ML20217D169
| ML20217D169 | |
| Person / Time | |
|---|---|
| Site: | Waterford  |
| Issue date: | 07/15/1991 |
| From: | Westerman T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML20217D153 | List: |
| References | |
| 50-382-91-18, NUDOCS 9107220080 | |
| Download: ML20217D169 (15) | |
See also: IR 05000382/1991018
Text
_ _ _ . . . . - -
, - _ _- ----- -_ _____-.- _ -_ _--- _
.
-
,
\\
-
_
,
APPENDIX
U.S. NUCLEAR REGULATORY COMMISSION
REGION IV
NRC !aspectio) Report No:
50-382/91-18
Docket No:
50-382
License No:
Licensee:
Entergy Operations, Incorporated
P. O. Box B
Killona, Louisiana 70066
Facility Name: Waterford Steam Electric Station, Unit 3 (Waterford 3)
Inspection At: Taft, Louisiana
Inspection Corducted: May 14 through June 17, 1991
Inspectors:
W. F. Smith, Senior Resident Inspector
Project Section A, Division of Reactor Projects
S. D. Butler, Resident Inspector
Project Section A, Division of Reactor Projects
7-/P7 /
Approved:
_ _ _ _ _
Chief, Pr6. ject Section A
bate
T. F. West
Inspection Summary
'
Inspection Conducted May 14 through June 17, 1991 (Report 50-382/91-18)
Areas Inspected:
Routine, unannounced inspection of plant status, followup,
onsite response to events, monthly maintenance observation, bimonthly
surveillance observation, and operational safety verification.
Results:
No violations or deviations were identified; however, there were
three unresolved items discussed in this report that have the potential of
becoming violations. An unresolved item is a matter about which more
information is required to ascertain whether it is an acceptable item,
deviation, or violation. These unresolved items and other inspection
highlights are summarized below:
In the plant operations area, house u ping practices were excellent as the
plant transitioned from refueling outage conditions to operating conditions.
Operator performance during startup and physics testing was exemplary; the
operators demonstrated many strengths in the response to events that occurred
during this inspection period. When the reactor coolant system developed a
20 gallon per minute (gpm) unisolable leak on May 17, 1991, the operators
l
followed the appropriate emergency plan, off-normal procedure, and operating
procedures to accomplish a safe and expeditious plant cooldown and
'
9107220000 910715
DR
ADOCK 050
2
l
l~
- . _ _ _ _ _ - - , , _ - , ,
.-
-
.
- _ - _ _ _ _ _ _ .
_
'
.
.
2
depressurization.
To further expedite depressurization, the shift supervisor
first consulted with plant management and then issued an order to isolate
safety injet'Jon tanks, thus entering Technical Specification (TS) 3.0.3.
The
a ._
inspector viewed this ac+1on as prudent; however, it remains unresolved whether
a violation of NRC regulations occurred (paragraph 4.1).
In the radiological controls area, the licensee's handling of the potential
overexposure incident on May 13, 1991, appeared adequate.
Following an
appropriate investigation, the licensee's determination that the individual had
not been overexposed had a sound basis..
The issue remains unresolved, pending
i
followup by a regi nal radiation protection specialist (paragraph 3.2.1).
Radiological work practices and cecontamination efforts at the end of the
refueling outage were a strength.
Maintenance and surveillance conducted during this reporting period did not
reveal any weaknesses, except in the assurance of qua 'ty of replacement parts
I
during repair of the governor on the turbine driven
gency feedwater pump.
An unresolved item was identified regarding this is>we (paragraph 5.1).
During
surveillance testing of High Pressure Safety Injection (HPSI) Pump A, the
licensee took extensive measures to resolve the reduction in recirculation
flow, thus demonstrating lessons learned from previous problems on HPSI Pump B.
This was viewed as a strength.
9
-
4
_
._ _ _ _ _ _ _ _ _ _ _ _ _ _
_
_ _ _ _ _ _
.
-
-3-
DE '
1.
PERSONS {0NTAClcJ
1.1 Principal Licensee Epp.lpyees
D. F, Packei, General Manager, Plant Operations
- T. R. Leonard, Technical Services Manager
- R. 5. Starkey, Operations and Maintenance Manager
- A. 5. Lockhart, Quality Assurance Manager
- R. E. Allen, Security and General Support Manager
- D. E. Baker, Director, Operations Support and Assessments
R. G. Azzarelin, Director, Engineering
W. T. Labc.ite, Radiation Protection Superintendent
- G. M, Davis, Events Analysis Reporting & Response Manager
- R.
F. Burski, Director, Nuclear Safety
- T. J. Gaudet, Operational Licensing Supervisor
L. W. Laughlin, licensing Manater
J. G. Hoffpauir, Maintenance Superintendent
D. W. Vinci, Operations Superintendent
A. G. Larsen, Assistant Maintenance Superintendent, Electrical
D. T, Dormady, Assistant Maintenance Superintendent, Mechanical
D. C. Matheny, Assistant Maintenarr.e Superintendent, Instrumentation and
Controls
- Present at exit interview,
in addition to the above personnel, the inspectors held discussions with
various operations, engineering, technical support, maintenance, and
administrative members of the licensee's staff.
