ML15224A691
| ML15224A691 | |
| Person / Time | |
|---|---|
| Site: | Oconee  |
| Issue date: | 06/19/1990 |
| From: | Binoy Desai, Shymlock M, Skinner P, Wert L NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML15224A689 | List: |
| References | |
| 50-269-90-12, 50-270-90-12, 50-287-90-12, NUDOCS 9006290229 | |
| Download: ML15224A691 (17) | |
See also: IR 05000269/1990012
Text
UNITED STATES
NUCLEAR REGULATORY COMMISSION
REGION II
101 MARIETTA STREET, N.W.
ATLANTA, GEORGIA 30323
Report Nos:
50-269/90-12, 50-270/90-12, 50-287/90-12
Licensee: Duke Power Company
P.O. Box 1007
Charlotte, N.C. 28201-1007
Docket Nos.:
50-269, 50-270, 50-287
License Nos.:
Facility Name:
Oconee Nuclear Station
Inspection Conducted: A ril 22 - May 19, 1990
Inspectors
,
P.
Skinner, Senior esi ent Inspector
Date
L. D. Wert, Residen Inspector
Date Signed
B. B. Desai, Resident Inspector
Date Signed
Approved by:
_
__
___/
M. B. Shymigck, Section Chief
Date Signed
Division of Reactor Projects
SUMMARY
Scope:
This routine, announced inspection involved inspection on-site in
the areas of operations including degraded grid voltage issues,
surveillance testing, maintenance activities, seismic monitoring,
outage activities, Notice of Unusual Event for a chemical spill and
inspection of open items.
Results:
One Violation was cited during this report period:
-
One violation addressed a failure to incorporate design basis
information involving degraded voltage protection
into
electrical relay procedures (paragraph 2.c).
-
Another violation addressed two examples of failure to follow
procedure.
These
examples
involved inadequate component
verification which resulted in incorrect component removal
(paragraph 4.c).
They are additional examples of a failure to
follow
procedures
addressed
in Inspection
Report
50-269,270,287/90-16.
The licensee's outage management continues to be very aggressive.
However, this aggressiveness may in part have contributed to some of
the activities which created the concerns discussed in this report.
The controls exhibited by all of the plant staff during the spill
from the spent fuel pool was expeditious and effective.
..*
- j
.,
REPORT DETAILS
1. Persons Contacted
Licensee Employees
- B. Barron, Station Manager
D. Couch, Keowee Hydrostation Manager
T. Curtis, Compliance Manager
- J. Davis, Technical Services Superintendent
D. Deatherage, Operations Support Manager
R. Dobson, Electrical Engineering Manager
- B. Dolan, Design Engineering Manager, Oconee Site Office
W. Foster, Maintenance Superintendent
D. Hubbard, Performance Engineer
D. Jamil, Electrical Systems Engineering Supervisor
- E. LeGette, Compliance Engineer
H. Lowery, Ch-airman, Oconee Safety Review Group
B. Millsap, Maintenance Engineer
- 0. Powell, Station Services Superintendent
- G. Rothenberger, Integrated Scheduling Superintendent
- R. Sweigart, Operations Superintendent
Other licensee employees
contacted included technicians,
operators,
0
mechanics, security force memblers, and staff engineers.
NRC Resident Inspectors:
- P. Skinner
- L. Wert
B. Desai
- Attended exit interview.
2.
Plant Operations (71707)
a. The inspectors reviewed plant operations throughout the reporting
period to verify conformance with regulatory requirements, Technical
Specifications (TS), and administrative controls.
Control room logs,
shift turnover records, temporary modification log and equipment
removal and restoration records were reviewed routinely. Discussions
were conducted with plant operations, maintenance, chemistry, health
physics, instrument & electrical (I&E), and performance personnel.
Activities within the control rooms were monitored on an almost daily
basis.
Inspections were conducted on day and on night shifts, during
weekdays and on weekends.
Some inspections were made during shift
change in order to evaluate shift turnover performance.
Actions
observed were conducted as required by the Licensee's Administrative
Procedures.
The complement of licensed personnel on each shift
2
inspected met or exceeded the requirements of TS.
Operators were
responsive to plant annunciator alarms and were cognizant of plant
conditions.
Plant tours were taken throughout the reporting period on a routine
basis. The areas toured included the following:
Turbine Building
Auxiliary Building
CCW Intake Structure
Independent Spent Fuel Storage Facility
Units 1, 2 and 3 Electrical Equipment Rooms
Units 1, 2 and 3 Cable Spreading Rooms
Units 1, 2 and 3 Penetration Rooms
Station Yard Zone within the Protected Area
Standby Shutdown Facility
Units 1, 2 and 3 Spent Fuel Pool Rooms
Keowee Hydro Station
Unit 1 Reactor Building
During the plant tours, ongoing activities, housekeeping, security,
equipment status, and radiation control practices were observed.
Unit 1 operated at 100 percent full power from the beginning of this
report period until April 26 when the unit was taken off-line to
begin the end of cycle 12 refueling outage.
