ML20153F521
| ML20153F521 | |
| Person / Time | |
|---|---|
| Site: | McGuire, Mcguire  |
| Issue date: | 04/29/1988 |
| From: | William Orders, Peebles T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML20153F513 | List: |
| References | |
| 50-369-88-09, 50-369-88-9, 50-370-88-09, 50-370-88-9, NUDOCS 8805100356 | |
| Download: ML20153F521 (12) | |
See also: IR 05000369/1988009
Text
-
[in Rio
- UNITED STATES
NUCLEAR REGULATORY COMMISSION
o
I ,^
REGION il
h8Y h ,
,
.101 MARIETTA STREET.N.W.
- { v /
g
ATLANTA GEORGIA 30323
g
.....
,
i
i
'
Report Nos.:
50-369/88-09 and 50-370/88-09
Licensee:
Duke Power Company
l
422 South Church Street
!
Charlotte, NC 28242
!
,
Docket Nos.:
50-369 and 50-370
License Nos.:
Facility Name: McGuire 1 and 2
,
Inspection Conducte g
arch 19, 1988 - April 22, 1988
Inspector [
AM/
<'w;mrders, Senior)dsidentInspector
/Da4V 5fgneo
Accompanying Personnel:
D. Nelson
R. Croteau
!
/[4
/
A
Approved by:E A. Peebles, Sectior Chief
D&te'5igned
'
DivisionofReactorFrojects
,
.
SUMMARY
.
Scopt,
This routine unannounced inspection involved the areas of operations
safety verification, surveillance testin
maintenance activities, and
follow-up on previous inspection findings, g,
.
!
i
Results:
In the areas inspected
three violations were identified.
One
I
violation was identified which ir$cluded two examples for failora to follow
'
!
procedure during safety valve testing and an inadequate procedure for slave
relay testing.
A second violation was identified which involved an inoperable
l
component cooling train and the *. aird violaticr. involved a failure to perform
i
post maintenance testing which rendered a nuclear service water train
j
,
1
1
l
J
s
I
I
i
l
8805100356 880429
ADOCK 05000369
'
Q
.i
-_
..
-
..
- - - - , _ - .
--- -.
-
- _ , - _ - - _ . _ -
. . _ _ - , - - _ _ _ -
- - _ - - ,
_
..
- .' .
'
.
REPORT DETAILS
1.
Persons Contacted
Licensee Empicyees
- T. McConnell, Plant Manager
8. Travis, Superintendent of Operations
i
- H. Suple, Superintendent of Maintenance
B. Hauilton, Superintendent of Technical Services
R. Sharpe, Compliance Engineer
J. Boyle, Superintendent of Integrated Scheduling
L. Firebaugh, OPS /NPE/MNS
- S. LeRoy, Licensing, General Office
- D.
Baxter, OPS /MNS/NPD
- S. Copp, Planning Engineer
R. Panner, Compliance
J. Snyder, Performance Engineer
- N. Atherton, Compliance
W.Reeside}AEEngineerOperations Engineer
R. White,
- G. Gilbert, MNS/NPD
'
,
Other licensee employees contacted included construction craftsmen,
technicians, operators, mechanics, security force members, and office
personnel.
- Attended exit interview
2.
Exit Interview (30703)
The inspection findings identified below were summarized on April 22,
i
1988, with those persons indicated in paragraph 1 above.
The following
items were discussed in detail:
l
1
(OPEN) Violation 370/88-09-01, Failure to follow procedure for
Pressurizer Code Safety Valve Testing and Inadequate Procedure for
Slave Relay Testing.
(Seeparagraphs5and9).
1
(OPEN) Violation 369/88-09-02,
Train Due to Inoperable Nuclear Service Water (RN) Valve. Inoperable Co
See
{
.'
paragraph 10).
!
(OPEN) Vio1rion 369/88-09-03, Inoperable RN Train Due to a Failure
to Test RN-21.
(See paragraph 10).
The licensee representatives present offered no dissenting comments, nor
did they identify as proprietary any of the information reviewed by the
inspectors during the course of their inspection.
- -
- -
-
.'
-
. .
.
'
2
'
3.
Unresolved Items
An unresolved item (UNR) is a matter about which more information is
required to determins whether it is acceptable or may involve a violation
or deviation.