2.
PLANT STATUS (71707)
During the reporting period, plant pressurization and heatup began following
the licensee's fourth refueling outage. The heatup was interrupted on May 17,
1991, when a 20 gpm reactor coolant system (RCS) leak from pressarizer spray
line check Valve RC-303 required the RCS to be cooled down and depressurized to
permit repairs.
Plant heatup was resumed on May 19 and a reactor startup was
performed on May 23 af ter completion of the necessary surveillance testing.
A
reactor trip occurred on May 26, from less than 0.1 percent reactor power,
during troubleshooting of a ground on the plant annunciator system. A second
reactor trip occurred on May 28, from 34 percent power, when nonsafety-related
Bus 381 failed to transfer to offsite power following a manual turbine trip.
The reactor was restarteo on May 29 and reached 100 percent power en June 2,
where it has remained through the end of the reporting Deriod.
_ - _ - _ _ _ _ _ _ _ _ _ __ _ ___- - - _ _
.-
_.4
_ _
-_ _
___ __
_
_ _ . _ _
_ _ _
+,
..
4
.
-
3.
FOLLOWUP
3.1 Followup _ of Previous Inspection Findings (92701)
3.1.1
(Closed) Inspector Followup _ Item (IFI) 382/9005-02
_
.This item was initiated to follow up on the licensee's actions following the
March 29, 1990, reactor trip caused by a power fault at a nearby substation.
Of particular interest was the licensee's consideration of providing a static,
uninterruptible power supply (SUPS) for the reetcar cutback system and the
steam bypass' control system.
The licensee imp Jmented Design Change
Package (DCP) 3320 during the fourth refueling out.ge, which connected those
systems to a SUPS. The systems were placed ta peration after the outage.
In
addition _to the above corrective actions, t.te licenseu d < eloped a 33-item
action tracking list with. Central Engineei ng, fo- the purpose of improving
i
reliability of the offsite power supplies. A few .najor examples of items
included:
control over maintenance by Ccatral Engineeriog, increased awareness
of the- equipment-out-of-service log, disp ,tcher 'rc iain;, a walkdown of the
nearby substation to determine condition, +ae i stab'11ty study of Waterford 3
pcwer busses.
The licensee actions in re>p;1st to the inspector's concerns
have:been satisfactory.
This IFI is clo'ed.
.
3.2 Other Followups (92701)
'
3.2,1 Apparent Overexposure Oue to Misplaced 1 crmoluminescent Dosimeter
On May 13, 1991,-the inspectors were informed that a contractor employee, who
,_
l
had permanently-left the site, may have received radiation exposure of
L
- approximately 3 rem, which was in excess of the limits allowed by 10 CFR 20.
-The initial determination was made by the licensee based on the dose recorded
i
by the employee's thermoluminescent dosimeter (TLD).
The_ licensee initiated
Potentially Reportable Ev'nt-(PRE)91-058 and began an investigation. The TLD
exposure was not consistent with the employee's self reading pocket
-
dosimeter'(SRD) readings, or with the 0.75 rem exposure received by fellow
,'
workers performing similar tasks.
It was subsequently determined that the TLD
reading did not appear to be as realistic as the employee's SRD readings, This
determination was based on interviews with the affected contract employee and
other employees working the same job, a review of work performed by the
employee, and exposure received.by fellow workers. ~lt was also reported that a
'
TLD was found-in the reactor containment building (RCB)'during the time that
the contractor employee was present in the RCB.
The cont ractor employee denied
losing his TLD, however, the licensee health physics personnel recalled that the
l-
individual had earlier lost and then recovered his TLD. A lost TLD report was
not_ initiated.
Since the licensee's administrative requirements for the lost
TLD were not followed, Quality Notice (QN) QA-91-116 was written.
The QN
documented the procedural violation and entered the licensee's corrective
action program.
Based on the licensee investigation, they assigned the worker
en exposure based on the SRD readings.
An NRC regional radiation protection
specialist was kept ir. formed by the licensee of their actions, and the
licensee's-approach to this issue appeared proper, pending further review
L
,
,
m
.,w
,
,e:
- . - - -
--
, - - ,
w
,,
,
--,
~
.
l
.
-5-
during a future inspection to be performed by a regional radiation protection
specialist.
This review shall be tracked by Unresolved Item 382/9118-01.
3.2.2
Low Recirculation Flow During Testing of HPSI Pump A
On May 20, 1991, during an operability verification surveillance test of HPSI
Pump A, the licensee noted that recirculation flow was 28.9 gpm.
This value
was below the " Action Required Limit" of 30.15 gpm per ASME Code,Section XI.
All other parameters recorded during the surveillance test were within
acceptable limits.
The pump was declared inoperable, and the licensee
assembled a cask force including maintenance and engineering personnel to
determine the cause of the low flow and establish corrective actions.
The task
force reviewed previous test data and noted that HPSI Pump A was full-flow
tested during the refueling outage on April 3,1991, and flow was consistent
with previous readings and was acceptable.
The flow measuring instrument was
vented and a calibration check was performed on the instrument.
