeting,
the licensee
acknowledged
the inspectors'oncern
and
committed to procedural
revisions to assure
that design basis capability is
maintained
as
a condition for returning
HOVs to service,
The inspectors
found
the licensee's
commitment to be adequate.
1
0
ATTACHMENT
1
PERSONS
CONTACTED
1.1
L'icensee
Personnel
- H. Hurgess,
Director, Technical
and Support Services
- W. Crockett,
Manager,
Technical
and Support Services
H. Frauenheim,
Engineer,
Electrical Maintenance
- R. Goel,
Engineer,
Nuclear Engineering
Services
- C. Groff, Director, Technical
Services
- K. Hubbard,
Engineer,
Regulatory Compliance
- C. Lewis, Engineer,
Nuclear Quality Services / Site Quality Assurance
- D. Hiklush, Manager,
Operation Services / Acting Plant Manager
- R. Powers,
Manager,
Nuclear Quality Services
L. Pulley,
Engineer,
Nuclear Engineering
Services
- J. Shoulders,
Director, Nuclear Engineering
Services
- D. Taggart, Director, Nuclear Quality Services / Site Quality Assurance
- A. Toy, Engineer,
Predictive Maintenance
- M. Williamson, Engineer,
Electrical Maintenance
1.2
NRC Personnel
- H. Hiller, Senior Resident
Inspector
- Denotes those attending
the exit meeting
In addition to the personnel
listed above,
the inspectors
contacted
other
personnel
during this inspection.
2
EXIT MEETING
An exit meeting
was conducted
on June
10,
1994.
During this meeting,
the
inspectors
reviewed 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.
1
~