ML16342C851
| ML16342C851 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 02/22/1995 |
| From: | Westerman T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML16342C850 | List: |
| References | |
| 50-275-95-01, 50-275-95-1, 50-323-95-01, 50-323-95-1, NUDOCS 9503070084 | |
| Download: ML16342C851 (38) | |
See also: IR 05000275/1995001
Text
ENCLOSURE
U.S.
NUCLEAR REGULATORY COHHISSION
REGION IV
Inspection Report:
50-275/95-01
50-323/95-01
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
Power Plant. Units
1 and
2
Inspection At:
Diablo Canyon site,
San Luis Obispo County, California
Inspection
Conducted:
January
13-17 and 23-27,
1995
Inspectors:
C. Hyers,
Reactor Inspector,
Engineering
Branch
Division of Reactor
Safety
H.
Runyan
~ Reactor
Inspector,
Engineering
Branch
Division of Reactor
Safety
Accompanying
Personnel:
H. Rathbun, Office of Nuclear
Reactor Regulation
Approved:
es
erman,
ie
.
ngineer>ng
nc
Division of Reactor Safety
Ins ection
Summar
Areas
Ins ected
Units
1 and
2
Special,
announced
inspection of'afety-
related motor-operated
valve testing
and surveillance
and followup of
engineering
issues.
Results
Units
1 and
2
~
The inspectors verified completion of the licensee's
commitments to
Generic Letter 89-10, contingent
on submittal within 60 days of the date
of this report,
a letter to the
NRC documenting additional justification
related to periodic verification and the capability of untested
motor-
operated
valves (Section
1. 1).
9503070084
950302
ADQCK 05000275
8
-2-
i
~
The inspectors
concluded that the licensee
had adequately
established
the design basis capability of motor-operated
valves that had not been
tested at or
near
design basis conditions.
Some additional effort was
needed to confirm generic
assumptions
used in support of untested
motor-
operated
valves
and to address
certain valves using the Electric Power
Research
Institute valve factor evaluation process
(Section 1.5).
~
The licensee's
motor-operated
valve program included consideration of
valve mispositioning (Section 1.2).
~
The licensee's
analysis
and corrective actions taken for potential
pressure
locking and thermal binding concerns
were thorough
and timely.
Modifications of susceptible
valves
had been completed.
(Section 1.3).
The inspectors
found that the licensee
had incorporated all appropriate
vendor information regarding diagnostic
system
measurement
accuracy.
Additional sources of error
had been identified (Section 1.4).
Grouping of valves for comparison of test results
was utilized to
justify the capability of untested
valves.
Additional information was
requested
regarding implementation of the grouping guidelines of Generic Letter 89-10.
Supplement
6 (Section 1.5).
Limited differential pressure testing
was included
as part of the
licensee's
plans for periodic verification.
The licensee
was requested
to provide additional information regarding several
issues
related to
periodic verification (Section 1.6).
Detailed maintenance fitup of valve internals
and performance trending
had been
implemented
and were considered
strengths
in the licensee's
ongoing program (Section 1.6.2 and 1.7.2).
Continuing strong quality assur ance oversight of the motor-operated
valve program was evident.
The technical
depth of the audits
was
considered
a strength.
Some recent audit findings had not been resolved
at the time of the inspection
(Section 1.8).
Summar
of Ins ection Findin s:
Inspection
Followup Item 275;323/9319-02
remains
open pending issuance
of a planned
NRC generic communication
on the issue of pressure
locking
and thermal binding (Section 1.9. 1).
Inspection
Followup Item 275;323/9139-02
was reviewed but left open
(Section 1.9.2).
Additional information was requested
to clarify ongoing program
commitments
as identified in Attachment 2.
Attachments:
-3-
~
Attachment
1
- Persons
Contacted
and Exit Meeting
~
Attachment
2 - Additional Information Requested
to Clarify Ongoing
Program
Commitments
'
4
DETAILS
1
GENERIC LETTER 89-10,
"SAFETY-RELATED MOTOR-OPERATED VALVE TESTING AND
SURVEILLANCE"
(2515/109)
On June
28 '989,
the
NRC issued
Generic Letter 89-10, which requested
licensees
and construction permit holders to establish
a program to ensure
that switch settings for safety-related
motor-operated
valves were selected.
set.
and maintained properly.
Subsequently,
six supplements
to the generic
letter have
been issued.
NRC inspections of licensee
actions
implementing
commitments to Generic Letter 89-10 and its supplements
have
been conducted
based
on guidance
provided in Temporary Instruction 2515/109
~ "Inspection
Requirements
for Generic Letter 89-10, Safety-Related
Motor-Operated
Valve
Testing
and Surveillance,"
Revision
1.
Temporary Instruction 2515/109
was
divided into two parts:
Part
1,
"Program Review;" and,
Part 2, "Verification
of Program Implementation."