The flow
instrument orifice had been previously disassembled and verified to be in good
condition and reinstalled properly on April 30, 1991.
Radiography was
performed on potentially blocked components (including the flow limiting
multi-stage orifice), but no obstructions were found.
Vibration data was
collected and reviewed, and no indication of pump damage was noted.
Historical
inservice test data taken since January 1935 was reviewed, and the licensee
noted some variations in recirculation flow that were not attributable to
changing pump performance.
The licenr^e concluded that the pump was operable on the basis of the above
reviews *" tests and attributed the variation in recirculation flow to the
multi-stage flow restricting orifice.
This device contained flow orifice
plates which had off-center orifices in two of the plates.
These plates were
free to rotate within the assembly, thus changing the hydrodynamic
characteristics of the device. With such a design, and based on historical
data, minor variations in recirculation flow could have occurred, which were
not a reflection of pump performance.
New baseline data was established and the pump was declared operable.
In
accordance with the ASME Code, the frequency of testing was doubled,
i.e., HPSI
3
Pump A would be tested every 45 days in lieu of every 90 days.
For the long
term, the licensee indicated that they were developing an alternate test method
for the HPSI pumps, which involved setting up a test loop which would bypass
the flow restricting orifice and provide a reference flow that was greater than
recirculation flow.
In this manner, reference flow would always be achievable
so that pump performance could be consistently evaluated in accordance with the
ASME Code.
This action would require procedure changes.
The inspectors
considered the above actions to be of sufficient depth to adequately address
the HPSI Pump A recirculation flow problem.
The licensee demonstrated
strengths in reflect bg lessons learned from recirculation flow problems
encountered on HPSI Pun.9 B in 1989.
See NRC Inspection Report
No. 50-382/89-09.
.
_ _ _ . _ _ . _ _ . _ _ _ . _ _ _ _ _ _ _ _ _ _ _ . _ _ _
_ _ _ _ _ _ _ _
_
_
_ -
.
- .- _.._ _.__
____
_ ._ _ _ _.__ _ _.
.
4
6-
3.2.3 Stop Screws Missing from Fire Isolation Switch
' On May 18,1931, with the plant in Cold Shutdown (Mode 5), postmodification
testing was being performed on Main Steam Isolation Valve (MSIV) A.
Part of
the testing included closing the valve using Fire Isolation Switch FR-4, in the
auxiliary relay room, to ensure that the MSIV control wiring changes made in
accordance with DC-3217 did not affect the function of the switch. The switch
was intended to be used in the event of a control room evacuation due to a
fire, and would provide control circuit isolation from the contiol room for the
MSIV and several other components and ensure that power is supplied to the
components to maintain them in the desired status. When the switch was
operated, it traveled past the isolate position, allowing two of the adjacent
make-before-break contacts to connect and short the positive side of one
circuit to the negative side of another circuit.
The short circuit caused the
power supply breaker for the switch to trip, but the switch was severely
damaged due to smoke and fire.
The fire was quickly extinguished.
During
replacement of the switch, it was determined that the stop screws that should
have been installed as part of the switch installation, were missing and
allowed the over-rotation of the switch,
The switch was a multi-deck rotary
switch manufactured by Electro-Switch.
It was determined that the stop screws
were supplied with the switch, but the licensee had failec to recognize that
the screws should have been installed. Several other isolatten , switches in
adjacent panels were inspected and at 19ast one other switch wat found to have
the stop screws missing. The licensee wrote a condition identificatior (CI)
report to inspect similar switches and to perform a root cause investigation.
The failed switch was replaced and successfelly tested.
The licensee was
considering notifying the industry via the nuclear repcrt: 1g network, and the
inspectors supplied pertinent data to NRC headquerters for consideration of
generic correspondence.
The licensee's actions wsre appropriate.
3.2.4
Inadvertent lift of Safety Injection Tank Relief Valves
On May 21, 1991, with the unit in Mode 3 at normal operating temperature and
pressure, the licensee entered the action statement for TS 3.5.1 when the
nitrogen relief valve on Safety Injection Tank (SIT) 1A lif ted and
depressurized the tank to approximately 300 psig. An attempt was made to
repressurize the tank, but the valve lifted again at approximately 500 psig.
SIT 1A was subsequently declared inoperable and RCS depressurization was
commenced to less than 1750 psla; a point where Mode 3 operation is allowed
with only three operable SITS maintaining pressure between 235 and 625 psig.
During RCS depressurization, the nitrogen relief lifted on SIT 18, rendering it
inoperable and placing the plant in TS 3.0.3.
The NRC Operations Center was
properly notified in accordance with 10 CFR 50.72. When RCS pressure was
reduced to less than 1750 psia, both TS 3.0.3 and TS 3.5.1 were exited.
SITS IA and IB were singularly isolated and vented and their relief valves
removed for_ inspection and setpoint checks.
The relief for SIT 1A actuated at
about 400 psig on the test bench, and the adjustment locknut was found loose.
It was reset to 700 psig (normal setpoint).
The SIT IB relief valve lifted at
about 710 psig and no adjustment was necessary.
Its locknut was found tight.
__
. .
_
_ _
.__ _ _
_ _ _ _ _ _ _ _
.
l
.
