ML15118A104
| ML15118A104 | |
| Person / Time | |
|---|---|
| Site: | Oconee  |
| Issue date: | 04/17/1996 |
| From: | Crlenjak R, Harmon P NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML15118A101 | List: |
| References | |
| 50-269-96-05, 50-269-96-5, 50-270-96-05, 50-270-96-5, 50-287-96-05, 50-287-96-5, NUDOCS 9605140289 | |
| Download: ML15118A104 (15) | |
See also: IR 05000269/1996005
Text
REG4
UNITED STATES
NUCLEAR REGULATORY COMMISSION
REGION II
101 MARIETTA STREET, N.W., SUITE 2900
ATLANTA, GEORGIA 30323-0199
Report Nos.:
50-269/96-05, 50-270/96-05 and 50-287/96-05
Licensee:
Duke Power Company
422 South Church Street
Charlotte, NC
28242-0001
Docket Nos.:
50-269, 50-270 and 50-287
License Nos.:
Facility Name:
Oconee Units 1, 2 and 3
Inspection Conducted: March 16-23, 1996
Inspectors:
i-
-_
__
P. E. Harmon, S 0or Resident Inspector
Date Signed
P. Humphrey, Resident Inspector
P. Fillion, Reactor Inspector
D. Forbes, Reactor Inspector
D. Jones, Reactor Inspector
L. Wie
Project Mapa
Approved by:
- 7
R. . Crlenja
,
ranch Ch f
Dafe Si ned
Division of Reactor Projects
SUMMARY
Scope:
This special inspection was conducted to review the circumstances
surrounding the Unit 3 reactor trip event on Saturday, March 16, 1996.
The trip occurred during the performance of a test on the electrical
system. The Senior Resident Inspector was on-site at the start of the
event and was subsequently supported by other NRC inspectors who
promptly arrived on-site to participate in the special inspection
effort.
The circumstances surrounding the event included several complications:
(1) a temporary loss of off-site power and delayed restoration of
off-site power; (2) tripping of the reactor coolant pumps and subsequent
natural circulation core cooling for approximately 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />; and
(3) Condensate System pressure surge (water hammer) and the spread of
condensate polishing resin throughout the system. The scope of the
inspection included:
(1) root cause determination; (2) evaluation of
trip transient analysis and system response; (3) evaluation of licensee
ENCLOSURE 2
9605140289 960417
ADOCK 05000269
G
III)
2
performance; and (4) corrective actions for the root cause and other
equipment problems.
Results:
The licensee accurately identified the root cause. The root cause was
determined to be a vibration-induced actuation of a load shed relay.
The relay was found to be improperly assembled.
(paragraph 2.3)
The licensee's event analysis properly identified all abnormal
indications and their causes. The Post Trip Review was complete and
thorough. (paragraph 5.0)
Corrective actions were comprehensive and well managed. All identified
pre-startup items were adequately addressed.
(paragraph 5.0)
Operators were very effective in managing the trip. Management was
effective in providing oversight and applying resources. The
Significant Event Investigation Team compiled an accurate sequence of
events and provided valuable input for corrective actions.
(paragraphs 3.2. and 3.3)
One Violation was identified for failure to make a 4-hour report as
required by 10 CFR 50.72.
(paragraph 3.1)
A weakness was identified concerning the length of time required to
regain off-site power. (paragraph 2.2)
A concern was identified regarding the schedule for implementation of a
modification to defeat the fast transfer of the reactor coolant pump
buses.
(paragraph 4.0)
ENCLOSURE 2
REPORT DETAILS
Acronyms used in this report are defined in paragraph 8.0.
