IR 05000219/1993003
| ML20044C431 | |
| Person / Time | |
|---|---|
| Site: | Oyster Creek |
| Issue date: | 03/09/1993 |
| From: | Rogge J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML20044C427 | List: |
| References | |
| 50-219-93-03, 50-219-93-3, NUDOCS 9303230024 | |
| Download: ML20044C431 (32) | |
Text
.
-
...
,
,
e U. S. NUCLEAR REGULATORY COMMISSION
REGION I
Report No.
93-03 Docket No.
50-219 License No.
DPR-16 Licensee:
GPU Nuclear Corporation 1 Upper Pond Road Parsippany, New Jersey 07054 Facility Name:
Oyster Creek Nuclear Generating Station Inspection Period:
January 19,1993 - February 22,1993 Inspectors:
Dave Vito, Senior Resident Inspector John Nakoski, Resident Inspector Steve Pindale, Resident Inspector, Salem Tim Frye, Reactor Engineer, DRP Dan Moy, Reactor Engineer, DRS M
/
Approved By:
~
/
bate Rogge, Section Opf(/
Reactor Projects Section 4B
-
Inspection Summary: This inspection report documents the safety inspections conducted during day shift and backshift hours of station activities including: plant operations; radiological controls; maintenance and surveillance; engineering and technical support; emergency preparedness; security; and safety assessment / quality verification.
t Results: Overall, GPUN operated the facility in a safe manner. One unresolved item I
concerns a potential weakness in the licensee's equipment alignment and verification controls.
Another unresolved item noted a need for clarification of operating parameters and '
.,
surveillance test acceptance criteria for the station batteries.
!
,
&
D e
9303230024 930310 PDR ADOCK 05000219
,
G PDR v
,
}
'
EXECUTIVE SUMMARY Oyster Creek Nuclear Generating Station Report No. 934)3 Plant Operations
A January 23,1993, event resulted in thermal stratification in the core region due to degraded operation of the shutdown cooling system. An NRC Augmented Inspection Team
'!
(AIT) was sent to the site to investigate this matter. The AIT findings are documented in a i
separate report (50-219/93-80). On January 30,1993, a full scram signal was received
[
during system restoration efforts after the containment integrated leak rate test (ILRT). A
>
step within the ILRT procedure prescribed two actions which were in the improper order for restoring the scram discharge volume (SDV) to service. Performance of the procedure step j
caused a full scram signal due to high water Icvel in the scram discharge instrument volume which was not remedied before deactivating the SDV high water level bypass switch.
i I
Plant restart commenced on February 14, 1993, after completion of the 14R refueling outage.
The outage was completed within GPUN's predetermined outage schedule. Plant startup was
effectively accomplished and generally uneventful.
!
Radiolocical Controls j
Radiological controls continued to be effective in reducing. worker exposure and the incidence of radiological events during the refueling outage.
l Maintenance / Surveillance The replacement of the A and B station batteries was reviewed by the inspectors. Installation
,
and testing of the new batteries were well done. While the station batteries are appropriately tested in service, the inspectors noted that the minimum requirements for cell voltage and specific gravity in the Oyster Creek technical specifications were not consistent with the
,
normal battery operating characteristics. The loss of power test was completed successfully for both emergency diesel generator (EDG) trains. Operators responded well to EDG problems experienced during the test. A successful containment integrated leak rate test was performed. The test was well controlled and the test results were calculated appropriately.
.
!
The inspectors independently verified the test results.
i
!
,
e ii I
,
,.
i Engineering and Technical Supnort
.'
The licensee performed a thomugh evaluation of a previously identified problem with the fuel zone level indication system and implemented an appropriate plant modification. The licensee took appropriate compensatory actions in response to the unavailability of the redundant fire water system. A change to the de high-potential testing requirements of the
5 kV cable testing program did not adversely effect the ability of the program to detect cable
'i degradation. Engineering effectively resolved problems encountered with both EDGs during the loss of power test.
Safety Assessment and Ouality Verification Overall management control of outage activities was good. Effective use of integrated
.
scheduling and plant risk configuration guidance contributed to the completion of the 14R refueling outage within its predetermined schedule. Safety significant emergent work was not delayed or postponed to support plant restart. The licensee performed several transient analysis scenarios to support decisions regarding concerns with the fuel zone level indication system.
iii
,
- _.
'(
TABLE OF CONTENTS Pace EX EC UTIVE S UMMARY....................................... ii 1.0 OPERATIONS (71707, 93702)...............................
I 1.1 Outage / Operations Summary............................
1.2 Scram Signal Received During System Restoration After ILRT.......
1.3 Stanup Following the 14R Refueling Outage..................
1.4 Degraded Shutdown Cooling Event Summary..................
1.5 Facility Tours.....................................
2.0 RADIOLOGICAL CONTROLS (71707).........................
,
3.0 MAINTENANCE / SURVEILLANCE (62703, 61726, 62705)
............
3.1 Replacement of Station Batteries A and B (URI 50-219/93-03-01)
.....
3.2 Average Power Range Monitor Trouble Shooting............... 10 3.3 Loss of Power Test
.................................
'
3.4 Containment Integrated I enk Rate Test
.....................
4.0 ENGINEERING AND TECHNICAL SUPPORT (71707, 40500, 60710)
.....
4.1 Fuel Zone Level Instrumentation......................... 16 4.2 Review of Cycle 14 Safety Analysis Report................... 19 4.3 Redundant Fire Water System Contingencies................... 19 4.4 5 kV Cable Testing Program Change....................... 20 5.0 S ECURITY (71707)..................................... 23 oi 6.0 SAFETY ASSESSMENT / QUALITY VERIFICATION (90700)........... 23 6.1 Review of Licensee Event Reports
........................23
i 7.0 REVIEW OF PREVIOUSLY OPENED ITEMS (92702)
...............
8.0 EXIT MELTINGS (40500, 71707)
............................
8.1 Preliminary Inspection Findings.......................... 28 8.2 Attendance at Management Meetings Conducted by Other NRC -
In spectors....................................... 28
,
iv
.
DETAILS 1.0 OPERATIONS (71707, 93702)
1.1 Outage / Operations Summary During the inspection period the licensee completed the remaining 14R outage activities to support a February 14,1993 plant restart. Section 1.3 of this report contains a description of restart activities.
,
While performing the nuclear steam supply system (NSSS) leak test on January 18,1993,a rapid depressurization occurred when the de powered condensate return valves for the isolation condenser opened. As reactor pressure increased above the hi-hi pressure setpoint (1050 psig) during the test and the anticipated transient without scram (ATWS) circuit initiated, the condensate return valves opened. The surveillance procedure controlling the leak test inappropriately required these valves to be in automatic, allowing them to open when the ATWS signal was received. NRC review of this occurrence was in progress at the end of the inspection period.
When the licensee performed the local leak rate test (LLRT) of the inboard main steam isolation valves (MSIV) on January 23,1993, reactor coolant temperature increased above 212"F. This was due to a degradation in shutdown cooling through the core. Section 1.4 contains a summary of this event that caused thermal stratification within the core region. An augmented inspection team (AIT) was sent to the site to investigate this matter. NRC inspection report number 50-219/93-80 documents the AIT findings. Both inboard MSIVs passed their LLRTs before this event was identified by the licensee.