The unit was in the
outage for the remainder of the reporting period.
Unit 2 and 3 operated at 100 percent for the duration of this report
period.
b.
Reduced Inventory Activities (71707)
The inspectors completed the actions required by the Midloop/Reduced
Inventory Activities checklist (promulgated by L. Reyes on April 11,
1990).
the licensee's responses and
Inspection Reports 269,270,287/89-17 and 89-25 which documented the
inspection required by TI 2525/101,
were reviewed.
The inspectors
observed that operators displayed a high level of attention and
focused on meticulous procedural compliance when proceeding to and
operating in midloop conditions.
OP/1/A/1103/11: Draining and
Nitrogen Purging of the Reactor Coolant System (RCS) contains most of
the administrative controls committed to in the GL 88-17 response.
This procedure specifically provides guidance for containment closure
capability, RCS temperature and level indications, inventory addition
paths and alternate power supplies as required by the checklist. The
inspectors noted that in addition to Oconee's installed reactor
vessel level instrument (LT-5),
two temporary ultrasonic level
detectors had been installed and provided a limited range of RCS
level indication in the control room.
One of these detectors was
mounted on an RCS cold leg, the other on a hot leg.
Information
3
provided by these indicators was very useful to the operators during
the draindown evolution.
Additionally, these level indications,
along-with information from the Reactor Vessel Level Indicating
System (RVLIS)
helped resolve a long standing problem involving
inadequate RCS venting during draining.
Apparently the venting
problem is caused by a particular type of valve (Kerotest) installed
in the vent flowpaths.
These valves will be changed to another type
of valve during this outage.
Procedure changes will be made to
ensure adequate vent paths exist during draining.
The inspectors
verified portions of the alternate inventory addition flowpaths and
insured that no work was in progress which knowingly could lead to
perturbations to the RCS or to systems necessary to keep the RCS
stable.
Particular attention was given to electrical system lineups
and evolutions.
During the report period the licensee identified one instance of an
apparent failure to meet a GL 88-17 response commitment.
The
licensee committed to maintaining two makeup paths available (during
reduced inventory operations)
in addition to the Low Pressure
Injection pumps.
Since one of the required available alternate
makeup flowpaths is gravity flow from the Borated Water Storage Tank
(BWST)
via valves 1LP-21 or 1LP-22, a minimum level of 46 feet is
required to
be maintained in the
BWST.
Enclosure 4.7 of
OP/1/A/1103/11: Draining and Nitrogen Purging of the RCS, lists these
requirements as prerequisites to lowering Reactor Vessel (RV)
level
to less than 50 inches on LT-5. These conditions had been met prior
to entering reduced inventory conditions. During filling of the Fuel
Transfer Canal (FTC) from the BWST on May 3, the level was allowed to
decrease to less than 46 feet.
PT/1/A/0600/01:
Periodic Instrument
Surveillance requires verification of this level once per shift when
in reduced inventory conditions. A Problem Investigation Report was
generated to address this issue.
The 46 feet level is the level
required by TS 3.3.4 for normal operations. BWST level of less than
46 feet may not make the BWST makeup path inoperable. It should be
noted that AP/1/A/1700/07:
Loss of Low Pressure Injection System
lists this path to be utilized only if the primary alternate path is
not operable (Bleed Holdup Tank path).
Additionally, Unit 1 was
maintaining one additional alternate LPI pump suction path available
at this time.
The inspectors will follow the licensee's actions on
this issue. Once resolution of a minimum BWST level is determined it
is expected that training and more stringent procedural controls will
be utilized to ensure BWST level is maintained as required.
c. Degraded Grid Issues (71707)
Inspection Reports 50-269,270,287/90-10 and
11 discuss an issue
regarding degraded grid protection which was initially identified in
March 1990 by Design Engineering (DE) as part of the ongoing Design
Basis Documentation
(DBD)
program.
That issue involved past TS
violations and a potential single failure vulnerability in degraded
grid voltage conditions due to the startup transformer breakers (the
4
"E" breakers)
opening on undervoltage
(UV).
At that time the
licensee stated that the plant was protected from degraded grid
voltage by this undervoltage feature.
On April 23,
1990,
the
inspectors were informed by the licensee that the undervoltage
setting on the breakers was not properly set.
Due to a 3 percent
tolerance in the sensing relays and a nonconservative setpoint, even
if switchyard voltage decreased to approximately 207 KV, the breakers
may not have opened.
The minimum acceptable grid voltage level to
insure all safety related loads would be supplied with adequate
voltage under Loss of Coolant Accident (LOCA) conditions is 219 KV.
(The
208 V level safety-related valves are the limiting loads,
information on their performance under degraded voltage conditions is
limited.)
The inspectors questioned if this issue had been reviewed
for reportability in accordance with 10 CFR 50.72 since this is
potentially a more significant issue than the previously identified
problem. With the existing UV setting, if a degraded grid condition
existed and a LOCA occurred (inrush of Emergency Safeguard (ES) loads
actuating), inadequate voltages could be applied to the ES loads. On
April 24,
1990, it was determined that the problem was reportable in
accordance with 10 CFR 50.72(b)(2)(iii)(D) and notification was made.