The Temporary Instruction 2515/109,
Part
1,
program review inspection
was conducted at Diablo Canyon during October
1991
and was documented
in NRC Inspection Report 50-275/91-39;
50-323/91-39.
The
Temporary Instruction 2515/109,
Part 2. implementation
review inspection
was
conducted at Diablo Canyon during July 1993 and was documented
in NRC
Inspection Report 50-275/93-19;
50-323/93-19.
A followup Part
2 inspection
was conducted during June
1994 and was documented in NRC Inspection
Report 50-275/94-17;
50-323/94-17.
The licensee notified the
NRC in a letter
dated
November 28,
1994. that Diablo Canyon's
commitments to Generic Letter 89-10 had been completed.
The principal focus of this inspection
was to evaluate the licensee's
process
for qualifying the design basis capability of each of the motor-operated
valves in the Generic Letter 89-10 program.
Though most valves were tested
under conditions applying
a high percentage of the design basis differential
pressure,
some were tested only under static or low differential pressure
conditions.
As discussed
in Generic Letter 89-10, these
were considered to be
valves for which
a two-stage
approach
should
be utilized.
The first stage
was
.
to set
up the valve using the best available
information.
The second
stage
required
a more precise
methodology.
which could include comparison to a
similar valve, prototype testing,
use of the Electric Power Research
Institute (EPRI) testing results,
or other methods.
1. 1
Summar
Status of Generic Letter 89-10 Motor-0 crated
Valves
At the time of the inspection,
155 motor-operated
valves were included in the
Generic Letter 89-10 program.
Approximately 68 percent of the motor-operated
valves
had been tested
under differential pressure
conditions.
The remainder
of the motor-operated
valves were tested only under static conditions.
-5-
Based
on the documents
reviewed during this inspection
and discussions
with
cognizant licensee
personnel,
the inspectors verified completion of the
licensee's
commitments to Generic Letter 89-10. contingent
on submittal of
additional justification and information related to several
areas of ongoing
program activities
as described
in this report.
and subsequent
NRC review.
1. 1. 1
Generic Implications Report
The inspectors
reviewed Procedure
ICE-12,
"18C Engineering
Procedure for
Preparation of Rotor Operated
Valve Sizing and Switch Setpoint Calculations,"
Revision 10.
This procedure
was developed
and implemented to ensure that
motor-operated
valves were properly sized
and setup prior to differential
pressure testing.
In addition,
Procedure
ICE-12 provides guidance
on
evaluating the results of differential pressure testing.
As required
by Procedure
ICE-12. the licensee
developed
a "Report on Generic
Implications of 89-10 Testing" following completion of the differential
'ressure
testing
recommendations
The inspectors
reviewed the licensee's
report dated
December
9,
1994.
This document
summarized the methods
and assumptions
used to qualify each valve in the
program as being capable of performing its safety function under
design basis
conditions.
The report discussed
the results of differential pressure testing
for each valve type and included
an evaluation of assumed
valve factors.
rate
of loading.
and stem friction coefficient.
Test valve factors
and rates of
loading were compared to design assumptions.
In cases
where test results
indicated marginal motor-operated
valve capacity,
the licensee
provided an
engineering evaluation to support design basis capability.
The. inspectors
reviewed the available margin identified by the licensee
for
each of the motor-operated
valves at the conclusion of their program.
Generally,
the inspectors
found that the licensee
had adequately
demonstrated
the existing design basis capability for each of the valves consistent with
their
program plan commitments.
1. 1.2
Actions Taken for Valve Factor Exceeding Assumptions
Valve factor is
a measure of the condition of the valve internals affecting
valve operation.
Valve factor is defined
as the ratio of the actuator thrust
to the differential pressure
force.
The actuator
thrust setpoint
range
was
established
using the design valve factor.
The existing margin based
on as-left switch settings
was termed the "ICE-12
margin," after Procedure
ICE-12, which defines the motor-operated
valve
testing program.
A positive
ICE-12 margin was required for motor-operated
valve operability.
The licensee did not routinely revise the design valve factor for
a motor-
oper ated valve group to account for measured
valve factors that exceeded
the
original assumption.
This feedback of test information was
a
common method to
validate design basis capability of the entire allowable range of thrust
-6-
settings.
Rather,
long-term maintenance of acceptable
valve settings
was
controlled by using what was termed the "effective margin."
The effective
margin was based
on the lowest permissible thrust setting within the setpoint
window.
The fact that
some motor -operated
valves
had positive effective
margins
even though the measured
valve factor exceeded
the design valve factor
was attributable to conservatisms
in estimating
pa'cking
and ejection loads
and
rates of loading.
Some motor -operated
valves,
however,
had negative effective
margins
(meaning that. if the valve were set at the low end of the setpoint
window, it may not have adequate capability).
The licensee
had implemented
administrative controls to prevent changing the switch setting for these
motor-operated
valves until the setpoint window has
been appropriately
adjusted.
The inspectors
found the licensee's
actions to be adequate.