7-
When the relief valves were reinstalled, the tanks were repressurized to normal
operating pressure and monitored.
No additional problems were identified.
The
relief valves-are 3/4 x 1-inch Crosby spring-loaded, nozzle-type relief valves.
At the time the valves lifted, scaffold removal was in progress at both of the
SITS; however, the licensee was not able to initially demonstrate, with
significant confidence, the connection between the scaffolding activity and the
relief valve actuations.
The licensee reviewed logs and alarm princarts,
interviewed operators on duty at the time of tne event, and concluded that an
actual-over pressure condition had not occurred in the SITS. A spare relief
valve of the same type was tested under similar conditions as the SIT relief
valves, and it was demonstrated that mechanical agitation would cause the valve
to lift. Although the scaffold carpenters did not recall hitting the valves or
the tail pipes during scaf fold removal, the licensee concluded that it was
possible that the ,alves could have been inadvertently agitated during the
evolution.
No further problems were experienced with the valves.
The
licensee's actions to solve this problem were satisfactory.
3.2.5
Reactor Trip Breakers Fail to Close_Due to Operating Spring Anomaly
On May 20, 1991, during routine surveillance testing of the reactor trip
circuit breakers (TCB), TCB-4 would not reclose af ter it was tripped open.
Troubleshooting by the electrical maintenance department revealed that one of
two operating springs had become disconnected when the breaker tripped and thus
prevented the breaker from closing properly.
After receiving proper
authorization, the spring was reconnected and the breaker was satisfactorily
tested.
The inspector determined that this was the f ailure mode of General Electric
AK-Series circuit breakers described in NRC Information Notice (IN) 91-15 and
in a 10 CFR 21 report submitted by Maine Yankee on December 28, 1990.
The
inspector discussed the matter with electrical maintenance and system
engineering personnel and learned that they had just completed an inspection of
all of their AK-2-25 TCBs during the recent refueling outage. Based on the
reported TCB problems, they found the springs incorrectly oriented on four of
the eight breakers.
A previous work authorization had been implemented to
correctly install the springs on the four breakers, including TCB-4.
Further
review of the spring installation description in the 10 CFR 21 report and
IN 91-15 revealed that, without diagrams being included, the descriptions were
confusing and appeared to conflict.
The licensee inspected their breakers and
oriented the springs based on the 10 CFR 21. report and verbal discussions with
a vendor representative and concluded that they were installed correctly.
They
intended to contact the vendor with a written request for additional
information and possible design improvements, since it appeared that the
springs could become disconnected even when installed correctly.
In addition,
they were considering procedural changes which would require electrical
maintenance personnel to be present when the breakers were closed to verify
that the springs were properly in place. The apparent conflict between the
10 CFR 21 report and IN 91-15 was reported to cognizant NRC personnel.
The
inspectors will continue to follow up ca this issue until it is resolved.
The
TCBs were capable of performing their intended safety function regardless of
which way the closure springs were oriented.
-
- - _ - _ _ _ _ _ _ _ _ _ _ .
..
_ _ .
_____ _______________ ____
_
_
.
.
_ _ _ _ _ _ _
.
-8-
3.3
In-Office Review Of Licens2e Event Reports (LERsl__{90712J
The following.LERS were reviewed.
The inspectors verified that reporting
requirements had been met, causes had been identified, corrective actions
.ppeared appropriate, generic applicability had been considered, and that the
uER forms were complete. The inspectors confirmed that unreviewed safety
questions and violations of TS, license conditions, or other regulatory
requirements had been adequately described.
The Region IV staff determined
that an onsite inspection followup of the event was not appropriate.
The NRC
tracking status is indicated below.
3.3.1'(Closedl_LER 382/90-003, " Reactor Trip Due_to Grid Disturbance"
_(This LER was closed on the basis of the foflowup inspection of
IFI 382/9005-02 above)
3.3.2
[C_losed) LER 382/91-004, " Failure to Complete Technical Specification
Surveillance due to Inadequate Attention to Detail"~
4.
ONSITE RESPONSE TO EVENTS (93702}
4.1 Unusual Event Due to High RCS Leakage
On May 17, 1991, Waterford 3 declared an Unusual Event (UE) in accordance with
their Emergency Plan, due to an identified RCS leak rate of about 20 gpm. which
was in excess of the TS 3.4.5.2 limit of 10 gpm.
The plant was in Hot Shutdown
(Mode 4), approaching startup from the fourth refueling outage.
RCS temperature
was at 291*F and pressure at about 1600 psia with a steam bubble in the
pressurizer.
During pressurization, minor leakage had been reported coming
from the 4-inch pressurizer spray line check valve (RC-303). Because the
bonnet bolts and the seal had been replaced during the outage, the valve was
being monitored for leakage, with maintenance personnel checking torque on the
bonnet bolts every 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />. Typically, on this type of valve, leakage has
decreased or stopped with increasing pressure, because pressure tends to seat
the seal tighter.
After workers in the RCB reported that a popping sound
occurred, and that it was steam emanating from the pressurizer cubicle,
excessive RCS leakage was confirmed. The RCB was evacuated and operators
commenced depressurization of the RCS in accordance with their operating
procedures.