There were no unresolved items identified in this report.
4. Plant Operations (71707, 71710)
The inspection staff reviewed plant operations durir:g the report period to
verify conformance with applicable regulatory requirements. Control room
logs, shift supervisors' logs, shift turnover records and equipment
removal and restoration records were routinely aerused.
Interviews were
conducted with plant operations, maintenance, clemistry, health physics,
and performance personnel.
Activities within the control room were monitored during shifts and at
!
shift cMnges. Actions and/or activities observed were conducted as
prescribed in applicable station administrative directives. The complement
of licensed personnel on each shift met or exceeded the minimum required
by Technical Specifications.
Plant tours taken during the reporting period included, but were not
limited to, the turbine buildings, the auxiliary building, Units 1 and 2
electrical equipment rooms, Units 1 and 2 cable spreading rooms, and the
.
station yard zone inside the protected area.
During the plant tours, ongoing activities, housekeeping, security,
equipment status and radiation control practices were observed.
1
Use of overtime by operations was reviewed to verify cornpliance with TS
requirements.
Documentation showed that the maximum overtime limits were
exceeded approximately nine times in 1987-1988, primarily for outage
support.
It is allowable to exceed the maximum limits for very unusual
circumstances.
Operations Management Procedure 1-7, Shift Manning and
Overtime Requirements, requires that overtime worked in excess of
guidelines be authorized in advance by the station manager or his designee
(another high level of management).
The inspector noted that in September
of 1987 two instances of exceeding the maximum overtime limits were not
authorized in advance.
The licensee received a violation for this issue
(see Inspection Report 369, 370/87-26) in the September 1987 period and
corrective actions have been taken.
a.
Unit 1 Operations
Unit 1 began the reporting period at full power.
On March 23, 1988,
the unit emerienced a saurious safety injection (SI), main steam
isolation ar,d reactor tr p.
The spurious SI signal was generated in
a Solid State Protection Syetem (SSPS) caLinet containing the
circuitry for A train low steam line pressure SI and main steam
isolation.
Licensee technicians had just completed testing in this
'
. - - _ _ _ _ _
__
_
_
'
.-
-
..
. .
.
3
cabinet and were closing the door.
The spurious signal occurred when
the door was shut.
All A train SI components actuated and the hiah
head pump injected into the primary coolant system.
isolation and feed water isolation occurred, generating a turbine
trip and resultant reactor trip.
Operators determined that the SI
was inadvertant and secured the injection in approximately nine
minutes.
No major problems occurred during the transient.
All A
train SI components functioned as designed during the transient,
except that the reactor trip was caused by the turbine trip above P-8
instead of directly by the SI actuation as would be expected.
The
cause for this was determined later and is discussed below.
During
the event a low steam generator level in the C steam generator caused
auxiliary feedwater (CA) to initiate a second time.
After the unit was stabilized, Instrumentation and Electrical (IAE)
technicians attemnted to determine the cause of the SI signal.
The
spurious signal was duplicated several times by agitating the SSPS
cabinet.
However, the signal could not be repeated following the
shutting of a nearby heavy door.
Further investigation could not
determine the exact cause of the problem.
The licensee considers
that a sina11 piece of loose wire or other conductor had shorted or
grounded the circuitry u)on agitation of the cabinet.
According to
4
the licensee, the additional agitation conducted during trouble-
.
shooting, and finally the shutting of the nearby door served to
'
vibrate the conductor away from the electrical centacts.
A thorough
cleaning of the cabinet was performed.
This arod'ned some small wire
fragments which could have caused what the l'censee postulates.
The
licensee also considers that the points of contact occurred in the
low steamline pressure SI circuit downstream from the point where the
reactor trip function branches off, thereby causing the SI to
initiate without causing a direct reactor trip.
Following the
cleaning a complete functional surveillance was performed with no
problems.
The unit was restarted and achieved full power the
,
following day.
No similar problems in SSPS have developed.
On March 27, load was decreased to 86 percent power due to decreased
load demand on the grid and was back at full power at 1247 a.m. on
i
March 28.
Later on March 28 power was reduced to 46 percent to
support removal of a voltage regulator control drawer.
Unit 1
-
returned to 100 percent power on March 29.