1.0
PERSONS CONTACTED
Licensee Employees
- B. Peele, Station Manager
- E. Burchfield, Regulatory Compliance Manager
D. Coyle, Systems Engineering Manager
- J. Davis, Engineering Manager
- W. Foster, Safety Assurance Manager
J. Hampton, Vice President, Oconee Site
D. Hubbard, Maintenance Superintendent
C. Little, Electrical Systems/Equipment Manager
J. Smith, Regulatory Compliance
G. Rothenberger, Operations Superintendent
R. Sweigart, Work Control Superintendent
- L. Azzarello
Other licensee employees contacted included office, operations,
engineering, maintenance, chemistry/radiation, and corporate personnel.
- Attended Exit Interview
2.0
EVENT DESCRIPTION (93702)
2.1
System Description and Alignment Immediately Prior to the Event
The on-site emergency power supply consists of two hydro units. When
the control system senses the need for emergency power, one hydro unit
is automatically aligned to supply power to plant auxiliaries via an
isolated 230 KV (Yellow) bus, also referred to as the Overhead Line.
The other hydro unit is automatically aligned to supply power via a
13.8 KV cable, referred to as the Underground Line. A 100 KV
transmission system, supplemented with an off-site gas turbine, can
supply power to the plant buses.
At the onset of the March 16, 1996, Unit 3 reactor trip event, the
licensee was in the process of performing Periodic Test PT/O/A/0610/22,
Degraded Grid, Switchyard Isolation and Keowee Overfrequency Protection
Functional Test. This test verifies, in part, that the emergency
electrical system has the capability of performing its function when
called on to do so.
The three Oconee units were at full power. In
accordance with the test procedure, both 87.5 MVA rated hydro units were
generating to the grid at about 74 MW each. The 100 KV system was in
ready standby with the gas turbine running. Automatic fast transfers of
plant auxiliaries and RCP buses from the unit auxiliary transformer to
the startup transformer were disabled by placing transfer switches in
the manual mode.
Three 230 KV motor operated disconnect switches at
breakers 21, 24, and 33 were open to help protect the units from
potential interaction with portions of the system under test.
II
2
2.2
Trip Description and Sequence of Events
At 1:17:47 pim., on March 16, 1996, simulated signals were injected to
initiate emergency alignment of the emergency power system. Plant
auxiliaries in all three units should have continued to operate, being
powered from their respective unit auxiliary transformer. Instead,
there was a simultaneous spurious actuation of a single set of contacts
in a Unit 3 auxiliary relay which caused tripping of a hotwell pump, a
booster pump, and two heater drain pumps in the Unit 3 secondary system.
One condensate booster pump started automatically as designed. Within
about 14 seconds, another condensate booster pump was manually started.
These events were followed by a condensate system pressure transient
which led to feedwater pump trips due to low suction pressure, and a
Unit 3 reactor/turbine trip at 1:18:45 p.m. Operators entered the
Emergency Procedures for the plant trip. Units 1 and 2 were unaffected,
remaining at full power.
The Condensate System transient was accompanied by a water hammer in the
Condensate lines which ruptured several heat exchanger gasket seals and
allowed dislocation of powdex resins. As designed, power was
automatically restored to the plant buses 21 seconds after the unit
trip. The power source was one Keowee unit via the 230 KV Yellow bus
(Overhead Line). This particular alignment was the result of being in
the test procedure when the trip occurred. Grid power was available at
the 230 KV Red bus. The RCS was in natural circulation, because
automatic fast transfer of the RCPs had been disabled as explained
above. Operators did not start the RCPs at the earliest opportunity
allowed by the emergency operating procedures.
Instead, the decision
was made to first parallel the Keowee unit and the Yellow bus to the
grid. The decision to wait until off-site power was restored was
primarily based on a concern that if a problem occurred during
paralleling, having the RCPs already running would create another loss
of forced circulation. With a lower heat load, natural circulation
would be more difficult to initiate the second time around. In
addition, the natural circulation in progress was stable and was an
acceptable operating mode.
At about 4:48 p.m., the Yellow bus and Keowee Unit supplying it was
synchronized to the grid (Red bus).