The primary containment integrated leak rate test (ILRT) was conducted starting on January 27,1993. The ILRT was completed on January 29,1993, after difficulties were
-
resolved. See section 3.4 for a discussion of the ILRT.
Loss of power testing (LOPT) commenced on February 3,1993, for emergency diesel generator (EDG) No.1. The EDG successfully passed the LOPT, however, the crankcase i
overpressure switch prevented idling of the diesel following the test. During the initial LOPT for EDG No. 2 on February 4,1993, the diesel failed to start as required after opening the startup transformer breaker powering the 4160V bus.~ Incorrect wiring of a selenium rectifier caused the start failure. EDG No. 2 performed as required during the LOPT on February 6,1993, following replacement of the selenium rectifier. Section 3.3 discusses the events associated with the LOPT and the licensee's corrective actions to address the deficiencies noted on the EDGs.
I
_
_
,
.
,
q
!
l-
,
1 y
h Restart comme"ced on February 14, 1993. He unit achieved full power on
'
February 19, 1993, following a controlled power ascension plan. Section 1.3 provides a.
j
"
discussion of restart activities. Synchronization of the main generator on February 16,.1993,
.
marked the completion of the 14R refueling outage. The outage, which started on.
-l November 28,1992, was completed in 80 days. This outage duration was within GPUN's -
predetermined schedule, including allowances for emergent work.
l
,
The inspectors observed the plan of the d y and scope control meetings throughout the j
outage. Emergent work was adequately integrated into the schedule. Safety significant z
,
emergent work was not delayed or postponed to support restart. Daily plant risk'
,
configurations were provided to the operations personnel in' the control room. Emergent.
.
i l
work on the MSIVs resulted in delays in performing the inboard MSIV LLRTs. While these
,!
'
delays contributed to the degradation in shutdown cooling event on January 23,' 1993, outage.
-
management decided when to perform the test based on the opportunities provided in the ~
.
outage schedule and engineering assurance that the testing could be done during those time 1 j
frames.
j l
Based on the inspectors' observations during the outage, the licensee's' use of integrated; j
scheduling, risk management based plant configurations, and the team building conducted.
~
_
before the outage contributed to completion of safety related outage work within the
'
predetermined outage schedule. Inspectors observed adequate performance of safety i a
significant work during the outage. Outage management maintained good control'of outage
!
activities.
!
!
1.2 Scram Signal Received During System Restoration Alter ILRT o
.
On January 30,1993, at 12:58 p.m., a full scram signal we received during system I
restoration efforts following the containment integrated le:1 rate test (ILRT). - No actual -
scram occurred because the reactor.was in a cold shutdown condition with the control rods
= '
inserted. Control room operators were performing procedure step 7.21.9_ of the ILRT -.
procedure (666.5.007) which provided direction for restoring 'the control rod drive hydraulic.
.
!
system when the scram signal occurred. The procedure step contained two actions.' %e first :
action was to return the scram discharge volume (SDV) high water level bypass switch to the1 i
normal position. The second action was to open the SDV vent and drain valves.' During the
'{
ILRT, however; test pressure had forced water past a few control rod drive hydraulic control:
ti unit scram valves which had experienced some minor leakage. This scram valve _ leakage j
caused the scram discharge instrument volume to fill, bringing in SDV_ high level alarms
.
j
.
during the ILRT. When the control room operators placed the SDV high level bypass switch t
in normal before opening the vent and drain valves, the scram' signal occurred.
'
'
I
,
t
-
!a
,
a
- 7..
<
The inspectors interviewed the control room operators and reviewed the ILRT procedure step in question. While the operators felt that the multi-action procedure step contributed to the problem, they acknowledged that they could have precluded the occmTence by a more thorough check of the control room annunciator panel lights before performing the procedure step. The licensee appropriately reported this occurrence to the NRC under 10 CFR 50.72 (6) (2) (ii) as an event that resulted in automatic actuation of the reactor protection system.
ILRT test personnel stated that the procedure would be changed to separate these actions and assure that they are performed in the proper order.
1.3 Startup Following the 14R Refueling Outage On February 12, 1993, the unit was in a condition to support restart. Shutdown nurgin and the drywell airlock local leak rate tests (LLRT) were required to be performed before reactor startup commenced. Successful completion of shutdown margin testing occurred late in the day on February 12. Initial attempts to complete the drywell airlock LLRT test failed. The licensee found the cause of the milure to be the inner airlock door. The inner door was opened and rescaled, and the door seal inspected and replaced twice. After the second replacement of the inner
,r seal, the licensee pressurized the drywell airlock with personnel inside the drywell to find the leak. Leakage from the inner door handwheel shaft was noted.
Disassembly of the shaft found a burr on the shaft that had damaged the shaft seal when operating the handwheel. The burr was removed and a new handwheel shaft seal was installed. On February 14,1993, at i1:15 a.m., the licensee successfully completed the drywell airlock LLRT.
Reactor startup commenced at 11:30 a.m. on February 14, 1993, using procedure 201.1, Revision 62, " Approach to Criticality." At 1:24 p.m., the reactor was critical with a 100 second period. Control of the startup was transferred to procedure 201.2, Revision 45,
" Plant Heatup to Hot Standby " By 1:45 p.m. the core was at the point of adding heat, with reactor coolant temperature exceeding 212"F at 2:45 p.m. on February 14, 1993. Reactor heatup continued until reactor pressure reached 1000 psi on February 15,1993. At 1000 psi tne licensee conducted an inspection inside the drywell for leakage. No significant leakage was identified during the 1000 psi inspection. Following the 1000 psi inspection, the licensec rclested the drywell airlock door successfully.
The initial vibration data for the main generator obtained on February 15,1993, were higher than expected. GPUN discussed the causes of the high vibrations with General Electric (GE)
and determined that the cause was shaft rubbing. Based on discussions with GE, GPUN secured the turbine and placed it on the turning gear for four hours to remove the bow from the turbine shaft caused by heating of the shaft from shaft rubbing. After the four hour period, the licensee placed the main generator back on line at 12:05 a.m. on February 16,1993, to support overspeed trip testing of the turbine. Vibration readings during this period were well within prescribed specifications during the four hour stabilization period required for turbine overspeed testin *
.,
,
!
'
'
Turbine overspeed testing started at 4:45 a.m. on February 16,1993, using procedure 625.4.001, Revision 7, " Turbine Overspeed Test and Calibration." Overspeed testing was complete by 6:53 a.m. and the output breaker for the main generator was closed at 7:07 a.m.
'
on February 16, 1993. With the main generator on line the 14R refueling outage was complete.
Drywell inerting started following completion of the 1000 psi drywell inspection on February 15. The licensee had trouble lowering the oxygen level inside containment below
'
the 4% required by technical specification (TS). Investigation by the licensee determined that the sample pump diaphragm for the containment atmosphere particulate and gaseous radiation monitoring system (CAPGRMS) had failed. This provided a pathway between the secondary containment and inside primary containment. CAPGRMS was isolated and oxygen levels decreased below 4% by 10:45 a.m. on February 16, 1993, within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after placing the mode switch in the run position as required by TS 3.5.A.6.