,
i
On April 16, Unit I was manually tripped from 100 percent
power due
to decreasing le<el in the C steam generator (SG) caused >y the C
feed regulating nice (FRV) failing shut.
The C FRV shut due to a
blown fuse on a wd controlling the valve.
The card was later
"
tested and found to be the cause of the blown fuse.
Auxiliary feed
water (CA) auto started on low steam generator levels but 1SA-49,
I
steam supply from the 8 SG to the turbine driven CA pump, did not
indicate open due to problems with the position indicating limit
i
switches.
The valve was actually open.
The limit switches were
.
-
-
.
.
-
-
-
-
,
- .l
i
-
..
-
.
-4
l
~
'
adjusted, tha' FRV card was replaced, and the unit returned to 100
t
percent power oa April 18.
l
b.
Unit 2 Operations
_
Unit 2 operated at full power for the entire period.
The SI on Unit
!
i
1, discussed above, did lowever affect Unit 2.
Due to the alignment
of the common portion of the Nuclear Service Water (RN) System, the
,
operating Unit 2 RN train was isolated by the single train SI on Unit
'
l
1.
Had both SI trains on Unit 1 actuated, the common portion of RN
would have realigned to ensure continued RN operation on Unit 2.
Various Unit 2 component temperatures elevated, but Unit 2_ operators
diagnosed and corrected the problem in time to prevent any additional
,
a
required actions.
No violations or deviations were identified.
5.
Surveillance Testing (61726)
Selected surveillance tests were analyzed and/or witnessed by the
inspector to ascertain procedural and performance adequacy and ccnformance
with applicable Technical Specifications.
4
Selected tests were witnessed to ascertain that current written approved
procedures were available and in use, that test equipment in use was
calibrated, that test prerequisites were met, that system restoration was
completed and test results were adequate.
l
,
Detailed below are selected tests which were either reviewed or witnessed:
PROCEDURE
EQUIPHENT/ TEST
,
PT/0/A/4150/05
Pressurizer Safety Valve.Setpoint Test
PT/1/A/4403/007
RN 1A Flow Balance Test
i
PT/2/A/4200/28A
SSPS Slave Relay Tests
PT/1/A/4208/03A
Train 1A NS Heat Exchanger Performance Test
PT/1/A/4252/018
CA Pump IB Performance Test
PT/1/A/4601/088
SSPS Train B Periodic Test
PT/1/A/4403/018
RN Train IB Performance Test
PT/1/A/4403/01A
RN Train 1A Performance Test
PT/1/A/4206/01A
NI Pump 1A Performance Test
PT/1/A/4252/01A
CA Pump 1A Performance Test
PT/1/A/4204/01B
ND Pump 1B Performance Test
PT/2/A/4209/01A
NV Pump 2A Performance Test
PT/2/A/4206/01A
NI Pump 2A Performance Test
PT/0/A/4350/38
125 VDC Battery Service Test
See paragraph 9 for further information conce ning PT/0/A/4150/05.
- .l
. .
.
5
..
On March 22 at 10:02 AM, a procedure error in a performance seriodic test
procedure caused re-alignment of power sources to severa'
Unit 2
'
non-safety containment ventilation systems.
A step in grocedure
i
PT/2/A/4200/28A, Slave Relay Test, directs the opening of a sliding link"
to prevent ventilation units from tripping during the test.
This occurs
in the section of the procedure that tests slave relays in the Train A
Safety Injection SSPS circuitry.
The step specifies opening sliding link
H-3 in cabinet 2ATC8.
When the system was actuated in a subsecuent step,
,
several non-safety containment ventilation systems experiencec a shunt
trip to re-align their power sources to non-safet
systems were lower containment ventilation (VL)y buses.
The affected
Upper Containment
,
Ventilation (VU), and Control Rod Drive Mechanism Cooling" Ventilation
(VR).
These non-safety systems are designed to "load shed in the event
of an ESF actuation to lessen the electrical load on safety system power
sources.
The loid shed took 31 ace because the wrong sliding link was
specified.
A step at the comp:etion of the test specifies closing sliding
link I-2 in the same cabinet.
I-2 is the correct sliding link that should
have been opened originally.
Having the wrong sliding link open during
the test did not othenvise adversely effect the test or plant operation.