The delay of approximately 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />
to restore off-site power from when it was available was due to several
factors:
The transfer procedure required transferring control of Keowee
from local to remote. A previous control transfer of Keowee from
Oconee (remote) to Keowee (local) had resulted in a Keowee unit
trip and damage to a breaker closing coil due to an error by the
Keowee operator making the transfer. Although this transfer was
in the opposite direction, operators wanted to review the
circumstances to prevent another Keowee trip. This was determined
not to be a concern after reviewing the previous event.
ENCLOSURE 2
3.
Several discussions were conducted with management and engineering
concerning whether to perform a dead bus or live bus transfer.
Operators wanted to review the corrective action taken for the
problems with paralleling during the 1992 Loss of Off-site Power
event.
Slow or delayed response of the Keowee speed control during the
actual synchronizing steps caused extra caution on the part of the
operators making the transfer.
The cumulative effect of these delays resulted in an inordinate amount
of time to restore off-site power. Since no complications or problems
(other than slow Keowee speed response) were encountered, the time taken
to complete the maneuver is considered a weakness. While each of the
actions taken could be considered conservative, the inspectors
determined that some of the training and decision-making during this
period should be examined to facilitate off-site restoration during
future events.
RCPs 3B1 and 382 were started at about 9:12 p.m. and 9:36 p.m.,
respectively. This allowed the operators to exit Emergency Procedure
EP/1/A/1800/01. Operators had not exited the EP because the Subsequent
Actions section requires starting RCPs before exiting. Following the
restoration of forced circulation, efforts continued to clean up
displaced condensate polishing resins. This was accomplished by short
cycle recirculation of the condensate through a restored resin bed.
At approximately 11:00 p.m., management decided to cool down and
depressurize the RCS to a point where decay heat removal via the LPI
system could be placed in service. This decision was reached after the
extent of the resin contamination was determined. The licensee
considered that continued use of the secondary side systems as a heat
sink for the reactor would be a challenge to the operators, especially
while trying to clean up these systems at the same time.
Actual unit cooldown began at approximately 5:50 a.m., on March 17.
During the cooldown, the licensee attempted to use the Unit 2 MDEFW pump
.and water supply to feed Unit 3 steam generators. This would allow more
flexibility in cleaning up the Unit 3 Condensate System and provide an
additional source of water for the cooldown.
The procedure for
implementing this alignment was revised just prior to use. The
procedure change allowed a single Unit 2 MDEFW pump (rather than two
pumps) to supply both Unit 3 steam generators.
When this arrangement
was initiated at 11:47 a.m., operators had difficulty controlling
levels.
The Unit 2 MDEFW pumps were stopped and the Unit 3 pumps were
placed back in service at approximately 12:57 p.m. The LPI system was
placed in service at approximately 5:15 a.m., on March 18. While the
LPI system was being aligned for the DHR mode, operators observed level
decreases in the Letdown Storage Tank (LDST), indicating leakage past
ENCLOSURE 2
4
isolation valves from the LDST into the LPI system. The cause was later
determined to be leakage past manual isolation valves 3LP-40 and 3LP-41.
The valves were manually tightened shut, and the leakage stopped.
2.3
Root Cause of Initiating Transient and Reactor Trip
The root cause of the event was determined to be incorrect assembly of
auxiliary relay 3RLS1X. The relay was a style D23MRD704A1 manufactured
by Cutler-Hammer Co., having 120 VDC coils. This relay had a base
assembly, a four-contact front deck (middle section) and a four-contact
front attachment (upper section).
The armature shaft was composed of
three pieces threaded together. The licensee determined that the
armature shaft pieces for the middle and upper sections were not
completely threaded together. This had the effect of decreasing the
contact gap of one contact in the upper section from the normal 50 - 70
mils to 5 mils. The contact with the 5 mil gap momentarily closed when
it was subjected to vibration. The vibration was caused by simultaneous
operation of several relays in an adjacent cabinet. The momentary
contact closure caused one of three "load shed" relays to operate, which
caused the tripping of pumps described in Paragraph 2.2. As a result,
relays which were being operated as part of the test caused the
vibration which resulted in actuation of a "load shed" relay that was
not part of the test. As of the end of the report period, the licensee
had not been able to identify why the armature shaft was not completely
threaded together. The relay and its component parts could have been
improperly assembled by the vendor, by the licensee, or could have
undergone age/use related movement after being placed in service. The
resident inspectors will continue to follow this item via the licensee's
corrective action program.