Manual heat balances were performed by the licensee at 25%,50%, 75%, and 100%. All the manual heat balance calculations were conservative when compared to the plant computer generated heat balance. Reactor power was held at 25% and 60% during the startup for fuel conditioning and to allow xenon equilibrium to be established. After returning the main generator to senice following overspeed testing, the licensee noted that the B condenser vacuum was lower than the vacuum in the A and C condensers. Reactor power was held at 60% until the cause of the vacuum tilt was determined. The licensee found that the steam trap downstream of the after-condenser for the steam jet air ejector (SJAE) for the B condencer had failed closed. This caused the drain line and after-condenser to fill with water, reducing the effectiveness of the SJAE. The line was drained and the trap repaired. B condenser vacuum increased to the same value of the A and C condensers (about 29" vacuum).
.
Reactor power was slowly increased following the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> hold at 60%, and on
.
February 19, 1993, the unit was operating at 100% power.
Inspectors provided extended startup coverage from February 12 through February 16,1993.
During the extended coverage, the inspectors observed control room operators (CRO), group shift supervisors (GSS), group operating supervisors (GOS), shift technical advisors (STA),
core engineering personnel, and operations management respond to changing plant conditions
,
and emerging challenges. The inspectors observed performance of and/or reviewed the
following procedures throughout the startup:
1001.27, Revision 16, " Shutdown Margin Measurement Test"
201.1, Revision 62, " Approach to Criticality"
201.2, Revision 45, " Plant Heatup to Hot Standby"
q
--
- ,
,
1001.6, Revision 13, " Core Heat Balance - Power Range"
+
620.4.005, Revision 21, " Intermediate Range Monitor Test and Calibration (Front
Panel Test)"
602.4.003, Revision 15, "Electromatic Relief Valve (EMRV) Operability Test"
625.4.001, Revision 7, " Turbine Overspeed Test and Calibration"
During the startup the inspectors also observed the CROs placing the systems necessary to support plant operation in service using the appropriate normal operating procedures
'
associated with the systems.
As an aid in performing an effective startup, the licensee had prepared a power ascension plan. The power ascension plan provided detailed guidance to the GSS directing shift activ: ties on what actions to accomplished on each shift. He inspectors reviewed the power ascension plan, which was consistently referred to by operations depanment personnel throughout the startup. The inspectors concluded that the power ascension plan provided for an orderly approach to reactor startup following the 14R refueling outage.
Before startup, the inspectors reviewed the precritical checkoff of procedure 201.1. Required approvals and cenifications were made before the startup commenced. Shutdown margin testing adequately demonstrated that the TS required shutdown margin was available.
Inspector review of the manual heat balances performed using procedure 1001.6 found that the computer generated heat balance was conservative in each case. At 50% power the
manual heat balance was within 3.45 MWth of the computer generated heat balance.
As the licensee was preparing to do the shutdown margin test, power was lost to a lighting
distribution panel (l Al lighting distribution panel). This caused a loss of power to turbine building lights and fire detectors in the 4160V switchgear mom and the emergency diesel l
generator building. The cause of the power loss was work inside a unit substation (USS) that provided power to the lighting distribution panel. The licensee was changing a temporary power supply from the USS to the turbine building operating floor to a permanent power supply. He power supply lead was being removed from the USS when it contacted energized components inside the USS. Rubber sheeting covered the energized components to prevent contact, however, it appears that the precautions were not adequate to prevent the inadvertent ground when removing the temporary power lead. The licensee inspected and cleaned the USS internals, and reset the power supply to the lighting distribution panel.
While the fire detectors were inoperable, the licensee established fire watches as required by TS. Power was restored before the licensee commenced the shutdown margin testing.
,
t
..
.
,
Overall, activities observed by the inspectors during startup were well conducted. High vibrations encountered on the initial start of the turbine were addressed in a timely and appropriate manner. Difficulties with performing the drywell airlock LLRT were eventually
'
resolved by the licensee. The CAPGRMS sample pump diaphragm failure was identified and
,
corrected appropriately. Opemtors controlled changes in plant conditions (increasing power, placing equipment in service, etc.) using the appropriate plant procedures and as directed by the guidance provided by the power ascension plan. Core engineering personnel, STAS and CROs worked well together during the approach to criticality. Operations Management provided good oversight during the startup. Surveillances observed by the inspectors were completed successfully and done in a professional manner, meeting the acceptance criteria specified in the procedure. Inspector review of the completed EMRV surveillance procedure (602.4.003) found that all five EMRVs operated as designed during the test.
The inspectors concluded that the plant startup following the 14R refueling outage was well controlled, effectively accomplished, and generally uneventful.
1.4 Degraded Shutdown Cooling Event Summary On January 23,1993, the licensee placed the plant in a condition that eventually led to thermal stratification within the core region. To perform a local leak rate test (LLRT) of the inboard main steam isolation valves (MSIV) in parallel with recirculation system work, the licensee established an alternate shutdown cooling (SDC) configuration in the plant. Reactor vessel isolation permitted pressurization for the LLRT. Imwering reactor vessel level to the normal operating band of 155 to 165 inches above the top of active fuel (TAF) established a vent path using the isolation condenser lines (180 inches TAF) to restore plant conditions following the LLRT. An open recirculation loop provided hydraulic communication between
'
the core region and the annulus as required by technical specifications. Recirculation pumps were secured.
Normal opemtion of the shutdown cooling (SDC) system is with level above 185 inches TAF
.
with all recirculation loops idled (the pump discharge valves closed) or if below 185 inches TAF, a recirculation loop is open with the loop recirculation pump running forcing flow through the core. In the alternate configuration, part of the SDC system flow bypassed the core through the open recirculation loop.
GPUN technical functions engineers were developing an analysis that supported operation of the SDC system at levels below 185 inches above TAF without a recirculation pump running r
by requiring increased SDC flow to compensate for the flow bypassing the core through the l
open recirculation loop. The temporary procedure change (TPC) developed to support operation in the alternate configuration used the draft engineering document provided by the technical functions engineers. The increased SDC flow requirement was omitted from the TPC to the SDC system operating procedure.
i
-
D
'.
At 4:30 p.m. on January 25,1993, a site operations engineer and a shift technical advisor noted elevated reactor vessel metal temperatures on a recorder that was moved to the reactor building from its former location in the control room. After being notified of the elevated metal temperatures, operators responded quickly to increase SDC flow and establish primary containment. Technical specifications required primary containment integrity when reactor
.
water temperature increased above 212 degrees F. The estimated reactor coolant temperatures in the core region were between 250 and 270 degrees F.
The NRC sent an augmented inspection team (AIT) to the site on January 26,1993; to review this matter. The team found the root cause of the event to be an inadequate SDC system operating procedure that did not include the flow requirements provided by the -
engineering evaluation and made no provisions for monitoring the effects of the plant
.'
configuration change. Although the event was found to be of low safety significance, the AIT noted a significant programmatic weakness in the licensee's implementation of the i
temporary procedure change process. These findings were discussed with the licensee at a
,
public meeting on February 8,1993.
1.5 Facility Tours i
The inspectors observed plant activities and conducted routine plant tours to assess equipment conditions, personnel safety hazards, procedural adherence and compliance with regulatory
,
requirements. Tours were conducted of the following areas:
intake area
'
control room
reactor building cable spreading room
turbine building diesel generator building
a vital switchgear rooms new radwaste building
access control points old radwaste building
drywell transformer yard
Control room activities were well controlled and conducted in a professional manner.