The procedure error constitutes an examp/88-09-01).le of an inadequate proce
is therefore an apparent violation. (370
The licensee states that the procedure error occurred in a recent re-write
of the procedure, but has no explanation for why it occurred.
Another
case of an unexplained procedure change occurred recently in an operations
procedure which was discussed in NRC inspection report 50-369,~370/88-04.
In that case two procedure steps were interchanged during re-write.
A
violclion was issued for that occurrence, but the licensee's corrective
action appeared to be limited to the operations organization where the
problem occurred.
When the corrective actions and lessons learned are
shared with other departments, similar problems may be prevented.
6.
MaintenanceObservations(62703)
Routine maintenance activities were reviewed and/or witnessed by the
resident inspection staff to ascertain procedural and performance adequacy
and conformance with applicable Technical Specifications.
The selected activities witnessed were examined to ascertain that, where
a3plicable, current written approved procedures were available and ir use,
t1at prerequisites were met, that equipment restoration was completed and
maintenance results were adequate.
No violations or deviations were identified.
7.
Follow-up on Previous Inspection Findings (92702)
The following previously identified items were reviewed to ascertain that
the licensee s responses, were applicable, and licensee actions were in
compliance with regulatory requirements and corrective actions have been
1
_
.
.
.
.
6
completed.
Selective verification included record review, observations,
and discussions with licensee personnel.
I
(CLOSED) Violation 369, 370/85-06-04.
Failure to Take Prompt Corrective
Action to Notify Operations Personnel of Potential Degradation of
Auxiliary Feedwater System and Correct Improper Valve Installation.
.
Nuclear Station Modifications (NSM's) were completed by July 21, 1985
which installed temperature monitors to detect check valve leakage and
replaced the stop check valves with a different design valve.
The
completed NSM's were reviewed and selected check valves were physically
verified to be in place by the inspector.
The violation was initially
denied but the NRC determined that the violation occurred as stated in the
'
The corrective actions stated in the response have
been completed and this item is closed.
(CLOSED) Inspector Followup Item 369, 370/86-28-04, Testing of Safety
Valves.
Procedure PT/0/A/4150/05, now contains instructions on observing
the trend of safety valve lift setpoints.
See section 9 for more
information on this PT.
,
,
(CLOSED) Unresolved Item 369/86-28-07, Blocking of Safety functions.
This
issue dealt with blocking of the low pressure safety injection signal when
a safety valve opened and caused excessive blowdown while in hot standby,
Mode 3.
The licensee has reinforced the policy of not blocking automatic
safety actuations except when directed by approved procedures or
This : tem is' closed.
(CLOSED) Violation 369, 370/86-28-01, failure to Report.
Corrective
actions have been taken and this item is closed.
8. Licensee Event Report (LER) Followup (90712, 92700)
The following LER's were reviewed to determine whether reporting require-
ments have been met, the cause appears accurate, the corrective actions
i
appear appropriate, generic applicability has been considered, and whether
t1e event is related to previous events.
Selected LER's were chosen for
more detailed followup in verifying the nature, impact, and cause of the
event as well as corrective actions taken.
l
(CLOSED) Licensee Event Report 369/86-17, Both Trains of Hydrogen
Mitigation System Inoperable.
Multiple failures of hydrogen ignitors
during quarterly surveillance testing resulted in both trains of the
hydrogen mitigation system being declared inoperable.
The licensee has
evaluated the life expectancy of the ignitors and estimates six years as a
conservative life expectancy.
Corrective action includes replacing the
ignitors every four years.
(CLOSED) Licensee Event Report 370/86-03, Unidentified Reactor Coolant
Leakage Due to Leaking Valves Resulting in Shutdown.
Four valves were
found to be leaking and subsequently repaired,
i
1
-
-
-
.:
.
.
.
.
7
(CLOSED) Licensee Event Report 370/86-06, Failure to Maintain Required
Boration Flow Path Due to Personnel Error.
The event was discussed in an
operations staff aeeting and shift supervisor meeting.
(CLOSED) Licensee Event Report, 370/86-19, Missed Surveillance on
Essential Auxiliary Power Systems.
Corrective actions have been completed
and this item is closed.
(CLOSED) Licensee Event Report 369/87-02, Both Trains of Containment Spray
System Inoperable.