The linkage between operation of the test relays and momentary closure
of the problem contact was demonstrated in a special test. This test
was reviewed and witnessed by the inspectors. After the cause had been
proven, the problem relay was replaced. The special test was repeated
to show that the problem had been corrected. The licensee has
determined that 210 of the relays are installed in safety-related
applications. It should be noted that the safety related function of
the subject relays, including the failed relay, are not affected by this
failure. The licensee has developed a schedule for inspecting a
representative sample of these relays. The sample will be expanded to
100% if any additional relays are found to not meet the acceptance
criteria currently being developed. The resident inspectors will
continue to review this item.
2.4
Root Cause of Resin Dispersement
The transient in the Condensate System caused resin to be backflushed
from the polishing demineralizers into the condenser hotwells. When the
hotwell pumps restarted, the polishing demineralizers were bypassed and
ENCLOSURE 2
5
the resin was transported throughout the condensate system rather than
being redeposited onto the polishing demineralizers.
The licensee determined that the resin was backflushed from the resin
beds when the Condensate System pressure surge (water hammer) occurred.
A bypass valve automatically opens to bypass the resin beds, as well as
the resin filter downstream of the beds, on a low booster pump suction
pressure. Hotwell pump check valves were believed to have failed to
fully seat during the backflow condition, allowing the resins to be
carried into the hotwell.
From there, the resins were pumped into the
Condensate System through the open resin bed bypass valve. The bypass
line does not contain a resin filter or trap. The hotwell pump check
valves were later radiographed to determine their position. No problems
were found with these valves.
The ruptured heat exchanger seals, caused by the water hammer in the
condensate lines during the transient, allowed leakage of water and
resin from the system in several locations of the turbine building.
Upon discovery of the resin in that leakage, the licensee promptly began
sampling of the secondary system to determine the extent of the resin
contamination and aligned the feedwater system to minimize further
spread of that contamination. The EFW pumps were used to provide
feedwater to the steam generators from the upper surge tank rather than
from the condensate hotwell which was known to be contaminated. A small
amount of resin was transported to the steam generators when water from
the condensate storage tank, which contained some resin material, was
used as makeup water for the upper surge tank.
3.0
ASSESSMENT OF LICENSEE PERFORMANCE (93702)
3.1
Assessment of Operator Performance in Response to the Event
The Senior Resident Inspector was on-site to witness the performance of
PT 0/A/0610/22. When the trip occurred at the initiation of the test,
the inspector was able to observe operator response and assess
performance. The control room shift crew was extremely effective in
controlling the plant after the trip. The SRO clearly directed the
whole crew, and communications were excellent. Conservative decisions
were made throughout the event.
Some exceptions to the effective operator performance were noted:
SRO Started a Condensate Booster Pump:
When the spurious operation of
the load shed relay caused tripping of several plant auxiliaries, the
SRO on shift attempted to mitigate the transient by quickly starting a
condensate booster pump. This was done without informing the shift
crew. The alarm printer shows the pump was started 14 seconds after the
load shedding occurred, and 12 seconds after the standby booster pump
automatically started.
Starting the additional booster pump may have
ENCLOSURE 2
- I6
caused the booster pump suction pressure to drop even lower, leading to
.a booster pump trip and the eventual Unit 3 trip.