1. 1.3
Rate-of-Loading Assumption Validated
The generic implications report summarized
an evaluation of the assumptions
made for rates-of-loading.
The rate-of-loading
assumption
was important for
the evaluation of static-only motor -operated
valves
and for other valves for
which
a accurate
rate-of-loading
was not obtained.
Rate-of-loading is the
percentage
change in the thrust at control switch trip from the static to the
dynamic test,
expressed
as
a positive number when the static control switch
trip thrust is greater than the dynamic thrust.
The licensee
had established
a margin of 15 percent to account for rate-of-loading.
Six gate valves
and
four globe valves
had rates-of-loading calculated to be greater
than
15
percent,
with the highest gate valve being 24.65 percent
and the highest globe
valve being 30.30 percent.
The average rate-of-loading for gate valves
was
a
ositive 2.39 percent.
For globe valves, the average
was
a positive
.27 percent.
The licensee
evaluated
each instance
where the measured
rate-
of-loading exceeded
the design value.
Two patterns
were noted:
one in which
high rates of loading were associated
with low test differential pressures.
and one in which high rates of loading were associated
with low measured
valve
factors.
The inspectors
reviewed this information and concluded that the use
of a design value of 15 percent rate-of-loading
was justified.
1. 1.4
Limited Validation of Stem Lubrication Assumption
The generic implications report also addressed
the consistency
between the
measured
and
assumed
values for stem friction coefficient.
The design
assumption for stem friction coefficient was 0.2.
Since the licensee's
diagnostic system did not regularly measure
only five tests
performed
with a Teledyne
stem mounted thrust and torque strain gage gave results
from
which a stem frictIon coefficient could be calculated.
The highest
measured
values for stem friction coefficient were 0. 12 in the closing direction and
0. 18 in the opening direction.
Based
on these tests
and industry experience,
the inspectors
considered
the design
assumption of 0.2 to be acceptable.
0
-7-
The licensee identified that for one motor-operated
valve.
1-9001B,
a stem
friction coefficient assumption of 0. 15 was used in lieu of 0.2 because
the
use of 0.2 would have yielded
a negative capability margin in the opening
evaluation.
This motor-operated
valve was
on
a special
3-month inspection
and
lubrication schedule.
The inspectors
noted that the licensee
also included
a
7 percent margin for
stem lubrication degradation
in thei r determination of motor-operated
valve
capability.
The inspector
found the licensee's
actions to be adequate.
1. 1.5
Inadvertent
Use of Neolube
on Valve Internals
The licensee
observed
unexpectedly
low pullout thrust requirements
during
initial static testing of several
motor-operated
valves immediately following
valve internal maintenance.
The licensee
discovered that maintenance
personnel
had applied the graphite
base lubricant "Neolube" to the internal
guides.
disks
and seats of 14 Generic Letter 89-10 motor -operated
valves
during valve reassembly prior to differential pressure testing.
The
maintenance
practice
had been
implemented
by procedure to assist in fitup
reassembly of the valve.
Engineering
was unaware of this practice at the time
of the differential pressure testing
and
became
concerned that this practice
may have affected the validity of differential pressure test results.
The
licensee
conducted
various static tests to determine the effects of neolube
on
valve performance.
The licensee's
testing
found that motor-operated
valve
performance
was not affected
by the use of neolube.
In addition. the licensee
concluded that neolube
was rapidly removed
by water
from the valve internals
after
a few valve strokes.
The licensee
considered that all valves
had been
stroked at least ten times prior to differential pressure testing.
The
licensee
concluded that valve performance
during differential pressure testing
was not affected
by the use of neolube.
~
The inspectors
found that the licensee's
static testing
and evaluation
focused
on disk pullout performance
and did not address
valve factor performance
under
differential pressure
conditions.
The inspectors
found no test-basis
for the
licensee's
conclusion regarding the effect of neolube
on valve factor
performance.
The inspectors
were concerned that lubrication of the valve
internals
may have resulted in nonconservative
determinations
of actual
valve
factors
from testing
soon after
reassembly.
The inspectors
noted that the
licensee
had generically
assumed that actual
valve factor did not change.
and
no margin had been specifically identified to accommodate
valve factor
degradation.
The inspectors
reviewed the test results for the
14 affected motor-operated
valves
and found that all valve factors were consistent with the results of
other
valves in thei r group.
None of the valves exhibited
a low valve factor.
0
-8-
In response
to the inspectors'oncern,
the licensee identified that one of
the neolubed valves
had been selected for differential pressure testing during
periodic verification.
According to the licensee,
changes
in valve factor
performance attributable to the previous
use of neolube will be evaluated at
that time to confi rm their opinion.
The licensee is being requested
to
provide additional detai
1 of thei r use of periodic testing to validate thei r
assumptions
regarding the use of neolube in response to this inspection
report.
1,2
~Ill
ltl
The inspectors
reviewed the licensee's
motor-operated
valve program to
determine the licensee's
current treatment of valves subject to
mispositioning.