This event resulted in violation 369/87-04-01 and
corrective actions are being tracked in followup to the violation.
9.
Pressurizer Safety Valve Setpoint Testing
During a review of the completed data sheets for PT/0/A/4150/05,
Pressurizer Safety Valve Setpoint Test, it was noted that the maintenance
personnel signing the data sheets for satisfactory lift checks were not
the personnel who were trained to perform the tests. In a letter to the
NRC dated July 22, 1987, the licensee committed to allowing only specially
trained personnel to work and test these valves.
This commitment was made
-
in response to violation 369/86-28-06 which involved a primary system
safety valve opening at a pressure outside the T.S. limit.
The tests in
question were performed on September 14, 1987; June 11, 1987 and June 13,
1987.
The licensee stated that the qualified individuals were present but
non qualified individuals signed the data sheet for the qualified
individuals.
The licensee stated that in the future the data sheets will
be clearly annotated if a non qualified person signs for a qualified
,
person and the data sheet will contain the name of qualified person
performing the test.
Another problem with the test performed on pressurizer code safety valve
2NC1 on June 11, 1987, was discovered.
The data sheet for this test
listed a lif t pressure of 2513 psi for the second lift of the valve.
The
procedure specified that each lift must be within the TS required range of
2485 psig plus or minus 1 percent (2461 to 2509).
The other lift
pressures were within the required band and the average of the three lifts
was also within the required range.
This is a second example of an
apparent violation (370/88-09-01) of T.S. 6.8.1 for failure to properly
implement the written procedure for pressurizer safety valve setpoint
testing.
Corrective actions for violation 369/86-28-06 should have
prevented this violation from occurring in that personnel were trained on
,
the specific requirements.
Contributing to the violation was the fact
that the data sheet did not clearly specify that each lift must be within
the required band.
.
The licensee initiated a Problem Investigation Report when this item was
brought to their attention by the inspector.
The Itcensee has stated that
the maintenance and quality control personnel involved believe the actual
lift setpoint was in the required band but the data was incorrectly
recorded.
All parties state they were aware each lift was required to be
i
-
.
_
-
.
.
.
. .
'
.
8
l
within plus or minus 1 percent of 2485 psig.
The proposed corrective
actions include changing the procedure, counciling the persons involved,
-
'
,
and expediting the development of a training program in this area.
During the review of the p(ressurizer safety valve setpoint test, Electric
Power Research Institute
EPRI) Report NP-4235, Set point Testing of
!
!
Safety Valves Using Alternative Test Methods, was reviewed.
This EPRI
'
report was prepared to present the results of tests performed and to
correlate alternate test methods for safety valve tests to testing using
>
i
full pressure steam as the test medium.
The full pressure steam test
method most closely simulates the actual conditions which the valve
'
experiences in the system.
It is noted, however, that McGuire has loop
seals in the lines from the pressurizer to the code safety valves so steam
is not actually on the valves.
The licensee uses an alternate test method
j
using nitrogen as the pressure medium rather than steam.
The EPRI test
>
results appear to indicate that:
1
a.
Tolerance bands using the nitrogen test method need to be much
.
tighter than plus or minus 1.0% in order to assure the valve will
'
lift at TS recuired 2485 plus or minus 1.0% psi while installed in
!
,
the system.
The licensee currently uses a plus or minus 1.0%
i
tolerance band using nitrogen.
b.
The actual valve lift setpoint using steam will be lower than the
setpoint using nitrogen.
It was discovered that generic correlations
'
3
could not be made to relate nitrogen tests to actual in place lift
setpoints; however, correlations on a valve-by-valve basis can be
-
'
made with a higher degree of confidence.
The method for determining
,
the valve correlations is given in Appendix E of the EPRI report.
1
The licensee currently does not use any correlations to correct the
!
'
lift point using nitrogen to the lift point using a steam medium.
j
-
T.S. 3.4.2.2 requires that pressurizer code safety valves have a lift
i
setting of 2485 psig plus or minus 1 percent and the lift setting pressure
i
i
shall correspond to ambient conditions of the valve at nominal operating
'
t
temperature and
aressure.
Based on the information available in EPRI
!
Report NP-4235, it is not clear that the pressurizer code safeties would
-
lift within the TS required range at normal system operating temperature
i
and pressure.