Delayed/Inadequate Notification:
Pursuant to 10 CFR 50.72(b)(2)(ii) the
licensee should have notified the NRC Operations Center of the reactor
trip within four hours of the event. However, notification was made
approximately 8k hours after the trip occurred. In addition to the late
report, the report which was eventually made did not describe the
extended natural circulation mode, the loss of off-site power, or the
resin displacement-event. This failure to make proper notification as
required by 10 CFR 50.72 is identified as Violation 50-269,270,287/96
05-01, Failure To Make Proper 10 CFR 50.72 Notification.
Flux/Flow Imbalance Trip:
On March 17, while in cooldown mode, a
reactor trip signal was generated due to flux/flow imbalance. This trip
occurred when operators stopped RCP 381 due to high vibration. The STAR
module sensed zero flow in RCP loop B, and initiated the trip as
designed. Operators were not aware of this design feature of the
recently installed STAR module. The fact this trip signal was generated
indicates there was inadequate training on and knowledge of a recently
modified system. At the time of this trip, the trip breakers had been
reset, but rods had not been pulled.
3.2
Assessment of Licensee Management Performance
Management oversight was extensive for the planned electrical system
test. Both the Plant Manager and the Engineering Manager were on-site
for the pre-job briefing and the conduct of the test.
The inspector
attended all team meetings following the trip and was able to directly
observe the decision-making process at each meeting.
Management provided clear direction to the operators throughout the
event. Operations personnel were involved and consulted in all decision
making.
Management was especially effective in three areas:
Assembling and Controlling Additional Resources: Bringing in and
controlling additional personnel, while coordinating with the shift
crew, relieved the shift crew of that responsibility. As a result, the
shift crew was able to properly focus on the plant conditions and
evolutions. Several maintenance and engineering disciplines were
already on-site during the test, and others were contacted and scheduled
as needed.
Secondary System Cleanup: The Chemistry Manager and other chemistry
personnel were brought in shortly after the displaced resin was
identified. This resulted in an accurate assessment of the extent of
the resin dispersion and an effective cleanup. This was effective as
evidenced by the prompt assessment that powdered resin had been
ENCLOSURE 2
7
dispersed throughout the Condensate Polishing System and alignment of
the feedwater system to minimize any further transport of resin to the
Chemistry management was also effective in recovery from the transient.
The secondary system was aligned to return water from the secondary
system to the condenser in order to remove residual resin from the
system using the condensate polishers. Soak and flush procedures were
also successfully used to remove residual resin from the steam
generators. Sampling results confirmed the success of both of those
actions. The inspectors reviewed analytical results for parameters
routinely used to monitor water quality in secondary system component:
and determined that the chemistry control parameters were quickly
returning to normal ranges and specifications as delineated in Chemistry
Manual 3.8. Chemistry management planned to continue the necessary
system cleanup actions in order to achieve normal secondary water
chemistry before restart of the unit.
Troubleshooting and Root Cause Determination: Management provided
excellent direction for quarantining, isolating, and preserving evidence
for the suspected components.
Even after the 3RLS1X relay was
identified as the probable cause, the troubleshooting effort continued
to investigate other possibilities. A Failure Investigation Process
(FIP) evaluation was used as the root cause and troubleshooting guide.
Special tests were designed and conducted to replicate the conditions of
the event which was suspected as producing the vibrations in the relay
panel.
These tests provided evidence that the relay was in fact passing
voltage and current for several milliseconds when the panel vibrations
were reproduced. As a result, the licensee preserved the condition of
the relay and was able to identify it as the clear root cause. After
replacing the relay with a bench-tested relay, the vibration-producing
test was rerun, and the new relay was not affected.
revealed the failed relay had contacts with out of tolerance gaps as
described in paragraph 2.3.
3.3
Assessment of SEIT Investigation
Shortly after the trip, licensee management considered the circumstances
of the trip and the possible relationship of the emergency power system
testing warranted an independent investigation through the SEIT proces..
The SEIT was assembled and arrived on-site during the morning of March
17.
The SEIT was chartered under a standard Duke Power Company charter
for such teams.