The differential pressure of the RCB atmosphere increased to about
10 inches water gauge (IWG) due to the leaking RCS flashing to steam, and when
nitrogen venting of the SITS was ccmmenced to reduce SIT pressure, RCB differential
pressure increased to about 20 IWG, with the TS limit being 27 IWG.
In order
to avoid having to purge the RCB, and to expedite depressurization, the shift
supervisor first consulted with plant management and then isolated the SITS
from the RCS.
This placed the plant in TS 3.0.3, because three SITS were
required to be operable by TS 3.5.1, and the action statements did not provide
for all SITS to be isolated.
Since the plant was already shut down and being
cooled down, the required actions of TS 3.0.3 were being met. When RCS
,
pressure was reduced to 392 psia, the pressure at which SITS were no longer
required to be operable, TS 3.0.3 was exited.
Cooldown was accomplished by
steaming to the main condenser until Shutdown Cooling Loop 2 was placed in
service. When the plant reached cold shutdown (Mode 5), the UE was terminated.
_.__ __
.
.
. - - -
.
.
-
-
-
-- .
.
..
.
. - - -
t,
.
+
.g.
The plant was depressurized and drained to 50 percent pressurizer level to
allow disassembly and repair of the valve.
RC-303 was a 4-inch Anchor Darling
i
swing check valve, and had a silver seal ring which was pulled in place _by the
bonnet bolts ~ The licensee disassembled the bonnet and found no damage, but
did find the bonnet seal cocked by about 54 mils, with two of the four bolts
loose,
This type of bonnet was typically installed by applying incremental
torque on the bolts.
No measurements were taken initially to verify alignment.
Subsequently, the licensee, in consultation with the vendor, developed a method
to measure al Snment.
This was applied to the reassembly of RC-303 on May 18,
when a new seal was installed.
Plant heatup and pressu-ization was commenced
on May 19, with no observed leakage from the new seal _ installation.
The NRC and local agencies were notified of the UE by the licensee in
accordance with their Emergency Plan, and pursuant to 10 CFR 50.72. At the
time of the notification, the licensee also informed the NRC headquarters duty
officer that TS 3.0.3 had been entered as a consequence of isolating the SITS.
The inspector was in the control room at the time of this event and observed
the decision process to isolate the SITS. During the event, it was suspected
that the RC-303 bonnet was leaking.
Not knowing the exact location of the-
i
leak, the shift supervisor stated that his decision to isolate the SITS was
based upon the need to expedite RCS depressurization and minimize the
possibility of the condition degrading to a more serious nature.
He was also
concerned that RCB pressure could reach a point that would procedunily require
'
purging the RCB, which would result in a release of activity to the atmosphere.
The shift supervisor consulted with licensee management present in the control
room, and was aware that if an automatic safety injection actuation signal was
received, the SITS would be capable of injecting into the RCS, because the
isolation. valves would open automatically. On May 18, the licensee informed
the NRC that the decision to enter TS 3.0.3 was made pursuant to
10 CFR 50.54(x), which, in part, provides allowance to depart from TS in an
emergency to-protect public health and safety.
This decision was questioned by
the NRC staff, because the circumstances did not appear to meet the intent of
The licensee subsequently concluded that 50.54(x) was not
utilized during this event.
The licensee informed the hRC of that conclusion
by letter dated June 17,-1991, which transmitted LER 382/91-08. The Basis of
TS 3.0.3 delineates the measures to be taken for circumstances not directly
provided for in the ACTION statements, and whose occurrence would violate the
intent of a specification. This was the case with isolating the SITS when_ they
were required to be operable pursuant to TS 3.5.1; however, it is unresolved as
i.
to whether or not the licensee should have invoked 10 CFR 50.54(x), and whether
l
it was a TS violation when the licensee voluntarily entered, and complied with
L
TS 3.0.3 to mitigate the potential consequences of an off-normal or minor
i
emergency condition that did_not appear to be an immediate threat to public
safety. The Region IV staff has requested assistance from NRR on this issue
'
-and will be tracked as Unresolved Item 382/9118-02.
4.2 Reactor Trip During Startup Testing
On May 26, 1991, with the reactor less than 0.1 percent power, a reactor trip
occurred resulting from an electrical spike during troubleshooting of a ground
L
.
.
. .-
.
_ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _
'
.
'
-10-
on the control room annunciator system. Operators were maintaining the plant
critical at low power and waiting for an emergent design change to be completed
on the process analog cabinets, before increasing reactor power.
Electrical
maintenance was requested to troubleshoot a ground on the annunciator system
while the other work was in progress.
The trip was initiated by high log power
on Channels C and 0 of the plant protection system, and it was concluded that
an electrical spike during the ground isolation procedure caused excore nuclear
instrument Channels C and 0 to increase above the trip setpoint of
.
0.257 percent power.
The plant was stabilized at hot standby using energency
.d.
and normal operating procedures. The NRC Operations Center was notified as
required by 10 CFR 50.72.
Corrective action was taken prior to restarting the
unit, which included hanging a caution tag on the annunciator cabinet and
warning personnel not to open the cabinets with the reactor critical at i.a
power without shift supervisor permission.