The interim corrective actions for the previous problem (see
Inspection Report 50-269,270,287/90-11) appear sufficient to ensure
that continued plant operation under degraded grid conditions will
require initiation of extensive actions.
The licensee is continuing
progress on the urgent Nuclear Station Modification
(NSM)
as
described in Inspection Report 50-269,270,287/90-11.
A separate
10 CFR 50.73 report will be submitted concerning this latest issue.
In further review of this issue, the inspectors reviewed the past ten
years of procedure RTP/0/A/4980/27A:
Routine Test Procedure:
Type
CV-7
Relay.
This procedure is used by
Transmissions department personnel to periodically verify the proper
operation of the UV relays involved in this issue.
These relays
sense the voltage which is available from the switchyard.
Their
function is to open the normal
(N) or emergency (E) breakers and
block closure of the breakers if voltage is too low.
The following
information was noted:
-
The original procedure (1976) required the UV relays to be set
at approximately 68 percent of 4160V (2829V).
-
In 1977,
as a result of a DE evaluation, the settings were
increased to approximately 88 percent (3660V) to "ensure minimum
voltage required for continued operation of non-safety and
safety related equipment" (degraded voltage protection).
This
88 percent setpoint corresponds to approximately 212 KV in
switchyard voltage (conversion includes tap setting
and
transformer loss approximations).
-
In June of 1980 the settings were decreased to about 77 percent
3203V) to comply with a DE study concerning degraded voltage
5
operation of unit auxiliaries.
This was done to prevent the
opening of the "E" breakers on UV in certain scenarios if two
Oconee units were sharing a single startup transformer.
A commitment was made in correspondence dated February 5, 1982,
to the NRC to return the setpoints to 88 percent since that was
the minimum analyzed value.
A TS change was also made which
prohibited connection of two units to one startup transformer
except for short periods of time.
However,
the inspectors
identified that the UV setting actually remained at 77 percent
until 1985 when it was returned to 88 percent for an unrelated
reason.
This discovery was immediately discussed with both DE
and station management.
Further review of this issue was conducted including review of
portions of the licensee's response to and correspondence relating to
a Generic Letter issued in June 1977 addressing degraded grid
protection. The following observations were noted;
-
Oconee does not meet all of the recommended actions of Branch
Technical Position PSB-1:
Adequacy of Station Electrical
Distribution System Voltages.
The proposed modification being
initiated for this problem will not result in Oconee being in
full compliance with PSB-1 since with a degraded grid a
subsequent occurrence of an ES signal will not automatically
separate the class
1E
system from the offsite system.
Additionally,
no TS requirements for the degraded voltage
protection system exist.
While the External Grid Trouble
Protection System (EGTPS) (designed to actuate on a loss of the
grid )is not specifically addressed in TS,
the switchyard
isolation circuitry is an inherent support system of the Keowee
overhead path and, therefore, its operability is required by TS.
-
The formal response to the GL did not specifically list the UV
feature of the E breakers as degraded grid protection for
Oconee, however, other related correspondence did.
-
The role of these UV relays in degraded grid protection has not
been sufficiently acknowledged in the past.
-
The setting of the relays was not properly controlled and they
were set well below both the value specified by DE and the value
required to accomplish their safety function. Although lack of
formal control of relay settings in general has been previously
recognized by the licensee as a weakness, the formal setpoint
documentation is still being developed. Longterm intentions are
to review each vital relay setting to ensure the relay is
properly set after the setpoint document is developed.
The
inspectors were concerned that since the E and N breaker UV
settings, were.in error, other critical relay settings may also
be in error.
6
The recently completed Emergency Power Switching Logic (EPSL)
Self Initiated Technical Audit (SITA)
reviewed the historical
relay test data sheets on these CV-7 UV relays, but the improper
settings were not noted.
The SITA team did make an observation
concerning the CV-7 relays tendency to drift from setpoint and
to be vibration sensitive. Since the DBD review of EPSL had not
yet been completed, it
would have been difficult for the SITA
team to discover the incorrect settings.
On April 26,
1990, discussions were held with NRR and the licensee
regarding both of the two recently identified degraded voltage issues
and the proposed modification to correct the problems.
On May 2,
1990, the
inspectors met with station management and DE
representatives to further discuss the improperly set relay issue.
The following key points were included;
With the UV relays set at their current value (87.5 percent of
rated bus voltage which is about 209 KV switchyard voltage), if
the relay had actuated 3 percent below the setting under
conservative assumptions and simultaneous ES actuation occurred,
then voltage supplied to the 208 V safety-related valve motors
may not have been adequate.
This could result from the 3
percent tolerance and the nonconservative setpoint of the
relays.
DE stated that the 4160V and 600V ES loads would have
sufficient voltage to operate under these conditions.