The charter included preparation of a sequence of
events (SOE), evaluation of the root cause, and providing short-term
(pre-startup) and long-term recommendations. The SEIT exited March 21
with preliminary findings and recommendations. The licensee agreed to
complete all short-term items prior to starting up Unit 3.
ENCLOSURE 2
8
The inspectors reviewed the SEIT's SOE and determined that it was
complete and accurate. The SEIT's SOE was in close agreement with the
Oconee post-trip report SOE and a SOE compiled by the inspectors.
The SEIT concurred with the Oconee root cause determination.
Recommendations provided by the SEIT adequately addressed the actions to
be taken prior to startup. The inspectors concluded that the SEIT was'
an effective process which provided additional confidence that the
licensee's on-site assessment, root cause and corrective actions were
thoroughly addressed.
4.0
EQUIPMENT PROBLEMS
(93702)
The post-trip review indicated that the safety-related systems had
worked as designed, with the exception of the relay that initiated the
event.
The licensee identified several equipment problems, and
adequately addressed them prior to plant startup. Some of the problems
were caused by the water hammer, while others were revealed during the
event but were not caused by it.
LP-40/41 Leakage:
When the operators attempted to place the LPI system
in operation in the DHR mode, the water level in the LDST began
decreasing. The operators determined that two closed manual valves,
(3LP-40 and 3LP-41) were leaking past their seats.
This allowed LPI
pump discharge to recirculate back to the BWST. The valves were
manually tightened approximately one turn in the shut direction, and the
leakage from the LPI system stopped. The licensee believes the previous
alignment of these valves did not fully seat them in the shut position.
The valves are located in difficult positions for manual operation, and
the previous positioning may not have fully shut the valves. The
licensee checked the position of the same valves on Unit 1 and Unit 2
and found no problems. A Past Operability evaluation will determine if
the valves' position would have constituted an operability concern
during a hypothetical LOCA event.
Heat Exchanger Seal Failures:
The pumps in the Condensate System are
normally started with their discharge isolation valves shut to prevent
flow-induced pressure surges. During this event, the pumps were stopped
and restarted in rapid succession without normal system startup
alignments. This resulted in rapid flow decreases and pressure drops,
followed immediately by rapid flow increases and pressure increases.
Several sections of the Condensate System have long vertical runs which
are especially susceptible to water hammer effects. The water hammers
created intense, short-lived pressure spikes throughout the system. The
pressure spikes resulted in blown seals and gaskets in several heat
exchangers.
Water was observed leaking from the Generator Stator Cooler
and the Generator Hydrogen cooler. All gaskets were replaced and leak
checked.
ENCLOSURE 2
9
Water Hammer Damage To Safety-Related Components:
The licensee walked
down the Feedwater System and the Emergency Feedwater System and did not
find evidence of any damage. The licensee concluded that the water
hammer was limited to the Condensate System and the heat exchangers in
that system.
Unexpected Recloser Operation: Two 230 KV transmission lines running
between the Jocassee switchyard and the Oconee switchyard connect to the
Yellow bus through circuit breakers 12 and 15.
These circuit breakers
were equipped with a recloser programmed for one reclosing at two
seconds.
For all previous times that Periodic Test 0/A/0610/22 was
conducted, the procedure called for blocking the recloser during the
test. For the March 16 test, the procedure had been revised to remove
that particular step.
Plant engineers, based on review of the control
circuit diagrams, thought that the recloser would not generate a reclose
signal during the test. It appeared that the synchrocheck relay within
the recloser would not allow reclosing to a dead bus.
What actually
took place during the March 16 test was that the "switchyard isolate"
signal initiated-a-trip signal to breakers 12 and 15 (among other
breakers).
The Yellow bus became isolated as designed. Two seconds
following the trip signal, breakers 12 and 15 reclosed and immediately
tripped due to the permanent trip signal from "switchyard isolate."