A change was to be made to the
electrical maintenance procedure to caution against isolating grounds with the
reactor critical at low power.
The emergant design change was completed and
the annunciator ground cleared.
The reactor was then restarted and startup
testing was resumed.
The actions taken during this event were appropriate, and
,
the operator response to the trip was excellent.
4.3 Reactor Trip Following Turbine Trip
On May 28, 1991, with the reactor at about 34 percent power, a reactor trip
resulted from a turbine trip when the 6.9 Kv Nonsafety Bus 381 failed to
transfer from the unit auxiliary transformer to the unit startup transformer.
Loss of power to the bus caused the IB and 2B reactor coolant pumps (RCPs) to
trip and resulted in a core protection calculator (CPC) auxiliary trip of the
reactor.
Thc unit was in startup testing after the recent refueling outage, and was
holding power for excore calibration and troubleshooting of the digital
i
electro-hydraulic (DEH) control system.
Problems had been experienced earlier
in the day with load swings when attempting to transfer from single valve to
3
sequential valve control on the DEH,
At the time of the trip, the DEH was in
manual valve control while maintenance technicians were running diagnostic
routines on the DEH computer and software,
Shortly before the trip, Governor
Valve (GV) No. I began to open causing turbine load to increase.
The operators
decided to manually trip the turbine, and the reactor trip followed when the
bus transfer failure caused the loss of two of the four operating RCPs.
Following the trip, systems responded normally except that the lack of decay
heat and only two RCPs running resulted in a cooldown to approximately 510 F.
When steam pressure dropped to 764 psia, an automatic main steam isolation
signal (MSIS) occurred.
Prior to the MSIS, attempts were made to throttle feed
flow from the operating main feed pump, which was thought to be a major
contributor to the cooldown.
Once the 381 bus was manually reenergized, the
IB RCp was restarted and the plant was staoilized at hot standby.
The MSIVs
were left shut to allow repair of a steam leak that was identified earlier on
'
the HP turbine.
The cause of GV No. 1 oscillation was found to be an error in
the OEH diagnostic procedure, which failed to specify j'sconnecting the valve's
servo driver card from its digital / analog interface card prior to running the
diagnostic test.
This allowed a signal to be imposed over the manuci signal to
_
___ _ _ _ _ _- _ - _ _ _ _ _ _ - _ _ _ _
-
..,
-11-
the GV, causing it to open.
The cause of the failed bus transfer was found to
be a faulty contact in a follower relay (25x relay) for the synchronous check
relay which prevented the shut signal from reaching the feeder breaker from the
unit startup transformer to the 381 bus.
The 25x relay was replaced.
The unit
was restarted at 1:23 a.m. on May 29, after repair of the HP turbine steam leak
,
i
and replacement of the faulty relay.
Extensive troubleshooting of the DEH with
the unit shut down could not identify the original problem with DEH sequential
valve control, so a plan was developed by the licensee to continue
l
troubleshooting with the turbine online. With the aid of the vendor a ground
was subsequently found, on one position of the DEH test select switch, and
corrected. The DEH was transferred to sequential valve mode without any
problems and startup testing was continued.
Based on the licensee's root cause analysis, the 25x relay failed due to a
'
plastic part disintegrating.
The pieces fell into the relay contactor and
prevented-the contacts from functioning.
The failure mechanism appeared to be
o
!
aging, possibly accelerat'ed by heat.
The 25x relay was manufactured by
!
Struthers-Dunn, Model 602-4-351.
There were also identical relays installed in
!
the plant which were labeled Model 602-4-1007.
The licensee had 348
Struthers-Dunn relays of different types installed-at the plant (nonsafety) and
25 of them were. identical to the one described above that failed. The licensee
,
l
had an exact replacement spare in stock.
Since Struthers-Dunn no longer
L
supplied these relays, the licensee was seeking an alternative source.
The
!
l
licensee implemented a work authorization to inspect similar relays in the
.
other three busses (3A1, 3A2, and 382) for possible defects, so that fast bus
l
transfer won't be impacted again.
They were developing long-term actions, such
l.
as a review of applications of this relay to other circuits in the balance of
I-
plant that ceuld cause unwanted reactor transients upon failure.
Conclusions:
No violations or deviations were identified.
The operator's
,
!
response to events was always timely and appropriate; follewing their
emergency, off-normal, and operating procedures.
One unresolved item was
identified concerning -the voluntary entry into TS 3.0.3- to expedite
d6 pressurization of.the RCS to stop an unisolable leak.
5.
MONTHLY MAINTENANCE OBSERVATION (62703)
l
l:
The station maintenance activities affecting safety-related systems and
L
-components listed below were observed and documentation reviewed to ascertain
l
that the activities were conducted in accordance with approved work
l
authorizations (WAs), procedures, TS and appropriate industry-codes or
standards.
'
l
l
5.1 WA 01058865, 01065885
l
l
On May 21, 1991, the inspector observed work in progress on turbine driven
l_
emergency feedwater (EFW) Pump A/B.
The governor had been replaced and the
,
~
control oil system cleaned to ensure no problems would develop with the turbine
i
control system due to possible contaminants in the oil.