Additionally DE representatives verified that there would be no
problem with normally running loads if grid voltage decreased to
these existing settings.
-
In order to provide absolute certainty that all ES loads under
the most conservative assumptions
would receive adequate
voltage, the existing UV relays would have to be set above 92
percent of 4160V rated bus voltage.
This corresponds to about
219 KV switchyard voltage.
-
With the relays set at 77 percent (1980 to 1985 settings) the
plant was not protected from degraded grid voltage conditions.
The 77 percent setpoint corresponds to about 195 KV in the
If this voltage was supplied to the main feeder
busses under LOCA conditions, many ES loads would not start and
even could be damaged. There may also be some concern regarding
normally running loads if grid voltage actually decreased to the
77 percent setpoint.
-
The lowest value of degraded grid voltage recorded at Oconee was
a 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> period at 208 KV on the yellow bus in 1982. Other than
that period, switchyard voltage has been above 219 KV.
-
The inspectors expressed concern that other relay settings
throughout Oconee's electrical systems may be different than DE
setpoints. The licensees longterm corrective action involving a
7
formal relay setpoint document and subsequent verification of
each setting may not be adequate if
other critical relay
settings are presently set incorrectly. The licensee committed
to comparing some transmissions procedure relay setpoints to
those specified/assumed in DE documentation.
A total of 44
safety significant overcurrent relays were checked and no
discrepancies found.
The differential current relay settings on
the Main Feeder Busses (considered an important protective
feature) have also been verified as current. Apparently this is
a time consuming process with the major effort required to
determine the DE documented setpoints.
The issue of the 3 percent tolerance band and nonconservatively set
UV relays was identified by the licensee as a result of the DBD
effort.
While the significance of the issue and the discovery of
relay settings lower than the DE values were the results of the
inspectors interactions,
the inspectors would probably not have
examined this area if
not prompted by the licensee's previous
discoveries.
The temporary corrective actions listed in Inspection
Report 50-269,270,287/90-11 appear to be adequate to address the
degraded grid issue sufficiently until the proposed modification is
installed. The issue of improper control of relay settings remains a
concern.
Incorporating design basis requirements into implementing
- procedures
is essential to safe plant operation and is required by
Criterion III of 10 CFR 50, Appendix B. The failure of the licensee
to incorporate design basis requirements into implementing procedures
is identified as Violation 50-269,270,287/90-12-01:
Failure to
Incorporate Design Basis
Information
Into Electrical Relay
Procedures.
d. Unit 1 Reactor Protection System Actuation
At 7:32 p.m. on April 26, 1990, Oconee Unit 1 experienced a Reactor
Protection System (RPS) actuation.
RPS channels A, C, and D tripped
on number of Reactor Coolant Pumps (RCPs)/flux signal.
The unit was
in hot shutdown conditions cooling down to cold shutdown. Group one
rods were at fifty percent withdrawn (as required by procedure) to
ensure the availability to add negative reactivity during the
cooldown. In accordance with step 2.1 of Enclosure 4.2, Hot Shutdown
Conditions to 250 degrees F/350 PSI conditions, of OP/1/A/1102/10:
Controlling Procedure for Unit Shutdown, dated 2/10/89, the operators
secured 2 of the 4 running RCPs (1 in each loop).
The resultant 2
RCPs running condition initiated the RPS actuation.
Group One rods
.tripped and dropped into the core.
On December 29,
1989,
TS 2.3 (Limiting Safety System Settings,
Protective Instrumentation) had been amended to prohibit operation
above zero percent reactor power with 2 or less running RCP's.
Accordingly, Instrument and Electrical (I&E) procedures were revised
and the RPS setpoints were changed to ensure compliance with the
requirement (See Inspection Report 269,270,287/89-40).
Apparently
8
small fluctuations in the Nuclear Instrumentation circuitry caused
the RPS to receive a reactor power signal greater than zero. (Actual
power level was 200 cpm on source range).
As expected, with a
reactor power signal greater than zero and only 2 RCPs running, the
RPS was actuated. OP/1/A/1102/10 was revised to permit three RCP
operation until RCS pressure was reduced to 1700 psi where shutdown
bypass could be actuated which blocks this trip. Group One rods were
withdrawn and the cooldown continued.
I&E is investigating the issue
and will be proposing permanent action to resolve the issue.
e. Overflow of the Unit 1 and 2 Spent Fuel Pool
At-approximately 9:15 a.m. on May 17, 1990, the licensee informed the
resident inspectors that the Spent Fuel Pool
(SFP)
had overflowed
resulting in contamination of various areas in the plant including an
area in the Unit 1 and 2 combined control room.
It was determined
that the spill resulted from performing steps out of sequence in
OP/1/A/1102/15, Filling and Draining of the Transfer Canal.
Performing the steps out of sequence resulted in shutting the
isolation valves between the SFP and the transfer canal with the SFP
pump still taking a suction from the transfer canal and discharging
into the SFP.
The details associated with this spill and review by
the NRC are contained in Inspection Report 50-269,270,287/90-16.