The licensee has indicated that the EPRI report does not
i
take into consideration the fact that McGuire has loop seals in the lines
)
to the pressurizer code safeties.
According to the licensee, the
,
temperature of the water at the code safety is 140 degrees F and testing
y
at Wyle Laboratories has confirmed direct correlation between nitrogen
j
lifts under ambient conditions and nitrogen with 140 degree F water at the
i
valve inlet.
The licensee stated that Catawba and Oconee send their
i
]
pressurizer code safeties to Wyle to have hot lifts
performed since
i
neither have loop seals.
This item was still being rev'ewed at the end of
'
the inspection period.
I
\\
i
I
a
l
,'
.
. .
'
.
9
10.
Nuclear Service Water Valves Inoperable
On March 9,1988, during a system walkdown a licensee engineer discovered
Unit 1 Nuclear Service Water (RN) valve 1RN1908, (RN to component cooling
(KC) heat exchanger 18 throttle valve) to be inoperable.
A travel stop
that limits the maximum 03en position of the valve had become repositioned
toward the closed direction and was found to be loose.
If the valve had
been called upon during an Engineered Safety Features (ESF) actuation, the
repositioned stop would have prevented required RN flow to the KC heat
exchanger from occurring, and would have possibly prevented the KC system
from adequately cooling critical com)onents in a design basis accident.
The travel stops are precisely set curing RN system flow balance tests
designed to balance RN system flow among all the RN heat loads, including
KC.
The licensee estimates that RN flow to the KC heat exchanger would
have been reduced approximately 1500 gpm below the required 6000 gpm in
the event of an ESF actuation.
The licensee considers that the heat
removal capability of the 8 train of KC, although impaired, was adequate
due to low RN temperature and the clean condition of the KC heat
exchanger. This hypothesis is based on design engineering analysis.
Upon
discovery, the licensee declared the KC train inoperable and took prompt
action to restore the travel stops to the most recent tested position.
'
The last RN flow balance test was conducted on January 29, 1988, at which
time the valve stops were adjusted.
The licensee could not produce any
documentation or evidence of authorized work conducted on 1RN1908 that may
have affected the travel stop positions since this last test.
It is known
that the travel stops have been routinely used to secure the valve in the
-
i
shut position to facilitate isolation for other work. A work recuest
usual'y documents these occasions.
The licensee concluded that tie hex
nuts securing the travel stop had vibrated loose allowing the stop to
drift in the closed direction.
In normal operation the valve is throttled
significant'ly in the closed direction with the travel stop performing no
function, thus it was free to drift upon loosening of the nuts.
Technical Specification 3.7.3 requires that both trains of KC be operable
during operation in Modes 1, 2, 3, and 4.
One train may be inoperable for
-
up to 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> in these modes.
The mis positioned travel stops on IRN1908
resulted in train B of KC being technically inoperable from the last
documented position of the valve on January 29, until the discovery of the
problem on March 9.
This is an apparent violation (369/88-09-02) of the
action statement requirements of Technical Specification 3.7.3.
In an unrelated event, on March 28, 1988, the licensee determined that
i
valve IRN-21, RN Strainer IA automatic backwash valve, underwent
1
maintenance on February 4,1988, without subsequent retest.
This valve is
,'
designed to automatically open upon high differential pressure across the
1A RN strainer thereby allowing backwash flow to clean the strainer.
This
i
flow is diverted from the total A train RN flow.
Upon an ESF signal,
!
RN-21 shuts, if open, to ensure that all RN flow is supplied to ESF heat
loads cooling critical components in accident situations.
A packing
,
!
l
.
-
-
-
-
,'
- '
. .
.
10
adjustment was performed on RN-21 on February 4,1988, under a work
request to investigate and correct a packing leak.
The work request
incorrectly identified that a retest was not required.
Maintenance
personnel tightened the packing but determined that the )acking leak could
not be stopped without over tightening the packing there)y impairing valve
stroke. The work request had a contingency to be re-scheduled until an
outage if tightening the packing was unsuccessful.
The work request was
!
returned to planning for this purpose.