The mechanical
l
overspeed device trip setpoint was being calibrated in accordance with
(
Procedure MM-03-016, Revision 0, " Emergency Feedwater Pump Turbine Mechanical
l
l
l
-
_ ___ - _ _ _ _
'
.
1
'
l
-12-
l
l
t
Overspeed Trip Test and Calibration."
Simultaneously, the turbine speed
control circuitry was being calibrated in accordance with M1-5-425, Revision 2,
,
" Emergency Feedwater Pump Turbine Governing Control System Calibration." The
inspector verified that the work had been properly authorized and was being
performed by qualified individuals in accordance with approved work
instructions.
Equipment operation was being performed by a qualified operator
as directed by the control room.
The inspector learned that an unsuccessful attempt had been made to test run
the pump turbine on the previous day.
Troubleshooting by mechanical
maintenance personnel revealed the oil pump for the replacement governor that
had been installed was not configured to rotate in the direction required by
tnis turbine (i.e., clockwise).
The result was that no oil pressure was being
supplied to the governor valve servo.
The matter was discussed with
maintenance personnel and it was determined that the turbine manufacturer
supplied two different Woodward
vernors for their turbines with either
clockwise (CW) or counterclockwise (CCW) rotating oil pumps. The only
difference in the part numbers was the letter "A"
prefix on the CCW units.
Each governor also has a rotation direction arrow placed on the outside of the
case.
Maintenance personnel originally began governor installation with the correct
governor, as specified in the work instruction. When the old governor was
removed, maintenance personnel observed that the rotation directinn arrow did
not match the arrow for the new governor. Guidance was requested from
maintenance engineering and, after it was found that the second governor in the
warehouse had a rotation direction arrow which matched the installed unit, the
decision was made incorrectly to install the second governor.
This was later
determined to be the wrong unit when the turbine's governor valve would not
function.
It appeared that the installed governor was originally an incorrect
CCW unit that had been field modified during a previous governor repair to
operate as a CW unit, and had the
"A" prefix scratched off of the part number
without reversing the direction of rotation arrow.
There was a second,
incorrect CCW unit in the warehouse that was never identified and properly
dispositioned.
Since the installed ;overnor actually was turning CW but still
had a CCW rotation arrow, the dec'-ion was made, incorrectly, to install the
CCW unit. When the error was ditcovered, after the turbine governor valve
failed to operate properly, the correct CW unit was installed and the turbine
was successfully tested.
Maintenance engineering indicated that the two spare
governor units will be sent to the vender to be refurbished and correctly
ported for CW rotation and then returned to the warehouse for future use.
This
matter will remain unresolved until it can be determined if the field
modification of the previously installed unit was properly authorized and
documented correctly and why the other incorrect spare governor unit in the
warehouse was not properly dispositioned. Additionally, the licensee indicated
that they will address why their quality assurance program did not prevent
installing the incorrect governor during this maintenance activity (Unresolved
Item 382/9118-03).
Conclusions: Although no violations or deviations were identified, the errors
identified during the maintenance on the A/B EFW pump demonstrates a potential
_ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ - -
-
_ _ _ _ _ _ _ _
.
.
.
.
-13-
weakness in a quality aspect of maintenance.
It was unclear why incorrect
parts had been procured for the EFW pump turbine and actually installed on two
different occasions. After the recent occurrence, the error was revealed by
the licensee's maintenance retest program before the pump was returned to
service,
Therefore the safety significance of the error was minimal.
6.
BIMONTHLY SURVEILLANCE OBSERVAT10N__(61726)
The inspectors observed the surveillance testing of safety-related systems and
components listed below to verify that the activities were being performed in
accordance with the TS.
The applicable procedures were reviewed for adequacy,
test instrumentation was verified to be in calibration, and test data was
reviewed for accuracy and completeness.
The inspectors ascertained that any
deficiencies identified were properly reviewed and resolved.
6.1 Procedure OP-903-008, Revision 3, " Reactor Coolant System Isolation
Leakage Test"
On May 22, 1991, the inspector observed the performance of Section 7.4.8 of
OP-903-008 for Valve SI-142B.
The test was performed to satisfy TS
surveillance requirement 4.4.5.2.3.
The test was properly authorized and
performed by qualified individuals in accordance with an approved procedure.
Good coordination was maintained during the test between the control room and
the containment where the leakage was being collected and measured.
The test
acceptance criteria and TS requirements were met for the valve.
No problems
were identified.
6.2 Procedure NE-02-020, Revision 2, " Control Element Assembly (CEA)
Insertion Time Measurement"
On May 22 and 23, 1991, the inspector observed the performance of NE-02-020 to
meet the TS surveillance requirements 4.1.a.3 for CEA reed switch position
transmitter calibration and functional testing and 4.1.3.4 for CEA drop timing.
The tests were properly authorized and performed by qualified individuals in
accordance with an approved procedure.
The inspector verified that the initial
conditions and prerequisites specified in the test procedure were met.
The
test acceptance criteria and TS requirements were both met. No problems were
identified.
6.3 Procedura OP-903-101, Revision 4, "Startup Channel Functional Test"
On May 28, 1991, the inspector observed the performance of OP-903-101, for
nuclear instrument Startup Channel 2, in preparation for a reactor startup
planned for later in the day.