3. Surveillance Testing (61726)
a. Surveillance tests were reviewed by the inspectors to verify
procedural and performance adequacy.
The completed tests reviewed
were
examined for necessary test prerequisites,
instructions,
acceptance criteria, technical content, authorization to begin work,
data collection, independent verification where required, handling of
deficiencies noted,
and review of completed work.
The tests
witnessed, in whole or in part, were inspected to determine that
approved procedures were available, test equipment was calibrated,
prerequisites were met, tests were conducted according to procedure,
test results were acceptable and systems restoration was completed.
Surveillances reviewed and witnessed in whole or in part:
PT/1/A/600/22 Motor Driven Emergency Feedwater Pump Suction Check
Valve Test
IP/2/A/305/3C RPS Channel 'C' On-Line Test
PT/2/A/115/08 Rx Building Containment Isolation and Verification
b. Unit 1 ES Actuation During Nuclear Station Modification (NSM) Testing
On May 16, 1990 at 2:18 p.m., an inadvertent automatic actuation of
the Engineered Safeguards
(ES)
channels 1 through 6 occurred on
Unit 1. The unit was in a refueling outage with fuel loading in
progress.
A low pressure injection (LPI)
pump,
both Keowee hydro
units,
and several
ES valves actuated as expected.
Refueling
9
operations were immediately stopped by the Refueling Senior Reactor
Operator.
The cause of the ES actuation was a procedure deficiency.
NSM
Procedure
TN/1/A/2682/OO/AK1,
Replacement of Existing
Reactor
Building Pressure Switches was being performed.
The ES system is
designed such that the three analog channels receive signals from
various Reactor Coolant System pressure transmitters and Reactor
Building pressure transmitters.
When the measured parameter reaches
a certain setpoint it
trips the associated analog channel.
If
2-out-of-3 analog channels trip, the digital ES channel is tripped.
As per the procedure, the trips associated with replaced pressure
switches (PS)
21 and 23 were blocked to accommodate testing, which
prevented digital channels 7 and 8 from actuating.
However, the
procedure did not block the signal from the pressure transmitters
associated with digital channels 1 through 6 from actuating during
testing.
When pressure was introduced to test PS 21,
"B" analog
channel tripped.
This trip was not reset by the operators.
Later
when PS 23 was tested, "C" analog channel tripped.
This fulfilled
the required 2-out-of-3 trip logic for the ditigal channels and
actuated ES digital channels 1 through-6.
Coincident with this testing,
I&E personnel were performing ES
calibration on the other ES circuits.
Analog channel "B" was not
reset by the control room operator since the trip condition was
attributed to the ongoing calibration.
The licensee notified the resident inspectors immediately and
reported this occurrence to the NRC duty officer pursuant to the
requirement of 10 CFR 50.72(b)(2)ii.
The licensee reset the trip and
returned components to their required conditions.
The licensee is
investigating the circumstances associated with this problem.
Fuel
movement was also restarted after the investigation was completed.
4. Maintenance Activities (62703)
a. Maintenance activities were observed and/or reviewed during the
reporting period to verify that work was performed by qualified
personnel and that approved procedures in use adequately described
work that was not within the skill of the trade.
Activities,
procedures,
and work requests were examined to verify; proper
authorization to begin work, provisions for fire, cleanliness, and
exposure control,
proper return of equipment to service, and that
limiting conditions for operation were met.
Maintenance reviewed and witnessed in whole or in part:
Replacement of Inner Tier Connector Cables On
Keowee Unit 2 Battery
WR 571130 Perform P.M. On Main Steam Emergency Feedwater Pump
Turbine
10
Perform Votes Testing On Valve 1C 391
P.M. Motor Driven EFW Pump 'lB' Breaker
WR 57865A Visual Inspection and Test lB LPI Pump Motor
WR 51883J CT-1 Transformer 4160V Bus Maintenance
WR 27982C
Investigate BKR Feeder On 3X5F "Alternate Feeder For
1XSF"
WR 64789C Repack 1LP22
Additionally, various portions of the activities listed in paragraph 7 of
this report were periodically observed by the inspectors.
b. Near-Miss Accident Involving Work In Energized Switchgear
On May 8, 1990, at approximately 10:30 a.m. a near-miss accident
occurred during preventive maintenance work on selected Motor Control
Center
(MCC)
panels.
While no injuries actually occurred the
accident had the potential for causing serious injury or death to
personnel involved.
The incident, which involved maintenance on the
shutdown unit, resulted in conditions which could have had adverse
effects on the operations of Unit 3 which was operating at 100
percent power.
A safety investigation was performed by the
licensee's safety group.
Two Instrument and Electrical
(I&E)
specialists were assigned to perform work involving torquing of
electrical connections inside the cabinet for 208V MCC 1XSF located
in the Safe Shutdown Facility (SSF).