As recommended
in Generic Letter 89-10. the scope of licensee
motor-operated
valve programs
was to include motor-operated
valves which were
not prevented
from inadvertent mispositioning from the control
room.
The inspectors
found that the licensee's
program continued to include motor-
operated
valves which were not prevented
from inadvertent mispositioning.
The
licensee
had not change'd their consideration of mispositioning in establishing
the scope of their
program.
The inspectors
found the licensee's
position regarding the consideration of
mispositioning within their Generic Letter 89-10 program to be adequate.
1.3
Pressure
Lockin
and Thermal Bindin
During the first Part
2 inspection,
the inspectors
reviewed the licensee's
program to address
pressure
locking and thermal binding, which was documented
in Nuclear Engineering
and Construction Ser'vices
memorandum to Nuclear
Operations
Supports
"Pressure
Locking of Gate Valves," dated October 6,
1992.
As of the date of this inspection,
no additional
program documentation
had
been
issued
on this subject.
The licensee
had identified 26 motor-operated
valves in each unit which met
their review screening criteria for ya1ves susceptible to pressure
locking and
thermal binding.
As
a result of their evaluation.
the licensee identified six
valves considered
susceptible for pressure
locking:
1/2-8703,
2-8801A/B,
and
2-8803A/B.
No valves were considered
susceptible to thermal binding.
The licensee
review recommended
three options
as corrective actions for each
of the valves:
(1) Drill a hole in the high pressure
side of the disk.
(2) Install
a bonnet leakoff line and block valve,
or
(3) Install
a bonnet relief line and discharge line.
During this inspection,
the inspectors
reviewed closed Action
Request
A0316042,
which identified that modifications to the six identified
valves
had been completed in Refueling Outages
1R6 and 2R6.
In addition to
-9-
these modifications, the licensee
stated that,
based
on information obtained
from industry sources.
modifications (drilled upstream valve disks) were
performed
on the six power-operated relief valve block valves,
1-8000A/B/C and
2-8000A/B/C.
The inspectors
concluded that the licensee
had taken sufficient steps to have
met its commitments associated
with pressure
locking and thermal binding for
Generic Letter 89-10 closure.
The inspection followup item (275;323/9319-02)
tracking this issue will remain open pending issuance
and subsequent
reviews
of a planned
NRC generic communication.
1.4
Actions in Res
onse to Generic Letter 89-10
Su
lement
5
The inspectors
reviewed the licensee's
actions in response
to Supplement
5 of
Supplement
5 had requested
information regarding the
diagnostic
systems
being used
by the licensee
during their Generic Letter 89-10 program.
The inspectors
found that the licensee
had responded to Supplement
5 and
identified that the
VOTES diagnostic test system
was used for diagnostic
testing during thei r Generic Letter 89-10 program.
The licensee
uses either
the
VOTES thrust transducer
(yoke-mounted)
or the Teledyne quick stem sensor
transducer
(stem mounted).
Several
unexpected
sources of error were
identified by the licensee during their testing,
as described
below.
The inspectors
found that the licensee
had incorporated all appropriate
vendor
information regarding
measurement
accuracy using the
VOTES system.
The inspectors
found the licensee's
actions to be adequate.
1.4. 1
Transition Zone with Split Stem Design
The licensee identified two incidents of gross calibration errors using the
yoke-mounted,
VOTES thrust sensor
on valves with split-stem designs.
The
discrepant
measurements
were observed
Globe
Valves 1-LCV-107 and 2-LCV-106.
The stem design for Valves
LCV-106 and
-107
consisted of a coupled
assembly of a threaded portion and
an unthreaded,
portion.
The coupling also served
as the torque restraint.
For purposes of
calibrating the
VOTES thrust sensor
mounted
on the valve yoke. the licensee
had treated this configuration
as
a solid stem geometry change
and applied
vendor recommendations
to avoid locating the "mini-c" calibrator in the
transition zone.
Later confirmatory thrust measurements
used
a strain gage
transducer
mounted directly on the unthreaded
portion of
the valve stem and analytically calibrated,
indicated only half of the
previously measured thrust.
After extensive investigation of the cause of the observed
discrepancy
including communications with the vendor (Liberty), the licensee
concluded
that transition
zone effects for split-stem valve designs
extended
much
farther beyond the range identified by the vendor,
introducing an error in the
e
0
'
i
-10-
output of the mini-c calibrator.
Due to space limitations in the area
between
the valve yoke and the stem,
the mini-c clamp was required to be placed in an
orientation that caused
a three-point contact with the stem.
The licensee
concluded that having the mini-c clamp in three-point contact with the stem
contributed to the observed diagnostic testing errors.
The licensee
concluded that
VOTES mini-c calibrator could not be used to
calibrate the
VOTES thrust transducer
or the quick stem sensor.
The licensee
had changed their practice to use
an analytic calibration of the quick stem
sensor
supplied
by the manufacturer for LCV-106 and -107.
No other valves
incorporated the split-stem design.