Inspectors verified operator knowledge of ongoing plant activities, equipment status, and existing fire watches through random discussions. Housekeeping activities after completion
of the 14R refueling outage recovered the majority of areas contaminated during the outage very quickly. Reactor building, turbine building, and yard areas were restored to pre-outage j
conditions shortly following the outage.
,
,
2.0 RADIOLOGICAL CONTROLS (71707)
During entry to and exit from the radiologically controlled area (RCA), the inspectors verified that proper warning signs were posted, personnel entering were wearing proper
,
dosimetry, personnel and materials leaving were properly monitored for radioactive contamination, and monitoring instruments were functional and in calibration.. Posted extended Radiation Work Permits (RWPs) and survey status boards were reviewed to verify i
,
-
.~
.
-
.
.
.
.s
-
.,
!
i
..
that they were current and accurate. The inspector observed activities in the RCA and _
{
verified that personnel were complying with the requirements of applicable RWPs and that -
workers were aware of the radiological conditions in the area.-
3.0 MAINTENANCE / SURVEILLANCE (62703, 61726, 62705)
3.1 Replacement of Station Batteries A and B (URI 56-219/93-03-01)
!
-
The inspectors observed portions of the replacement and post-modification testing of the A
>
and B main station batteries. The inspectors also reviewed the documentation related to these efforts.
.
The A and B station batteries were replaced with AT&T Lineage 2000 Round Cells, Model l
KS-20472 List IS. The new battery cells have cylindrical PVC shells containing horizontal
circular plates. The lead-calcium batteries are mounted in a rack made of a polyester-glass
material with a metal frame around it. The batteries and racks are environmentally and seismically qualified. The AT&T Round Cell 1600 ampere-hour discharge capacity at a
!}
nominal 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> discharge rate (188 amperes) is larger than the 1200 ampere-hour capacity for j
the Gould FTA-21 lead-antimony batteries which were replaced. The nominal operating j
voltage of the new batteries under float charge is 2.17-2.22 volts / cell. The nominal specific
!
!
gravity is 1.215 i.005. These nominal operating values for cell voltage and specific gravity :
are essentially the same as those for the C station batteries, which were not replaced this outage. The C batteries are Gould NCX-1200 lead-calcium batteries with a 1200 ampere-t hour capacity at the eight-hour discharge rate.
'
v Overall, the A and B station battery replacement work was done well and initial chhrging and
post-modification testing of the new batteries were completed successfully. The new storage
.
racks were assembled and the battery cells installed within the racks in accordance with the -
,
manufacturer's specifications. The work package contained appropriate' guidance to - ~
accommodate both the safe removal of the old batteries and the installation and testing of the I
new batteries. Cable terminations were done appropriately and connecting cables were
+
adequately supported.
t While reviewing the engineering safety evaluation for the modification which was included with the work package, the inspectors noted that no change to the plant technical.
specifications (TS) was being pursued because the operating parameters for the new B station q
batteries were essentially the same as the batteries that.were replaced (the A station batteries l
are not safety-related and are not addressed in the TS). The inspectors verified that the.
~
operating' parameters for the old and new batteries were the same but questioned.the
. ;
application of the existing TS requirements to the operating characteristics of the batteries.
-!
TS surveillance requirements 4.7.B.1.b. and 4.7.B.1.d. require weekly checks of battery pilot j
cell voltage and specific gravity. TS surveillance requirements 4.7.B.2.b. and 4.7.B.2.c.
require quarterly checks of cell voltage under float charge and specific gravity for each cell.~
ti Although not stated in the TS, the inspectors verified that the weekly pilot cell checks are -
,
!
,
&
'.
.
!
,
'
also done under float charge. The TS limits for both the weekly and quarterly tests are cell voltage greater than or equal to 2.0 volts per cell and specific gravity greater than or equal to
,
1.190. The minimum TS value for specific gravity appeared to be a reasonable value for the y ower operating range of this type of battery under float conditions. However, the minimum
'
j TS value for individual cell voltage did not appear to be a reasonable number, particularly l
since the weekly and quarterly surveillances are performed with the batteries under float conditions.
,
<
The licensee requested the battery vendors (Gould and AT&T) to provide information regarding the relationship between cell float voltage and specific gravity as well as the
,
relationship between specific gravity and open cell voltage. The vendor responses described a
general relationship between cell float voltage and specific gravity for lead-acid batteries. In r
general, there is a 3% change in battery capacity for every 0.010 change in specific gravity.
For example, if a cell operating at 2.22 volts under float and 1.215 specific gravity has its cell voltage reduced to 2.0 volts, the specific gravity will drop to about 1.178. There is also a general relationship between specific gravity and open cell voltage for all lead-acid batteries. In a given cell, open cell voltage equals specific gravity plus 0.84. For example,
at the Oyster Creek TS minimum value for specific gravity (1.190), open cell voltage would
be 2.03 volts.
After receiving the information from the vendors, the licensee acknowledged that the TS limits for cell voltage and specific gravity were not interrelated. While the TS limit for specific gravity appeared to be related to battery-specific operating characteristics, the licensee described the TS limit for cell voltage as being related to the design bases for Oyster Creek de loading requirements. In other words, assuming a loss of all ac power subsequent to a loss of coolant accident (LOCA), the batteries at 2.0 volts per cell should have enough capacity to support the de loading requirements that result from this plant condition. At 2.0 volts per cell, battery capacity is effectively reduced by about 10%. The inspectors reviewed the de load profile calculations for the B and C station batteries to determine the available capacity margins. The B station battery has about 50% excess capacity and the C station
,
battery has about 160% excess capacity based on the load profile calculations.
The inspectors also reviewed the current surveillance procedures for the weekly and quarterly battery tests to determine whether and how they addressed the TS acceptance criteria. The TS minimum values for cell voltage and specific gravity are included as the procedural acceptance criteria for procedures 634.2.002 and 634.2.003, the respective procedures for the weekly and quarterly battery surveillance tests. However, further review of the procedure showed that these values are not actually what is used to determine acceptability of the battery cells during testing. Steps within the body of each procedure (Step 6.11 for procedure 634.2.002 and Step 6.10 for procedure 634.2.003) addressed battery operability in terms of normal operating characteristics. Specifically, these procedure steps require that if a cell voltage less than 2.13 or a specific gravity less than 1.205 is measured, then an equalizing charge shall be placed on the battery to restore the battery to a fully charged
'
condition.
,
t
.
'.
The inspectors concluded that performance of the weekly and quarterly battery surveillance tests will assure that the batteries are maintained in an operable condition. The inspectors also concluded that operating the batteries at the minimum TS value of 2.0 volts per cell would represent adequate battery capacity to meet design bases de loading requirements, even though the minimum TS value for specific gravity would not be met in this condition.
However, the inspectors noted that the minimum values for cell voltage and specific gravity
,
in the Oyster Creek TS and in the station battery surveillance procedures are inherently confusing due to the inconsistency between these values and the normal battery operating characteristics. The following inconsistencies were noted:
1.