208V MCC
1XSF is normally
powered from 600V MCC 1XSF but contains a Kirk Key interlock feature
to allow it to be fed from 208V MCC 3XSF.
deenergized with both its supplies (600V Load Center OXSF and 600V
Load Center 1X8) tagged out. Apparently, due to some communications
problems between the I&E specialists and the Unit 1 Operations
Supervisor, the specialists did not realize the line side of the
Kirk Key enclosure on 1XSF was being supplied with 208 volts from
3XSF.
Although the specialist did check for voltage present on the
bus prior to work he didn't detect any, apparently, because he did
not make good contact with the test leads.
The cabinet involved is
very small and confined.
Protective gloves were being worn during
the work in accordance with the procedure. At the conclusion of the
work his screwdriver touched the Kirk Key housing and caused an
electrical arc.
This.caused the 100 amp breaker between 600V MCC
3XSF and 208V MCC 3XSF to trip, deenergizing 208V MCC 3XSF and 208V
MCC 3XSF-1. This deenergized several Unit 3 valves:
3HP-3,4
Letdown Cooler Outlet Valves
SSF-3HP-20
Reactor Coolant Pump Seal Return Isolation Valve
SSF-3RC-5,6
Pressurizer Water and Steam Space Isolation
Valves
SSF-3SF-97
SSF Reactor Coolant Makeup (RCMU) Pump Supply
Valve
SSF RCMU Pump Return and Letdown Valve
11
Indications of the valves being deenergized were received in the
Unit 3 Control Room and personnel in the SSF Control Room called the
Unit 3 Control Room and informed them of alarms received. No valves
shifted position.
The loss of power to the valves did not result in
any immediate operational concerns or transients since either the
valves were already in the required safety position or another valve,
operable from the control
room,
was available in the flowpath
involved.
The proposed Limiting Conditions for Operation (LCO)
Action Statements of TS 3.18 (Standby Shutdown Facility) and TS 3.6
(Reactor Building) for containment isolation valves out of service
were both entered. Approximately 15 minutes after the loss of power
to the 3XSF panels they were reenergized by shutting the tripped
breaker.
Security was initially informed to take the measures
required if the SSF is degraded. Due to an apparent misunderstanding
by some of the security personnel involved, no security measures were
taken.
The licensee's safety investigation concluded that the I&E personnel
were complying with their procedures but failed to verify all
alternate power sources deenergized despite review of an electrical
diagram.
Additionally,
poor communications
between
the I&E
specialists and the Unit 1 Operations Shift Supervisor along with a
personnel error while checking for voltage prior to work, resulted in
personnel working on energized gear without realizing it
was
energized.
The inspectors also noted that the 100 amp breaker between 600V MCC
3XSF and 208V
MCC 3XSF tripped on the fault prior to the 50 amp
breaker between 208V MCC's 3XSF and 1XSF.
This is under
investigation by the licensee.
C. Activities Being Performed on Wrong Unit
On May 10, 1990, the licensee informed the inspectors that a valve on
Unit 2 had been inadvertently cut out of an operating system. This
mistake did not result in the loss of any safety-related function.
On May 9, two mechanical technicians were directed to cut out
1LWD498,
a drain valve on a Unit 1 Low Pressure Injection pump,
in
room 62 of the Auxiliary Building. This valve is in a portion of the
system that is isolated except during draining operations. Instead,
the technician cut out valve 2LWD498.
Unit 1 and Unit 2 components
are located in room 62 and are adequately labeled.
The technicians
were experienced and qualified to perform this function.
The work
which was to be performed was documented on work request (WR) 541261
which also contains a sign-off and a verification sign-off to assure
the craft conducting the WR are performing the desired work on the
correct component. The area was well lighted and no reasons could be
identified, other than personnel error, for this occurrence.
The
mistake was made late in the shift on May 9 and was recognized by the
technicians on May 10 and reported to their supervision.
The valve
was immediately re-welded into the system, as required.
12
A second occurrence of work being performed on the wrong unit
occurred on May 17,
1990.
Two I&E Technicians were assigned to
replace a solenoid valve in the operating system to 1PR4. This valve
was to be replaced in accordance with WR 546111.
This valve is
normally shut during operation with power removed so there was no
affect on unit operation. The valve is located in the Unit 1 purge
room adjacent to the combined Unit 1 and 2 purge room, which was
poorly lighted.
Working with a flashlight as the source of light,
the technicians removed the solenoid valve.
Upon exiting the area,
they noted that the valve was labeled as a Unit 2 valve and not
Unit 1. Their immediate action was to reinstall the valve, to leave
the electrical splices exposed for QA verification, and to notify
their supervision.
The work request for this activity also included
sign-offs for verification that work was being performed on the
correct component.
Oconee Nuclear Station Directive 2.2.2,
Independent Verification
dated 1/24/90, defines how this process is to be performed to assure
that activities are conducted to reduce human errors.
requires the licensee to have adequate procedures and specifies that
these procedures will be adhered to.
The failure to adhere to
SD 2.2.2 is a violation of TS 6.4.1 and is identified as violation
50-269,270,287/90-12-02.