The inspectors
reviewed the licensee's
root cause evaluation
and found the
licensee's
actions to be adequate.
The licensee
determined that the observed
problem was not reportable
under
Part 21.
The licensee
planned to formally notify the
VOTES diagnostic
equipment
vendor (Liberty) of their observations
and conclusions.
1.4.2
Translating
Torque Restraint Effect
The inspectors
reviewed Action Request
A0354065 which documented that during
the recent
2R6 refueling outage,
the licensee
had performed
a hydrostatic test
of Residual
Heat
Removal Suction Isolation Valve 2-8702 to demonstrate
design
basis capability.
The diagnostic trace indicated
a unexpectedly
large thrust
required for disk pullout during opening of the valve.
The other three
identical valves in this valve group did not display this anomaly.
The
licensee's initial root cause investigation determined the cause of the
anomaly to be
a pressure
locking condition existing during the test.
Motor-Operated
Valve '2-8702 was one of two pressure
isolation valves off the
reactor coolant system to the suction of the residual
heat
removal
pumps.
These pressure
isolation valves were normally locked closed with electric
power
removed during operation.
The licensee identified a closing safety
function to isolate residual
heat
removal pipe break during cooldown operation
and opening safety function to initiate cooldown for Appendix
R shutdown.
The inspectors
observed that the test data for Motor-Operated
Valve 2-8702
indicated that only marginal capability existed.
The inspectors
reviewed the
thrust signature
analysis for this valve and found that the licensee
had
discounted the excessive
opening thrust requirement
as
a test
anomaly in their
analysis of the capability of the motor -operated
valve.
According to the
licensee,
the valve had experienced
pressure
locking due to a valve alignment
unique to the testing which was not possible during normal valve alignment.
The bonnet vent valve to prevent pressure
locking had been closed during
testing.
The inspectors
were concerned that the capability of the motor -operated
valve
could not be demonstrated
based
on the measured
data without discrediting the
anomalous portion of the opening thrust.
While the licensee's
hypothesis of
i
the occurrence of pressure
locking appeared
reasonable
to explain the observed
excessive
opening thrust requirement,
the inspectors
found that the licensee
had
no confirmatory evidence
(such
as measured
bonnet pressure)
to validate
the pressure
locking occurrence.
Furthermore,
the valve had not been
disassembled
for internal inspection.
The inspectors
did not consider the
quality of the test to be an adequate
demonstration of design basis capability
and requested
additional justification.
On further review, in response to the inspectors'oncern,
the licensee
identified
a calibration error affecting the accuracy of the opening thrust
measured
during the test of these valves.
The licensee
determined that the
error was due to a yoke-torque effect during initial valve opening which was
not present
during calibration of the yoke-mounted
VOTES thrust sensor.
The
licensee
had calibrated the yoke-mounted
VOTES thrust sensor
by backseating
the valve to produce
stem tension.
The design of the valve incorporated
a
stem-mounted
torque restraint which translated
along the stationary yoke
during valve stroking.
This design feature subjected
varying portions of the
depending
on the location of the torque restraint.
During
. valve stroking the torque restraint translated
past the fixed location of the
VOTES sensor=
on the yoke.
As
a result, the yoke in the location of the
VOTES
sensor
was not subjected to torque during backseating.
However during initial
disk pullout, the yoke in the location of the
VOTES sensor
was subject to
The l,icensee
concluded that
a backseating
technique could not be used
for open calibration purposes
for valves with the translating torque restraint
design.
The licensee
reanalyzed
the test data for Motor-Operated
Valve 2-8702 using
the close calibration data obtained
from the
same test
and determined that
additional margin was available to justify design basis capability based
on
the actual test data irrespective of the potential pressure
locking
occurrence.
The licensee
reviewed all other uses of the backseating
calibration technique
and found no other
valves which would be similarly
affected.
The licensee
committed to formally notify the
VOTES diagnostic
equipment
vendor (Liberty) of their observations
and conclusions.
The inspectors
found that the licensee
had appropriately selected
the
VOTES
sensor
location based
on vendor
recommended
practice.
The inspectors
found
that the vendor
information did not address
the use of backseat calibration
techniques.
The inspectors
reviewed the licensee's
evaluation
and found the
licensee's
actions to be adequate.
1.5
~Grou in
The inspectors
reviewed the licensee's justification for the design basis
capability of motor-operated
valves which were not tested
under dynamic
conditions.
'
-12-
Approximately 68 percent of the motor -operated
valves in the licensee's
program were tested
under differential pressure
conditions.
and valve factors
were determined
based
on results of the individual valve tests'or
motor-
operated
valves which were not considered
practicable
or meaningful to test
under differential pressure
conditions,
the licensee
determined
design basis
capability
by analysis with an assumed
valve factor for each motor-operated
valve.
The inspectors
noted that the assumed
valve factors in the licensee's
capability analysis
were not validated
by comparison with the test results of
similar valves,
nor were they validated
by industry testing of identical
valves.