When breakers 12 and 15 closed, the "switchyard isolate complete" signal
was interrupted. This reset a timer, which caused breaker 9 to close at
10.5 seconds rather than the programmed 8.5 seconds. This additional
two-second delay in establishing emergency power was not significant in
that the accident analysis time was not exceeded. As a result of this
information obtained from the periodic test, the licensee initiated the
following three corrective actions:
All recloser circuits at the Oconee switchyard would be corrected
such that they would operate as shown on the drawings (i.e., not
reclose to a dead bus).
As an interim measure, the reclosers for breakers 12 and 15 were
disabled.
If necessary, a modification will be implemented to block the
reclosing upon a "switchyard isolate" signal.
As far as could be determined during the inspection, the recloser
problem existed since initial plant startup.
It represents a problem
because the actual operation of the circuit was different than shown on
the drawing. The inspector concluded that there was no opportunity for
the licensee to identify the problem prior to conducting the periodic
test in its latest form.
Automatic Fast Transfer of RCPs:
Circuit breaker control circuits
included an automatic fast transfer of the reactor coolant pumps from
ENCLOSURE 2
10
the unit auxiliary transformer to the startup transformer upon a
generator trip. This feature created a potential problem in relation to
one scenario. The scenario of concern was when there is an External
Grid Trouble Protection System actuation, but not a unit trip. This
results in the 230 KV Yellow bus being energized from a hydro unit
(Keowee) and isolated from any other buses.
The reactor coolant pumps
would be powered from the main generator through the unit auxiliary
transformer. Should a unit trip occur from this configuration, a fast
transfer of the reactor coolant pumps to the Yellow bus would take
place. This is a transfer between two unsynchronized buses not
supervised by a synchrocheck relay. Therefore, an out-of-phase transfer
must be assumed. This could result in high transient current, which
could result in saturation of current transformers in the startup
transformer differential protection. This would result in lockout of
the startup transformer, as well as a loss of off-site power and one
path for emergency power. The Abnormal Operating Procedures include a
step to place the breaker mode selector switch in manual upon a
"switchyard isolate" alignment, thus blocking the fast transfer
described above. Nevertheless, this leaves a time period of
vulnerability (perhaps 5 minutes) which was a concern. The licensee
stated that they identified this same concern and a plant modification,
aimed at automatically blocking the fast transfer of the reactor coolant
pumps whenever a "switchyard isolate" signal is present, had been
approved. The inspector noted this modification (identified as NSM
2983) was on the three-year schedule. It was scheduled for the Unit 3
March 1998 outage, the October 1998 Unit 1 outage, and the March 1999
Unit 2 outage. The inspectors expressed to licensee management that they
should consider expediting the schedule for this modification, because
it had greater safety significance than indicated by the present
schedule.
5.0
CONFIRMATORY ACTION LETTER AND PRE-STARTUP REQUIREMENTS (92703)
On March 18, 1996, the Regional Administrator for NRC Region II issued a
Confirmation of Action Letter (CAL) to the Licensee. The CAL delineated
the following conditions and actions to be taken by the licensee prior
to restart of Oconee Unit 3.
Conduct a comprehensive investigation to evaluate equipment and
personnel response of the trip event, including a detailed
evaluation of the root cause of the event.
- Complete
the investigation by the SEIT. Evaluate the findings of
the Oconee staff including the SEIT and implement appropriate
corrective actions.
Conduct a walkdown of Keowee systems verifying that the present
alignment is appropriate to supply emergency power to all three
Oconee Units. Notify NRC before using Keowee in a manner that
ENCLOSURE 2
11
will decrease the reliable supply of emergency power to Oconee.
This includes operating with both Keowee hydro units to the grid.
Verify the Lee Steam Station is available to provide electric
power to Oconee on short notice, and notify NRC of any subsequent
change in its availability.
Meet with the NRC to discuss the results of the above items,
including a discussion of future Keowee testing plans. Obtain
concurrence from the Regional Administrator prior to entering
Mode 2.
The inspectors confirmed full compliance with the provisions of the CAL.