These examples of failure to follow
procedures are another example of the violation identified in
Inspection Report 50-269,270,287/90-16 and will not be cited
separately.
d.
Grounding Strap Inadvertently Left On 1B2 RCP Switchgear
On May 16, 1990, at 5:46 p.m., when control room (CR) operators were
attempting to run a test on the 1B2
RCP motor with the pump
uncoupled, an electrical flash occurred in the 1B2 Reactor Coolant
Pump (RCP) motor switchgear (designated 1TB) followed by smoke coming
out of the switchgear compartment. Additionally, the rear panel had
blown clear of the compartment.
The smoke cleared immediately and
there was no indication of any fire.
At the time a non-licensed
operator in the area of the breaker saw some smoke coming out of the
corpartment and notified the CR.
When the CR operator attempted to
start the 1B2 RCP motor he received indications that the breaker had
not closed and immediately opened the incoming supply breakers for
the 6.9 KV 1 TB switchgear.
The licensee's investigation revealed that this incident occurred as
a result of a personnel error. Due to the error the grounding straps
were inadvertently left connected to the load side of the 6.9 KV
breaker supply to the 1B2 RCP motor.
Grounding straps had been
installed earlier in the outage to protect equipment and personnel in
case of inadvertent closing of the breaker. When the operator tried
to close the breakers,
current was shorted to ground. through the
straps.
The breaker opened on an overcurrent condition.
The test
was run successfully at a later date.
A high-potential test was
performed on the 1B2 RCP motor cables and gas pressure in the cable
13
penetration was checked.
No damage to the pump motor or cables was
identified. Minimal damage to the breaker occurred.
The CX transformer, which serves as a backup power supply to the
auxiliary loads of the Keowee hydro units, was also locked out by
actuation of its differential current relay. As a result, the plant
entered a 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> LCO in accordance with TS 3.7.2.a.3. The licensee
is continuing their investigation to determine why this occurred.
The transformer was declared operable after a ratio test was
performed and the LCO was exited.
The primary source of auxiliary
power to the Keowee unit transformers lX and 2X were operable at all
times during this event.
The inspector reviewed this event. This event was due in part to the
lack of a formal mechanism for controlling the installation and
removal of grounding straps on the non-safety 6.9 KV breakers. This
is considered by the licensee to be within the skills of the craft.
The licensee is evaluating this process- as part of their
investigation of this incident.
5. Seismic Monitoring Instrumentation
During this report period the inspectors reviewed the installation and
testing of the five peak accelerometer recorders in the Unit 1 Reactor
Building (RB).
Deviation 50-269/86-20-02:
Discrepancies Between the FSAR
and Installed Seismic Instrumentation addressed an earlier situation in
which the five peak recording accelerometers listed in Section 3.7.4.1 of
This item was closed in Inspection Report
50-269,270,287/89-22 after a Nuclear Station Modification and several
Exempt Changes were utilized to replace the old degraded detectors with 2G
Peak Accelerometer Recorder model PRA-400. These detectors are made by
Engdahl Enterprises and are rated for the temperature and other conditions
present in the RB.
These are self contained devices requiring no external
power.
Essentially each device consists of 3 metal plates which are
etched with a scribe upon seismic activity.
On a tour of the RB the
inspectors examined each of the five recorder locations.
The detectors
had been removed for calibration. The following observations were made;
-
Three of the recorders are mounted on components which experience
high temperatures during operation and thus are attached to thermal
mounts specifically designed for this purpose. The thermal mount for
the recorder attached to a pressurizer. support was loose and its
attachment blocks were broken.
Subsequent discussion with the
Instrument & Electrical (I&E)
technicians who removed the detector
for calibration confirmed that little useful information would have
been attainable from this instrument.
-
All of the recorder mounts with the exception of the one mounted on a
line hanger are aligned in a North-South location (the latter is
aligned East-West due to interference).
While the manufacturer's
data sheet indicates that the recorder should be mounted within 1
14
degree of North, the licensee's- installation procedures did not call
for such precision.
The inspectors reviewed the safety evaluation
addressing the alignment steps of the procedure and discussed this
issue with the responsible onsite engineer.
He stated that Design
Engineering had contacted the manufacturer (Engdahl) who indicated it
was not necessary to align the detectors within 1 degree North for
them to operate properly.
The inspectors discussed the calibration procedure and walked through
several steps of the procedure (IP/1/B/125/2A:
2G Peak Recording
Accelerometer Calibration) with several I&E technicians who perform the
calibrations.
The equipment required to interpret the plate indications,
was readily available and operable. The technicians were knowledgeable of
the monitors and how to interpret the indications. The plates, which had
been removed for calibration, were examined.
Several of the plates had
indications of motion,
some as high as 2G.
The technicians and the
engineer involved indicated that these had been caused by mounting the
recorders in areas of high vibration.
These indications could mask the
desired information in the case of an actual earthquake.
DE is currently
assessing this issue and considering mounting the recorders in more rigid
locations such as the RB floor or walls during the next outage.