The test was properly authorized and performed
by a qualified individual in accordance with an approved procedure.
The test
acceptance criteria were met and no problems were identified.
Conclusions:
No violations or deviations were identified. The licensee's-
surveillance program continued to perform well.
The licensee expended
considerable ef fort and utilized comprehensive checklists and administrative
controls following the recent refueling, to ensure that the required
i
. -
-
._
-
- _ .
.
-
_ - -
.-
.
_14
surveillances were performed satisfactorily prior to entering operating modes
where the equipment was required to be operable.
7.
OPERATIORAL SAFETY VERIFICATION (71707)
The objectives of this' inspection were to ensure that this facility was being
operated safely and in conformance with regulatory requirements and to ensure
that the licensee's management controls were effectively discharging the
licensee's responsibilities for continued safe operation.
The inspectors conducted control room observations, reviewed logs and reviewed
the equipment out of service log, and noted that plant conditions and equipment
status were being kept up to date.
TS limiting conditions for operation and-
action statements were being followed in an exemplary manner as the plant was
taken through startup and physics testing from the refueling outage.
The
inspectors observed reactor startups following the refueling, noting that
procedures were properly utilized and were current.
The inspectors conducted plant inspection tours on a daily basis, and found
that housekeeping was improving rapidly with the conclusion of the outage.
After the plant had been at full power for just a few days cleanup was
virtuall" completed. The. inspectors did note, however, that the boric acid
makeup' tank rooms did not -improve as a result of the outage.
,;.ere were still
large quantities of unused heat trace and sensing wires draped on the piping
and boric acid deposits were observed in numerous places.
This was brought to
,
the attention of licensee management, who was already aware of the condition
i
from previous tours.
The inspectors attended plan-of-the-day meetings and noted continued excellent
coordination and communications between organizations.
,
l
On May 16, 1991, the inspectors conducted an-inspection tour of the RCB to
'
assess conditions prior to heatup of the plant. _The emergency core cooling
sump was inspected to ensure that no debris was present which could block the
sump screens during an accident, and that the-sump baskets contained sufficient
pH additive as required by .S 4.5.2.d.3
The RCB was generally in good
-
condition but a list of minor discrepancies was provided to the licensee to be
corrected prior to plant startup. The items were corrected as appropriate.
On June 4,'1991, the inspectors performed a test of the Emergency Notification
System in-the control Hoom to confirm operability.
The results were
satisfactory.
On June 4, 8, and 13, the licensee experienced reactor power transients
resulting from a reduction of feedwater heating caused by the inadvertent
,-
isolation of extraction steam (ES) to the high pressure feedwater (FW) heaters.
The ES valves were designed to modulate extraction steam flow and shut
automatically when a high water level is present in the Rd heaters.
The June 4
and 13 events were associated with Ct to the No. 1 FW heaters and the June 8
event resulted when ES was isolate. so the No. 2 FW heaters.
The loss of FW
heating caused-reactor power to increase from 100 percent to as high as
.-
-
_
.
. _ . _ . . _ _ _ . .
___
_ _ _ _
._
_
.
.
,
.
_ _ . . _
__
.
,
-15-
104 percent for about 5 minutes.
The operators took prompt act
power to within the licensed limit in each case by reducing tur
reestablishing ES to the FW heaters.
Although such transients
under t'le circumstances of this incident, NRC does not consider
violation of the licensed steady state power limit of 100 perce
power.
The first two transients were attributed to work that was being
the vicinity of the FW heaters causing physical jarring of the
which actuated the ES isolation.
It was also determined that n
switch mechanisms that were installed in the Magnetrol level co
were very susceptible to actuation by normal system vibration a
physical contact.
The licensee had replaced the older dry cont
mechanisms with the mercury-type mechanisms during the recent r
because of premature aging and heat damage which made the dry c
unreliable.
Initial corrective action for the transients consisted of train
personnel and erecting barriers around the FW heater level cent
physical contact with +he switches.
Since the third transient
associated with actual physical contact, it was concluded that
switch mechanisms were unacceptable, and the vendor was contact
assistance.
The motor operator for the ES valve to the No. 1 F
left deenergized after the third transient.
This was done as a
until the problem could be resolved in order to prevent another
With the assistance of the vendor, the licensee obtained a diff
mercury-type switch mechanisms that were specifically designed
high vibration applications, and intended to install the new sw
when a spare parts equivalency evaluation report (SPEER) could
work instructions prepared.
The inspectors will continue to moni
activity.
Conclusions: No violations or deviations were identified.
The 1
continued to operate the plant in a safe and professional manner.
j
response to the repeated feedwater transients was considered go~-
taken to prevent further transients were appropriate. Housekee
,
at this plant continue to be excellent.
8.
EXIT INTERVIEW
The inspection scope and findings were summarized on June 20, I
persons indicated in paragraph 1 above.
The licensee acknowled
inspectors' findings. The licensee did not identify as proprie
material provided to, or reviewed by, the inspectors during thi
l
.
.
.-
-_
-. .-
.
.
. - - - . .