Rather,
the licensee
selected
valve factors which were considered to
be adequately
conservative
based
on engineering
judgement
and the best generic
test data available at the time.
Although not defined within their program
as grouping, the licensee
used
a
grouping methodology to validate assumptions
made for valve factors.
Nineteen
valve groups were established,
generally defined by
a specific manufacturer
.
size.
and pressure
class.
Some groups contained
valves with more than one
size or pressure
class.
The inspectors
considered
the groups to represent
an
acceptable
division of valves for the purpose of evaluating valve factors.
The licensee utilized groups of motor -operated
valves for the purpose of
establishing
the best available data for evaluating similar performance.
The inspectors
were specifically concerned with eight untested
isolation valves,
1/2-FCV-438,-439,-440.-441.
The licensee identified zero
,existing margin for these
Anchor Darling 16-inch flex wedge gate valves with
an assumed
0.6 valve factor.
The inspectors
considered that the use of 0.6
for valve factor was adequate
as the best available data for Stage
I
qualification of the valve.
But the inspectors
were concerned that the
generic valve factor of 0.6 may not be sufficiently conservative to bound or
have
a high statistical
confidence in representing
expected
valve performance.
Industry test results of'ome similar valves
have
shown valve factors
as high
as 0.8.
Although
a 0.6 valve factor has recently been considered
generally
conservative,
several of the licensee's
tests of flex-wedge gate valves
have
indicated valve factors greater than 0.6.
The inspectors
considered that
additional valve specific justification was requi red for the use of generic
valve factors to demonstrate
the design basis capability of untested
valves.
The inspectors
noted that the licensee
had changed their approach to
justifying Stage II motor-operated
valves
and
had decided not to utilize the
EPRI performance prediction methodology to more precisely estimate the valve
factors for the feedwater
isolation valves
and other valves lacking specific
test validation of valve factor assumptions.
During previous
NRC inspections,
the licensee's
program had intended to use the
EPRI performance prediction
methodology for justification of untested
motor-operated
valves.
-13-
In response to the inspectors'oncern,
the licensee
reviewed their approach
to justifying untested
valves
and stated that they planned to utilize the
performance prediction methodology for the eight main feedwater isolation
valves
and the six power operated relief block valves.
Further, the licensee-
stated
they would consider the
EPRI performance prediction methodology or
other justifiable sources of data
as part of the justification for the
remainder of the untested
valves in their program.
The licensee stated they would review their grouping method in view of the
guidelines
presented
in Supplement
6 of Generic Letter 89-10.
In general.. the
inspectors
found the licensee's
method. consistent with the guidance of
Supplement
6.
Exceptions
were noted regarding the use of the highest
individual valve factor as the group valve factor.
The licensee is being requested
to submit additional detai
1 of their program
enhancements
in a written response
to this inspection report.
The inspectors
concluded that the licensee
had adequately
established
the
design basis capability of motor -operated
valves that had not been tested at
or near design basis conditions.
Some additional effort was expected
on the
licensee's
part to obtain applicable data to confi rm generic assumptions
used
in support of untested
motor-operated
valves'.6
Periodic Verification and Post-Maintenance
Testin
1.6. 1
Periodic Verification
The inspectors
reviewed the method utilized by the licensee for periodic
verification of design basis capability.
The licensee
had previously
committed to conduct periodic static testing only for each motor-operated
valve every 2 to 6 refueling outages
depending upon,probablistic risk analysis
risk significance,
performance history, margin,
and control logic utilized.
The licensee
considered that static testing provided adequate
periodic
performance monitoring assuming that observed
valve factors
remain constant.
Prior to this inspection,
the licensee
also committed to conduct additional
testing under differential pressure
conditions to validate their calculation
assumptions
relating to valve factors.
The licensee
planned to test nine
motor -operated
valves
(seven gates.
one butterfly, and one globe) over the
next three refueling outages.
Two differential pressure tests
would be
conducted during the
1R7 outage,
three during 1R8.
one at power
between
1R8
and
2R7,
and three during 2R7.
After that time, the need for further
differential pressure testing would be evaluated.
The licensee
issued Action
Request
A0362274 to track their commitment.
The licensee is being requested to provide additional details following the
inspection regarding their programmatic
enhancements
regarding the use of
differential pressure testing in their periodic verification program or other
justifiable sources of data.
i
-14-
The inspectors
found that the licensee
had not included any specific margin
for valve factor degradation
in their determination of motor-operated
valve
capability.
However.
based
on
a review of the available
margin for all motor-
operated
valves,
the i'nspector considered that of the licensee's
motor-
operated
valves
had adequate
margin to accommodate
some degree of valve factor
degradation until confirmatory differential pressure testing could be
performed during periodic testing.
Some valves.
however,
had very small
margins that
my not be sufficient to account for short-term degradation.
Accordingly. the licensee is being requested
to identify the existing margin
for valve factor degradation
in all motor-operated
valves in their program
as
part of their written response
to this inspection report.