The interrelationship between cell voltage and specific gravity is not addressed. If a battery cell were actually operated at the minimum TS values (2.0 volts,1.190 specific gravity), the cell would be inoperable because the relationship of these values would be indicative of material degradation of the battery internals.
2.
Indicating the TS minimum values as the acceptance criteria for battery surveillance testing is not realistic based on the normal operating characteristics of the batteries.
3.
The current surveillance procedure acceptance criteria are at the TS limits. No margin is provided above the TS limits.
4.
TS surveillance requirement 4.7.B.I.b. does not indicate that the weekly pilot cell voltage check is done under float conditions.
The licensee acknowledged the inspector's concerns and indicated that they had already planned to clarify licensing basis, design basis, and operating and maintenance documentation related to station battery operating characteristics. This issue is considered unresolved (URI 50-219/93-03-01) pending licensee actions to clarify the TS surveillance requirements, procedural acceptance criteria and other system related information describing the station battery operating requirements for cell voltage and specific gravity.
3.2 Average Power Range Monitor Trouble Shooting
[
On February 11,1993, the inspector observed instrument and control (I&C) technicians perform troubleshooting on the average power range monitors (APRM). APRM 2,3 and 4 had failed a surveillance test (Procedure 620.4.002, Revision 8, " Average Power Range Monitor Front Panel Test") performed in preparation for restart following the 14R refueling outage. Job order 45570 controlled trouble shooting and adjustments to the APRMs.
For APRM channel 3 and 4 both failed the front panel test when the zero set point read below zero and the 150% reading was only 147%. APRM channel 2 failed the front panel test when the zero setpoint read below zero.
- -.
. -
-
- -
'.
.
>
The inspector observed the 1&C technicians working on APRM channels 3 and 4. During troubleshooting on APRM 4, the I&C technician readjustal the mechanical zero following the guidance provided by the vendor manual. APRM channel 4 was within the specified tolerances of the front panel test after the mechanical zero adjustment. To correct the
,
'
conditions for APRM channel 3, the I&C technician adjusted the negative voltage input to the
-
APRM's power supply using the guidance provided by the vendor manual. Following this-
,
adjustment, APRM channel 3 was within specifications.
During the work observed by the inspector, the I&C technician used the vendor manual and appropriate parts of the surveillance procedure to adjust or check the APRM response. The job order provided the necessary instructions to perform the work. Overall, the work was coordinated well with control room operators and adequately performed.
3.3 Loss of Power Test The loss of power test is a surveillance that must be performed prior to restart. Each emergency diesel generator (EDG) train is tested separately. A normal electrical system lineup is established to start the test with the two non-safety related 4160V busses (A and B)
powered from separate startup transformers. The safety related 4160V busses (C and D) are powered by the A and B 4160V busses respectively. To perform the test, the output breaker from the respective startup transformer (SI A and SIB) is opened, removing power from the
'
non-vital 4160V bus. The EDGs must start and power the safety related bus within twenty seconds of the loss of power.
Surveillance procedure 636.2.001, Revision 26, " Diesel Generator Automatic Actuation Test," provides the instructions for performing the loss of power test. The test of the EDGs performance during a loss of offsite power (LOOP) includes the loads that the EDGs powers during a loss of coolant accident (LOCA). The surveillance procedure directs the output breaker for the startup transformer to be opened, followed in seven seconds by a simulated reactor vessel lo-lo water level (LOCA signal). The 7 second delay prevents multiple core spray pump motor starts in rapid succession and thereby avoids potential motor damage.
Following the input of the LOOP and LOCA conditions, the containment spray and emergency service water (CS/ESW) pumps are sequenced on manually. This demonstmtes the ability of the EDGs to accept large loads without faults after the automatically sequenced loads are powered by the EDG.
The loss of power test for EDG #1 was performed on February 3,1993. During this test EDG #1 performed acceptably. While securing EDG #1 following the surveillance, the EDG tripped while placing the unit in the idle cycle. Engine crankcase pressure indicated positive and the overpressure switch tripped causing the EDG to sto '
,
.
The licensee replaced the crankcase overpressure switch and performed an operability sun'eillance before returning the EDG to an operable status. The crankcase overpressure switch is removed from the EDG protective circuitry during fast starts as evidenced by successfully starting the EDG during the loss of power test. Based on discussions with the vendor, the licensee determined that the most likely cause of the crankcase overpressure indication was oil impingement on the crankcase overpressure switch when the EDG was fast started for the test. At the next availabic opportunity, the licensee plans to install a baffle plate to prevent the oil slung off the camshaft from impinging on the crankcase overpressure switch.
!
On February 4,1993, at 8:30 p.m., the licensee attempted to perform the loss of offsite power test using EDG #2. When the output breaker for the startup transformer powering the
,
B 4160V bus was opened, EDG #2 did not start. Control room operators (CRO) and the group shift supervisor (GSS) took action to recover from the conditions established by the surveillance procedure and restore offsite power to the B and D 4160V busses. The B 4160V bus was reenergized by 8:43 p.m., and by 10:00 p.m., the D 4160V bus was reenergized following restoration of the electrical lineup and recovery from the loss of power test.
Review by the licensee determined that a selenium rectifier (EMD model number 8244517) in the fast start circuitry for EDG #2 had been installed incorrectly. The undenroltage signal was received by the fast start circuitry and the pulse relay generated the one second signal
'
required to seal in the fast start signal. However, because the selenium rectifier was installed incorrectly, the signal was not passed to the unit start relay.
During replacement of the selenium rectifier the licensee found that the leads on the installed rectifier were reversed. The licensee had performed an upgrade of the EDG control circuitry during the 14R outage that required rewiring of the effected rectifier. When the rectifier was rewired, the leads were swapped. With the leads swapped, the selenium rectifier polarity was reversed causing the rectifier to block the signal in the wrong direction. Licensee field testing of a new selenium rectifier, found that when subjected to a two second reverse direction current, the polarity of the selenium rectifier reverses. This allowed the rectifier to successfully pass two fast start signals during diesel post maintenance testing. Before performing the loss of power test, the licensee had tested the fast start circuitry for EDG #2 on two occasions without incident following the control circuit upgrade modification. After the leads were correctly installed, licensee testing verified that the selenium rectifier functioned properly.
I
.
!
i
A contributor to the incorrect wiring of the scienium recti 6er may have been poor polarity l
markings on the construction drawings. In addition, the markings on the rectifier may have
'
made it difficult to identify the incorrect wiring in the field. The licensee was still reviewing the event to determine the root cause of the incorrect wiring. The licensee had reviewed the installation of the same selenium rectifier on EDG #1 and found that the wiring was correct.
l Licensee review of the installation of other rectifiers in the EDG fast start circuitry identified
'
no discrepancies. Also, other rectifiers used in the fast start circuitry did not display the i
polarity reversal when subject to reverse current flashing displayed by the EMD model #
8244517 selenium rectifier. The licensee has provided this information to the industry.
Following replacement of the selenium rectifier, the licensee performed a successful
'
operability test of EDG #2. On February 6,1993, the licensee reperformed the loss of
,
power test for EDG #2. The EDG started and loaded as required.
j
.
Licensee review and analysis of the EDG #1 failure to idle was appropriate. The bypass of
EDG protective functions during a fast start was demonstrated during testing of EDG #1.