Since
Section 3.7.4.1 of the FSAR specifically lists the locations of the
recorders a change will be required if
the detectors are moved in the
future.
The licensee intends to remount detectors in their present
locations before the end of the current outage.
The licensee performed
an operability evaluation on the recorders in their current- locations and
determined them to be operable.
No violations or deviations were identified.
6. Notification Of Unusual Event Due To Chemical Spill
On April 26,
1990, at 1:10 p.m. an Unusual Event was declared due to a 16
gallon spill of hazardous chemicals in a pipeyard. The pipeyard is inside
the protected area but is not in an enclosed area.
The source of the
spill was a 55 gallon drum which had rusted through.
The spill was
stopped by uprighting the drum. The chemical involved was "M and S Safety
Solvent" which contains Tetrachlorethylene and Methylene chloride. The 16
gallons were spilled into soil under the drum. No personnel were injured
or exposed to the chemical.
RP/O/B/1000/01:
Emergency Classification requires that an Unusual Event
be declared if a spill of a chemical equal to or greater than ten times
the Reportable Quantity (RQ)
per RP/O/B/1000/17:
Hazardous Substance
Release, occurs.
Enclosure 4.2 of RP/O/B/1000/17 states that a spill of
greater than
.33 gallons of M&S Safety Solvent (30 percent
Tetrachloroethylene, 30 percent Methylene Chloride) is reportable. Local
agencies, South Carolina's Department of Health and Environmental Controls
(DHEC), and the National Response Center were all informed as required. A
report to the NRC in accordance with 10 CFR 50.72(a)(3) was made.
The inspector toured the pipeyard and examined the remaining barrels of
chemicals.
Several corroded barrels containing chemicals were noted.
15
This was brought to the licensee's attention.
Additional corrective
actions are being.considered by the licensee.
No violations or deviations were identified.
7.
Unit 1 Refueling Outage (71707)
Unit 1 commenced the end of cycle 12 refueling outage on April 26.
The
outage is presently scheduled for 41 days.
Major maintenance activities
during this period have been:
'B' Low Pressure (LP) Turbine Inspection
Refurbish 'B' LP turbine intercept valves (6)
Hi-Pot Test all 4160V breakers
Eddy Current Test Steam Generators
ISI inspections on Reactor Vessel Closure
Head Weld and Head to Flange Weld
Approximately 400 Valve and valve actuator overhauls
Local Leak Rate Testing and Integrated Leak Rate Testing
preparations
The inspectors have monitored various activities including the initial
shutdown and subsequent cooldown/draindown.
The schedule at present
indicates the unit should be returned to operation about June 6, 1990.
8.
Inspection of Open Items (92700)(90712)(92701)
The following open items were reviewed using licensee reports, inspection,
record review, and discussions with licensee personnel, as appropriate:
a.
(Closed)
P2189-18:
Limitorque Corporation Reported Failures of
Melamine Torque Switches on SMB-000 and SMB-00 Operators.
In
November of 1988, the Limitorque Corporation identified a common mode
failure of melamine torque switches installed in SMB-000 and SMB-00
actuators.
This information was transmitted to Duke Power Company in
correspondence dated November 3, 1989.
Problem Investigation Report
4-088-0250, was issued to determine if the problem existed at Oconee
and the corrective actions to be taken if the problem was identified.
The licensee identified 34 valves as needing inspection.
Of this
group 22 were identified as safety-related and 12 non safety-related.
Of these operators,
27
have been repaired or replaced,
16
safety-related and 11 non safety-related.
There are 3 left to be
corrected on Unit 1, all safety-related, which are scheduled to be
corrected during the outage presently in progress. The remaining 4
are on Unit 2 and are scheduled to be corrected in upcoming outage.
Based on this action, this item is closed.
b.
(Closed)
P2190-01:
PT 21 Associated With Findings and Potential
Concerns Based on ECOTECH/RAM-Q Source Verification and Performance
16
Based Audit of the Rockbestos Company. This item was discussed with
the Design Engineering (DE)
section.
DE stated no safety-related
applications of this material has been used at Oconee. In addition,
DE informed the inspector that Rockbestos had recently been audited
by the
NRC
and
no unsatisfactory findings were identified.
Rockbestos Co., is providing written documentation to Duke that the
wire which was the subject of this concern, is satisfactory to be
used in safety-related applications. Based on this information, this
item is closed.
9. Exit Interview (30703)
The inspection scope and findings were summarized on May 21,
1990, with
those persons indicated in paragraph 1 above.
The inspectors described
the areas inspected and discussed in detail the inspection findings. The
licensee did not identify as proprietary any of the material provided to
or reviewed by the inspectors during this inspection.
Item Number
Description and Reference
VIO 269,270,287/90-12-01
Failure
to
Incorporate Design
Basis
Information Into Electrical Relay Procedures
(paragraph 2.c)
VIO 269,270,287/90-12-02
Failure to Follow Procedures Resulting in
Incorrect Component Maintenance
0II