The previous
Inspection Followup Item 9139-02 concerning periodic verification remains
open.
1.6.2
Post-Maintenance
Testing
The inspectors
reviewed licensee
Procedure
AD13. ID4, "Post-Maintenance
Testing," Revision 1.
The inspectors
reviewed the licensee's
post-maintenance
test matrix which identified maintenance activities requi ring subsequent
diagnostic testing to assure that design basis capability is maintained.
The
inspectors
found that the matrix requi red appropriate diagnostic testing
following motor-operated
valve maintenance;
dynamic testing
was identified
generally for valve modifications which could negatively affect efficiency or
valve factor.
However,
no specific maintenance activities were identified.
The inspectors
considered
the lack of prescribed differential pressure testing
following valve maintenance
(e.g.,
valve disc replacement
or reversal,
dimensional
rework. of valve internals. etc.) to be
a weakness.
In response
to
the inspectors'oncern,
the licensee identified detailed maintenance
and
modification activities which would require differential pressure testing.
The licensee stated that the detailed
post-maintenance
testing requirements
would be added to the post-maintenance
test procedure.
The inspectors
reviewed selected
maintenance
records
and found that appropriate testing
had
been performed.
The inspectors
considered
the licensee
actions to be
adequate.
The inspectors
reviewed Maintenance
Procedure
HP M-51.38, "Inspection
and
Haintenance of Wedge Gate Valve Internal
Components."
Revision 0.
The
inspectors
found that the licensee
procedure
required
a detailed inspection
and fitup procedure for all wedge
and parallel disk valves.
The procedure
incorporated the lessons
learned
from industry experience
for good maintenance
pr actices
and critical areas of emphasis
for long-term control of valve
erformance.
The licensee
considered
the detailed inspection information to
e critical in their determination of required differential pressure testing
for post-maintenance
testing.
The inspectors
found the implementation of the
detailed
mai'ntenance
procedure to be
a noteworthy strength in the licensee's
program.
-15-
1.7
Failure Anal sis
and Trendin
1.7. 1
Failure Analysis
The inspectors
reviewed the licensee's
actions in response to recent
operational
problems
encountered
during testing of motor-operated
valves.
According to the licensee,
one operational
fai lure to perform on demand
had
been experienced
subsequent
to differential pressure testing.
The fai lure
occurred
on LCV-109.
The fai lure was considered
an isolated incident.
and the
root cause
involved auxiliary contacts
in the motor contacter in the motor
control center.
The licensee's
failure analysis
was addressed
during
a
previous
NRC inspection.
Based
on
a review of all action requests
for the past two years
regarding
motor -operated
valve problems,
the inspectors
found that the licensee
documented
problems in detail. thoroughly analyzed motor-operated
valve
failures'nd evaluated
design basis capability as
a condition of return to
service of the motor-operated
valve.
Several
noteworthy examples of licensee
failure analysis
are described
below.
1.7.1.1
Valve 1-8923A
The inspectors
reviewed the licensee
documented
results of,differential
pressure testing for Valve 1-8923A.
This Aloyco 6"-1501. split-wedge.
gate
valve displayed
a high closing force requirement
as compared to other
identical valves in its group.
The licensee
disassembled
and inspected
the
valve and found wear
and damage in the valve internals.
The licensee
attributed the high closing force to an inability of the valve disks to rotate
when wedging due to
a wear ridge on the ball of the ball-and-socket joint
which coupled the disks.
The licensee
concluded that the sharp corner of the
socket did not allow the valve disks to rotate to achieve its intended wedging
action.
The licensee's
corrective action involved chamfering the shoulder
on
the socket
and blending out the wear ridge on the ball.
In addition, the
~
licensee
has included two similar valves in the periodic verification program
to monitor for this wear mechanism.
1.7. 1.2
Directional Valve Factors
The inspectors
found that the licensee
had identified a significant flow
di rection effect on the valve factor performance of Aloyco, spit-wedge.
gate
valves with a ball-and-socket joint design.
Although considered to be
bi-directional valves,
the valve factor for flow in one direction was
consistently
observed to be twice the valve factor observed in the opposite
flow direction.
The licensee
analyzed the capability of the motor-operated
valve based
on the worst valve factor displayed.
Furthermore.
specific
procedural
requirements
were established
in Maintenance
Procedure
MP M-51.38
to assure
the orientation of the disk was controlled to prevent reversal of
the disk assembly.
-16-
The licensee stated they will notify the industry of the observed directional
effects
on Aloyco valves via the nuclear
news network.
Furthermore.
the
licensee
plans to notify the valve vendor of their observation
and
conclusions.
The inspectors
found the licensee
actions to be adequate.
1.7.2
Trending-
The inspectors
reviewed the licensee's
motor-operated
valve tracking and
trending program.
The licensee
had established
and was implementing
a
computer
based
program for trending motor-operated
valve data.
The inspectors
found the licensee's
program to be effectively implemented
and capable of
roviding meaningful
performance indicators.