Licensee analysis of the cause of the EDG #2 start failure was adequate. Corrective actions for the EDG #2 start failure were adequate to support performance of the loss of power test l
'
and demonstrate EDG #2 operability. Licensee review of the cause for the incorrect wiring of the selenium rectifier in the EDG #2 fast start circuitry was ongoing at the end of the inspection period.
During the response to the failure of EDG #2 on February 4,1993, the inspector observed
that the control room operators (CRO), group shift supervisor (GSS), and the group operating supervisor (GOS) appropriately used operating procedures to restore the electrical plant lineup to normal. Also, appropriate sections of the surveillance procedure were used to restore normal plant conditions after the failed test. The GSS took action to restore the electrical lineup within a reasonable time of determining that EDG #2 had failed to start. Overall, recovery from the failure of EDG #2 was well controlled.
The inspector observed the initial test of EDG #2 on February 4,1993, reviewed the results of the tests for EDG #1 and #2, and discussed the failures for both EDGs with the system
,
,
engineer. During the successful tests, the EDGs were shown to have adequate capacity to power vital loads in a loss of offsite power coincident with a loss of coolant accident. Minor discrepancies were noted in the shedding of non-vital loads (such as a waste collector pump and radwaste building floor drain sump pump) that did not increase EDG loading significantly. The licensee has issued work requests for each discrepancy noted during the test.
,
e
'
.
.
f
i 3.4 Containment Integrated Leak Rate Test
The inspector reviewed the test procedure (Procedure 666.5.007, " Primary Containment
Integrated Leak Rate Test," Revision 16, dated December 18,1992), and related documents for technical adequacy and to determine compliance with the regulatory requirements of Appendix J to 10 CFR 50, technical specifications (TS), and applicable industry standards.
'
The inspector witnessed a large portion of the integrated leak rate test (ILRT) and subsequent verification test. The inspector also performed an independent calculation of the test results.
t The inspector reviewed the "as-run" copy of the ILRT procedure with related changes, attachments and test log for technical adequacy and for consistency with regulatory requirements, guidance, and licensee commitments. Review of procedure acceptance criteria, test methods, and references indicated conformance with Appendix J to 10 CFR 50 and the
,
Oyster Creek TS. The ILRT valve lineups were reviewed to ensure that systems were properly vented and drained to expose the containment isolation valves to containment atmosphere and test differential pressure with no artificial boundaries. A large sample of valve lineups was physically verified by the inspector during tours taken before and during the performance of the test. The test was performed and the test log and test data were maintained in accordance with the procedure. A procedural step containing multiple actions partially contributed to a reactor scram signal which was received while restoring the control rod drive hydraulic system after completion of the test. This is discussed in more detail in Section 1.2 of this report.
The inspector reviewed the calibration records for the resistance temperature detectors, dewcells, and precision pressure detectors used for the ILRT. The calibrations met applicable accuracy requirements and were traceable to the National Bureau of Standards. The inspector
>
evaluated the ILRT data acquisition system and the program used to interpret the raw data and perform the leak rate test calculations. The test sensors were appropriately located and l
were calibrated in place after installation to account for signal loss imposed by the run of cable from the containment to the data acquisition unit. The data were then transmitted to a computer for calculation via a serial communications link. The inspector verified that the
.
computer program appropriately accounted for sensor volume fractions and included proper
,
calculations for the reassignment of volume fractions, if necessary during the test, due to sensor failure. The program was broken down into distinct test modes to accommodate pretest sensor verification, containment pressurization, verification of temperature stabilization prior to test commencement, the leak rate test itself, and the imposed leak verification test.
The inspector concluded that the data acquisition system was providing proper data input to the computer program and that the program performed the appropriate test calculations,
'
including the accommodation of potential test variables such as sensor failure and water level changes.
.
'.
'
,
The inspector also reviewed the licensee's calculations for the volume fractions for the RTDs and dewcells used for the ILRT. The inspector independently calculated some of the sensor'
,
volume fractions and verified the licensee's calculations.
A large portion of the ILRT test sequence was witnessed by the inspecto. The test -
-
chronology was as follows.
'
TEST CHRONOLOGY 1/21/93 Completed pre-test inspection of containment.
1/26/93 1200 Commenced pressurization of containment.
1/27/93 0245 Reached test pressure (38.7 psia)..
0800 Temperature stabilization criteria satisfied. Test commenced.
1430 Excessive leakage noted. Leak searches in progress.
Valve lineups being reverified. Valve pacidng w2s tightened for several valves. Other valve fitting adjustments were made. Drywell sump discharge and-drywell equipment drain tank discharge isolated, as drywell sump level variations were affecting test data.
1/28/93 A.M.
Reestablishment of temperature stabilization as licensee-acknowledged that repair efforts had immlidated prior test data.
i 1015 Test restarted.
1/29/93 1015 Completed data acquisition for ILRT. - Measured leak rate M = 0.372 wt%/ day. Leakage at 95% upper confidence limit (UCL) = 0.376 wt%/ day.
1130 Imposed leak of about 1.0 wt%/ day for supplemental verification test.-
1530 Acceptance criteria met for supplemental verification test.
F
.
-
t
,,
,
The licensee evaluated the test results for the 24-hour period starting at 1015 on January 28,1993, and ending at 1015 on January 29,1993. The calculated leakage as-left rate at the 95% upper confidence level was 0.376 wt % per day. The as-left test acceptance criterion at the reduced test pressure of 23.5 psig was 0.615 wt % per day (0.751 ). I, is
the adjusted total leakage limit based on the reduced test pressure (I3 = L (P/P.)")
The calculated as-left leakage rate was properly corrected for the leakages from the penetrations isolated after the aborted first ILRT attempt as well as the Type C leakage additions from other penetrations isolated or in use during the performance of the test. The as-left leak rate was also corrected for sump level changes which occurred during the test period. The final as-left leakage rate, including corrections, was 0.401 wt % per day. The technical summary report, required within 90 days of test completion by 10 CFR 50, Appendix J, will include both as-found and as-left integrated leak rate using pre-and post-repair leak rates to account for leakage improvement. The as-found integrated leak rate is calculated by adding the minimum pathway leakage improvement accomplished by outage maintenance as monitored by local leak rate test results to the corrected as-left ILRT leak rate (LLRT).
The inspector reviewed the LLRT results summary and discussed the analysis of the test results with the licensee. The inspector found that the licensee had appropriately reported the failure of Type C testing performed on main steam isolation valve (MSIV) NSO4B (outboard MSIV on B steam line) in a licensee event report (LER 92-013) as total Type B&C leakage was determined to be greater than the TS limit of 0.6La. The cause, corrective' actions, and safety significance of this LLRT failure will be reported in a supplemental LER. The inspector was satisfied with the licensee's understanding of the application of local leak rate test results to as-found and as-left conditions of containment as applied to ILRT results.
The inspector performed an independent calculation of the results of the second ILRT attempt using a sample of raw data from the test. The results of the leak mte and verification test independent calculations verified the accuracy of the licensee's leak rate calculations.