On March 22, the Regional Administrator.and members of his staff met
with the licensee at Oconee Nuclear Station. At the meeting, the
licensee provided a briefing on the status of Unit 3, details of the
SEIT findings, and discussed the root cause. All remaining items
required for startup were listed and presented at the meeting.
Following the meeting, the inspectors reviewed the licensee's completion
of all required items and confirmed their satisfactory completion. As a
result of the NRC review of the licensee's conformance with the CAL, and
completion of all required startup items, the Regional Administrator
approved closure of the CAL and startup of Oconee Unit 3 on March 25,
1996.
The unit was restarted later that day.
The licensee submitted to
NRC a letter, Response to Oconee Unit 3 Confirmation of Action Letter,
dated March 25, 1996. This letter detailed the compliance with the CAL.
As part of the inspection effort to ensure completion of the CAL items,
the inspector reviewed the licensee Post Trip Review Report. The report
independently identified the root cause, and listed the corrective
actions required prior to unit startup. The report was accurate and
comprehensive. All required actions were completed.
6.0
UFSAR REVIEW
A recent discovery of a licensee operating their facility in a manner
contrary to the UFSAR description highlighted the need for additional
verification that licensees were complying with UFSAR commitments.
During an approximate two month time period, all reactor inspections
will provide additional attention to UFSAR commitments and their
incorporation into plant practices, procedures and/or parameters.
While performing the inspections which are discussed in this report the
inspectors reviewed the applicable portions of the UFSAR that related to
the areas inspected. The inspectors verified that the UFSAR wording was
consistent with the observed plant practices, procedures and/or
parameters.
ENCLOSURE 2
12
7.0
EXIT MEETING
The inspection scope and findings were summarized on March 27, 1996, by
P. Harmon with those persons indicated by an asterisk in paragraph 1.
The inspector described the areas inspected and discussed in detail the
inspection results. A listing of inspection findings is provided.
Proprietary information is not contained in this report.
Dissenting
comments were not received from the licensee.
Item Number
Status
Description and Reference
VIO 269,270,289/
Open
Failure to Make Proper 10 CFR 50.72
96-05-01
Notification
8.0
ACB
Air Circuit Breaker
As Low As Reasonably Achievable
BHUT
Bleed Holdup Tank
BTO
Block Tagout
BWST
Borated Water Storage Tank
Confirmation of Action Letter
CFR
Code of Federal Regulations
Component Cooling
Condenser .Circulating Water
CR
Control Room
Design Basis Accident
Emergency Feedwater
EPSL
Emergency Power Switching Logic
End Of Cycle
Engineered Safeguards
FDW
Failure Investigation Process.
GL
Generic Letter
GPM
Gallons Per Minute
Health Physics
High Pressure Injection
Integrated Control System
I&E
Instrument & Electrical
IR
Inspection Report
KHU
Keowee Hydro Unit
KV
Killovolts
LCO
Limiting Condition for Operation
LDST
Letdown Storage Tank
LER
Licensee Event Report
Loss of Coolant Accident
Low Pressure Injection
Low Pressure Service Water
ENCLOSURE 2
Motor Driven Emergency Feedwater
MPManeac
Prcdr
MVAMillion Volts-Amps
MWMegawatts
NCVNon-Cited Violation
NLCNon-Licensed Operator
NSMNuclear Station Modification
NSDNuclear System Directive
EPOperating Experience Program
ONSOconee Nuclear Station
PSIDPounds Per Square Inch Differential
PSIGPounds Per Square Inch Gauge
PMPreventive Maintenance
PIPProblem Investigation Process
RCPReactor Coolant Pump
REMRoentgen Equivalent Man
RFORefueling Outage
SEITSignificant Event Investigation Team
SOERSignificant Operating Event Report
SFPSpent
Fuel Pool
Senior Reactor Operator
Technical Specification
Updated Final Safety Analysis Report
RaWork Control Center
R
iWork
Order
ENCLOSURE 2
Sinfcn4vn0IvsiainTa