The inspectors
considered
the
icensee's
trending activities to be
a strength in their ongoing program to
maintain design basis capability.
1.1 ~lit
1
The inspectors
reviewed "Site Quality Assurance
2R6 Technical
Support Outage
Assessment."
The licensee
performed this follow-up assessment
of items
identified in thei r Generic Letter 89-10 Management
Prerogative Audit
(Audit 94016I), the 89-10 High Impact Team's activities during the
2R6 outage.
and the Stage
2 qualification of the main feedwater
isolation valves.
The inspectors
found that the licensee's
self-assessment
efforts were
comprehensive
and indicated
a strong commitment to ensuring the design basis
capability of motor-operated
valves.
The inspectors
reviewed reports
summarizing the activities discussed
above
and noted that the findings and
observations
were substantive.
As
a result of the licensee's
self-
assessments,
several
action requests
were initiated.
Action Request
A0354168, "Qualification of Stage
2 Motor-Operated
Valves
(Generic Letter 89-10 Program),"
was
among the action requests
generated
by
the quality assurance
self assessment.
This recent action request identified
a need to prepare
adequate justification for the change in commitments to the
NRC regarding the basis for the Stage
2 qualifications of the main feedwater
isolation valves.
This action had not been completed prior to the
NRC closure
inspection.
The'icensee
was requested
to address
closure of all outstanding
quality assurance
findings in a written response to this inspection report.
The inspectors
found the strong quality assurance
involvement in the
licensee's
closure activities to be
a performance strength.
0
'
1. 9
~0en
Items
-17-
1.9. 1
0 en
Ins ection Followu
Item 93-19-02:
Pressure
Lockin
and Thermal
~Bindin
This issue is discussed
in Section
1.3 of this report
~
This item will remain
open pending issuance of a planned
NRC generic communication addressing
pressure
locking and thermal binding.
1.9.2
0 en
Ins ection Followu
Item 91-39-02:
Periodic
Ver ification
This item is discussed
in Section 1.6 of this report.
This item will remain
open pending the licensees
submittal of additional information and
NRC review.
0
ATTACHMENT 1
1
PERSONS
CONTACTED
1. 1
Licensee
Personnel
- H. Angus,
Manager,
Regulatory
and Design Services
- H. Baker,
Engineer,
Electrical Maintenance
- W. Crockett,
Manager,
Engineering Services
- H. Frauenheim,
Engineer,
Electrical Maintenance
W. Fujimoto. Vice President,
Nuclear
Power Generation
- R. Goel,
Engineer,
Nuclear
Engineering Services
- T. Grebel. Director, Regulatory Compliance
- K. Hubbard,
Engineer,
Regulatory Compliance
- C. Lewis, Engineer,
Nuclear Quality Services
~J.
Holden.
Manager,
Maintenance
Services
- H. Phi lips. Director, Technical
Maintenance
L. Pulley.
Engineers
Nuclear Engineering Services
K. Riches.
Engineer.
Regulatory Compliance
- A. Toy, Engineer,
Predictive Maintenance
- L. Womack. Vice President.
Nuclear Technical
Services
- M. Williamson, Engineer,
Onsite Nuclear Engineering Services
- J. Young.
Directors'uclear
Quality Services
1.2
NRC Personnel
- T. Westerman,
Engineering
Branch Chief, Division of Reactor Safety,
RIV
The personnel
listed above attended
the exit meeting.
In addition to the
personnel
listed above,
the inspectors
contacted
other
personnel
during this
inspection period.
- Denotes personnel
that attended
the exit meeting.
2
EXIT HEETING
An exit meeting
was conducted
on January
27.
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.
ATTACHHENT 2
Information Requested to Clarify Certain
Program Commitments
Regarding Generic Letter 89-10
Clarify intended
use of the Electric Power Research
Institute
Performance
Prediction Hethodology or other justified sources for
confi rming the design basis capability of motor -operated
valves which
have not been tested
under differential pressure
conditions in the
Diablo Canyon Generic Letter 89-10 program.
Specifically address
the
main feedwater isolation valves
and the block valves for power
operated
relief valves
(Section 1.5).
'.
Provide the results of the Pacific Gas
5 Electric's review of the
grouping of untested
motor- operated
valves consistent with the guidance
of Supplement
6 to Generic Letter 89-10.
In particular justify any
exceptions to the use of the highest individual valve factor as the
group valve factor (Section 1.5).
3.
Provide additional detail of the intended
use of periodic verification
testing to validate assumptions
regarding the effect of neolube
on valve
factors
(Section
1. 1.5).
4.
Provide additional detail of the use of differential pressure testing
as
periodic verification (Section 1.6.1).
s.
Identify the margin for valve factor degradation existing in all motor-
operated
valves in your Generic Letter 89-10 program (Section 1.6. 1).
Identify the status of any outstanding quality assurance
audit findings
related to motor-operated
valve program closure
(Section 1.8).
0'