4.0 ENGINEERING AND TECHNICAL SUPPORT (71707,40500,60710)
4.1 Fuel Zone Level Instrumentation Section 4.1 of NRC Inspection Report 50-219/92-23 discussed a design deficiency related to the fuel zone reactor vessel level (FZL) indication which was identified by GPUN in November 1992. The problem dealt with possible errors in observed level indication after the FZL instrument switched from the narrow range to the wide range detectors during accident conditions when the core region is voided and dynamic losses are occurring from steam or two-phase flow across the instrument level taps. After an initial engineering evaluation, GPUN provided guidance to the control room operators which described the problem and the potential magnitude of the indicated level ermrs. A graph of indicated wide range fuel zone vs. actual level, assuming the worst case error, was provided to the control-
._.
'
!
,
.
,
t room operators for use as an operator aid until the plant was placed in cold shutdown for the
'
14R refueling outage. Use of the operator aid was not necessitated prior to the plant reaching
'
cold shutdown on November 28,1992. Software changes were made to the programmable logic controllers for the FZL instrumentation during the 14R outage to resolve the problem.
The existing system consists of four narrow range and four wide range Rosemount detectors.
The narrow range FZL instruments measures level from +55 inches TAF to +180 inches.
i The variable legs for the four narrow range FZL channels, A through D, senses level at the
,
core spray spargers (+55 inches TAF). The wide range FZL instruments measure level from about -144 inches TAF to +180 inches TAF. The indicated wide range FZL for channels A i
and C are slightly different than that for channels B and D, due to the different locations of the reference pressure taps. The wide range FZL reference pressure tap for channels A and C is at the lower core plenum, while the pressure tap for channels B and D is at the bottom of the bypass region above the lower core plate. The difference in the instrument differential pressures (dp) is the dp across the lower core plate. Normally, channels A and C wide range FZL will be slightly greater than channels B and D with upward flow through the core. In a
reverse flow condition, channels B and D wide range FZL will be slightly greater than channels A and C.
In the plant condition prior to modification, the channel A and B FZL instrumentation
,
became operational when the following three conditions were satisfied: 1) the reactor j
recirculation pumps were tripped; 2) water level was within the range of the instruments; and,3) there was no flashing in the impulse lines. The channels C and D instruments did not employ the recirculation pump trip interlock, but instead were activated by a switch on the
,
control room panels. Similar control switches for channels C and D are located on the remote shutdown panel in the vital 480 volt switchgear room. If reactor vessel water level is greater than +55 inches TAF and the core spray system parallel isolation valves are closed, the narrow range instrument provides level indication. When level is less than +55 inches TAF or the core spray system parallel isolation valves are open, the wide range instrument provides level indication. If a saturation condition (flashing) was detected in the reference
leg, the instruments would turn off. The problem discovered by GPUN was related to the application of a fixed indicated level compensation factor to the wide range FZL whenever the control room (or remote shutdown panel) meter switched from narrow range to wide range indication. This compensation or K factor was intended to allow a 'bumpless" transfer of indication when the narrow range to wide range switchover occurred. However, GPUN review postulated that actual level could be changing more or less than the compensation provided by the fixed K factor, supplying potentially confusing information to the operators.
GPUN evaluated several plant transient scenarios using the existing FZL indication circuitry to assess equipment response and whether the information provided to the operators could be effectively used to determine level conditions in the core region. The thermal-hydraulics computer program and the instrumentation model for the Oyster Creek plant-specific simulator were used to perform the postulated scenarios. The seven transient scenarios
'
included variably sized loss of coolant accidents with and without feedwater, main steam line
!
>
v
..
_
--
.
.
>
.
.
!
!
breaks with and without feedwater, normal loss of feedwater, and a trip of all five i
recirculation pumps. As a result of the transient analyses, GPUN concluded that the existing
-
instrument logic would provide confusing information to the operators in transient conditions.
'
Application of the K factor to the wide range FZL indication could provide unnecessarily
non-conservative level indication to the operators. The transient analyses showed that
'
application of the K factor caused the indication to be out of the range of the meter for long
'
periods of time. The transient analyses also showed that the indication was turned off a number of times due to a portion of the instrument logic which shuts off the indication when (
the K factor is greater than 1.25. GPUN concluded that exclusive use of the wide range FZL l
instruments was preferable to either maintaining the existing indication circuit logic or
-
suppressing the use of the K factor while maintaining both narrow range FZL and wide range i
FZL in service.
During the 14R refueling outage, a plant modification was implemented to resolve this deficiency by simplifying the fuel zone level indication provided to the control room operators. The modification consisted of removing the indication provided by the narrow
,
range FZL channels, allowing the use of wide range FZL for all conditions. This was accomplished by software changes to the programmable controller logic which input dummy
'
,
signals to indicate that the core spray valves are open (for FZL channels A and B) and that l
the manual switch has been activated for FZL channels C and D. Use of these input signals
,
ensures that wide range FZL will always be displayed unless other conditions which cause the
'
FZL instruments to be inoperable are in effect, i.e., level outside of instrument range, saturation (flashing) in the reference leg.
l Operations and training department personnel discussed the effects of this plant modification with the control room operators prior to restart from the 14R refueling outage. The Boiling
Water Reactor Owners Groups (BWROG) guidelines for use of the fuel zone level instrumentation were stressed. This included maintaining an awareness of the indicated level
'
differences between the FZL instruments and those measuring level in the reactor vessel annulus due to differences in voiding in the instrument level columns. The operators were
,
instructed to be cognizant of level responsiveness as an indication that the meters were
.
!
reading properly. The operators were reminded that in dynamic plant conditions it can be difficult to know what the exact reactor water level is and that equally important factors in these conditions are knowing that level is above TAF as well as the level trend. No EOP changes were required as a result of this plant modification. However, the operators were reminded of the EOP guidance to initiate reactor vessel flooding if they feel they cannot determine the reactor water level.
Training modules were being developed for upcoming operator requalification classroom and simulator sessions. These discussions are to include level control issues in general and the FZL indication issue in particular. GPUN has expedited efforts to install the FZL modification on the Oyster Creek simulator. The simulator modification was to be installed and operational shortly after the end of this inspection period.
l
!
i
l
s
,
.
The inspectors reviewed the engineering safety evaluation and modification design description i
related to the FZL instrument modification. The inspectors reviewed the latest revisions of surveillance procedures 664.3.004, "FZL Monitoring System Processor Surveillance Test
'
(Channels A and B)," 664.3.005, "FZL Monitoring System Processor Surveillance Test (channels C and D)," and 664.3.006, "FZL Channels C and D Ancillary Components
'
Calibration and Test," to assure that the plant modification had been appropriately incorporated. The inspectors also interviewed operators, training personnel and engineering
,
personnel regarding their input to and knowledge of the effects of this modification. The
,
inspectors concluded that the licensee had performed a thorough evaluation of this issue and had developed an appropriate plant modification to resolve it. The plant modification appeared to provide the control room operators with the most useful and least confusing
'
information available from the FZL instrumentation.
!