Contributing to this problem was
the fact that maintenance clearance was also deemed to be not required
which resulted in operations being not fully informed of the extent of
work being conducted on the valve.
j
. Technical Specification 4.0.5 states that testing to ASME Code require-
ments is required to properly retest ESF components following maintenance.
Station Directive 3.2.2, Identifying and Performing Plant Retesting,
implements these requirements by identifying the components and types of
!
maintenance that require retests as well as identifying the retest
j
required.
RN-21 is identified as a component requiring retest.
!
Adjustment of stem packing is an example of maintenance requiring retest.
'
In tnis case a valve stroke timing test should have been performed since
the valve is required to shut within 60 seconds upon receipt of an ESF
signal.
!
On March 28, 1988, the licensee detected the error on the work request and
immediately added RN-21 to the TS Action Item Log for RN train A which was
currently declared inoperable for unrelated reasons.
The performance of a
stroke timing test at that point could have determined valve operability,
'
however, add"tional packing adjustment took place first.
During the
subsequent stroke timing test, the valve failed to shut.
Licensee review
of operator aids computer (OAC) data revealed that the valve was actually
required to automatically open and shut numerous times between February 4
and March 28 to correct high strainer differential pressure.
Typical
i
,
stroke times on these occas'ons, as recorded by the OAC, show that the
,
valve shut well within the maximum time permitted, (approximately 10
l
seconds vs maximum 60 seconds) The licensee has stated that it is likely
5
that satisfactory results would have been obtained had a formal stroke
timing test been conducted after the initial packing adjustment.
It was
i
likewise hypothesized that the valve would have performed its safety
j
function during the time of unknown inoperability.
The licensee considers
l
that the final packing adjustment caused the valve to fail to shut.
Technical Specification 3.7.4 requires that two trains of RN be operable
in modes 1, 2, 3, and 4.
One train may be inoperable for up to 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />
in these modes.
The failure of the work request to aroperly identify that
a retest was required caused the requirements conta:ned in TS 4.0.5 to be
omitted.
This resulted in RN-21 and thus train A of RN to be inoperable
from February 4 to March 28, 1988.
Mitigating factors, however, lessen
the significance of the RN-21 inoperability.
The total flow diverted with
RN-21 open is approximately 700 gpm.
This amount is a small portion of
the total RN flow of greater than 12,000 gpm available from one train
-
.
-
. .
,
.
11
f
during accident conditions.
Also, RN temperature during the time of
inoperability was in the range of 40 to 50F, well below the design
temperature of 78F. Design engineering evaluation indicates that ample
heat sink existed for A train RN to perform its safety function. The prime
concern and root cause of this particular event is the failure to retest
as described above.
This item is identified as an apparent violation
(369/88-09-03) of Technical Specification 4.0.5.
The inoperability of the two RN valves discussed above occurred over an
extended period of time resulting in numerous occasions when both trains
of RN or the systems they support were rendered inoperable.
Most notable
is the overlapping period (February 4 to March 9) when both RN valves were
inoperable rendering both trains of KC inoperable.
The NRC recognizes the
mitigat;ng factors discussed above and considers the safety significance
of these specific events to be minimal.
However, had conditions been
less favorable or other components been involved, the safety significance
could have been much greater.
The NRC is particularly concerned with
'
,
the events associated with the RN-21 problem from the stand 30 int of
maintenance work control and retesting.
The inspectors have initiated a
thorough study of the work control process to determine if sufficient
controls are in place to prevent missed retests and unknown inoperabili-
i
ties of safety system components.
11.
Ground Water Detection
On April 12, 1988, NRR technical staff reviewed the Groundwater Monitoring
System and conducted a walkdown of selected monitoring wells ( Auxiliary
Building East and West wall exterior monitors, Auxiliary Building north
wall interior monitors PP-51, QQ-56 and PP-61), and Auxiliary Building
drain sump
"C".
System operators and surveillances were found to be
consistant with T.S. 3.4. 7.13 requirements.
On April 13, 1988 design
calculations of hydrostatic and buoyancy influences and overturning
potential for the Auxiliary Building, Reactor Buildings, and Diesel
Generator Buildings were audited at licensee's corporate engineering
offices.
This review is part of the NRC's review of Duke's request for
Technical Specifications 3.4.7.13 changes dated January 27, 1988.
l
l
.
. -
.
.
.
.
.
.