4.2 Review of Cycle 14 Safety Analysis Report j
The inspector reviewed the reload information and safety analysis report for operating cycle 14 (Topical Report 090, Revision 0, dated 12/22/92). This report documented the expected Oyster Creek core performance during fuel cycle 14. The results of this review are detailed below.
i
'
The shutdown margin analysis included in calculation C-1302-226-541-267, Revision 0, was reviewed. The calculation documented that the minimum shutdown margin during cycle 14
)
operation would be 1.595% delta K/K at 7 GWD/MT. The inspector verified that the above shutdown margin was higher than the limit of technical specification section 3.2 (1.0% delta K/K). The inspector reviewed the fuel assembly mislocation analysis which evaluated the consequences of changes in the local pin power distribution and bundle reactivity due to a j
fuel assembly loaded in an improper location of the core. Sixteen control cells were analyzed for this event. With this transient, the operating minimum critical power ratio (MCPR) was determined to be 1.29. The inspector also reviewed the loss of feedwater heating analysis which determined the change in MCPR as a result of cooler water entering the core. The analysis considered a 100*F decrease in feedwater inlet temperature and a 10% increase in feedwater flow. With this transient, the operating limit for MCPR was determined to be 1.27. The technical specification MCPR operating limit for cycle 14 will be established as 1.47. Based on these selected samples, the inspector concluded that the results of cycle 14 safety analysis were well within the technical specification requirements.
4.3 Redundant Fire Water System Contingencies As a result of the failure of a mobile crane at the Oyster Creek intake structure on
'
December 2,1992, the redundant fire water was effectively rendered out of service. The top of the crane boom struck the redundant fire pump house structure and also damaged the discharge and recirculation lines of the redundant fire water pump. The redundant fire water pump and motor and the 350,000 gallon redundant fire water tank were not damaged.
Although the redundant fire water pump and tank are not safety-related, they provide a
'.
,
.
convenient means of backup fire suppression should the main fire pumps be unavailable.
y Specifically, Technical Specification (TS) 3.12.B.3.a. requires that in the event of failure of
'
the main fire pumps, backup fire suppression capability shall be provided within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
The inspectors assessed the licensee's contingency plans for providing backup fire suppression
,
while the system remains out of service, as well as plans for system restoration.
Shortly after the crane failure, site fire protection engineering provided the operations i
department with more specific instructions for actions to be taken if the main fire diesel pumps or the 14 inch main fire header were lost due to the degraded condition of the redundant fire water system and intake canal accessibility concerns at the time.
Procedure 333, " Plant Fire Protection System," has since been revised to reflect more-specific guidelines if the main fire water system is inoperable and the redundant fire water system is incapable of providing backup fire suppression capability. GPUN also chose this time to renew their support agreements with the local fire departments in Waretown, Lanoka Harbor, Forked River, and Bayville due to the increased possibility of needing offsite fire support should the main fire pumps or header become inoperable.
Since the redundant fire water system is not specifically required to be operable for compliance with the plant TS, engineering efforts to support system restoration were deferred until other priority projects were completed during the 14R outage. In early January 1993,
the redundant fire water tank was drained and inspections were performed by plant-engineering. At that time, plant engineering began efforts to partially' restore the system by developing a modification that provided for the installation of a 6 inch tee connection to the discharge piping of the redundant fire water tank. The tee connection will provide a flow path to an alternate pump, if necessary. The modification will remain'after the redundant fire water pump is returned to service. This connection will provide a means to access the water -
inventory in the redundant fire water tank, even if the redundant fire water pump is inoperable.
At the conclusion of the inspection period, the licensee was installing the tee connection and painting the inside surface of the redundant fire water tank in preparation for refill. The licensee plans to have the tank refilled by the beginning of March 1993, and complete restoration of the redundant fire water system by the end of March 1993. The inspectors concluded that the licensee had taken appropriate compensatory actions in response to the -
unavailability of the redundant fire water system.
4.4 5 kV Cable Testing Program Change In January 1993, the licensee notified the resident inspectors of a change in their cable testing program for safety-related and non-safety-related 5 kV cables. Based on industry experience regarding the destructive nature of higher de voltage high potential (hi-pot) testing of installed 5 kV cables, the licensee had reduced the voltage level of installed 5 kV cable hi-pot testing from 25 kV de to 10 kV dc. The inspectors discussed this issue with the licensee and reviewed the cable testing program to assess its ability to detect cable degradatio }
.
,
.
History In response to a 1988 Augmented Inspection Team that assessed the effects of a faulted cable between emergency diesel generator No. 2 and 4 kV vital bus ID, GPUN committed to developing a formal cable testing program. GPUN replaced the failed cable and tested the new cable with a 35 kV de hi-pot test and a 5 kV ac power factor test. The test results were satisfactory. As part of the long-term corrective actions in response to the cable failure, the licensee stated that higher levels of de hi-pot testing would be utilized as recommended by industry standards to provide more accurate data on cable condition. The cable testing program was formally implemented in early 1991 prior to the 13R refueling outage. The resident inspectors assessed the licensee's implementation of the program and noted that it had been developed using IEEE 400-1980, "IEEE Guide for Making Direct Voltage Tests on Power Cable Systems in the Field." IEEE 400-1980 calls for an inservice de hi-pot test voltage of 25 kV for 5 kV cables disconnected from equipment. The standant also allowed the use of lower test voltages for cables connected to equipment to comply with limitations imposed by the connected equipment.
Program Reevaluation Although cable is normally factory tested with hi-pot testing to detect manufacturing defects,
,
the cable may incur subsequent damage due to shipment or installation. Cable performance may also be degraded by errors in cable termination or splicing. As such, testing is often performed after shipment and installation, but befom the cable is energized, to ensure that the installed cable will perform satisfactorily. Additional testing after time in service can also be performed to monitor cable performance and detect cable defects or damage caused by operation in service. Hi-pot testing can be performed with either an ac or de voltage source.
The de voltages required to produce an equivalent voltage stress are about three times higher than the required ac voltages. Historically, the de iti-pot test has been viewed as advantageous for field application due to the portability of the related test equipment. The de test equipment is significantly smaller than equipment required for ac hi-pc4t testing because
,
charging does not have to be continually applied to the de equipment. Since developing the formal Oyster Creek cable testing program, the licensee had been performing post-shipment and inservice de hi-pot testing at the test voltages for 5 kV cable prescribed in IEEE 400-1980, i.e., 35 kV de (post-shipment / installation test) and 25 kV dc (in service / maintenance test).
-
In late 1992, based on more recent industry information on the destructive nature of de hi-pot testing, Oyster Creek plant engineering reevaluated the need to perform de hi-pot testing at
'
the higher test voltage values. The industry information, compiled by several sources including the Electric Power Research Institute (EPRI), indicated that the higher voltage dc
,'
hi-pot testing subjects the cable to voltage stresses which are not representative of normal service. It was noted that de hi-pot testing did not appear to become problematic until the
!
1960s, when cable insulation changed from the paper-oil type to solid extruded dielectrics. It is believed that the cable stressing effected by higher voltage de hi-pot testing may even be
!
,..,,
.
predisposing cables with solid dielectric insulation for future failure, causing the cable to fail sometime, often within months of returning the cable to service. Based on this information and Oyster Creek plant-specific experience with other types of cable testing, including lower voltage de hi-pot testing, plant engineering concluded that higher voltage de hi-pot testing should be excluded from the Oyster Creek cable testing program.
The Oyster Creek cable testing program, with the lower voltage de hi-pot test for installed 5 kV cables, was implemented during the 14R outage. New cable was 2.5 kV de megger checked on the cable spool before pulling. After the new cable was pulled, another 2.5 kV de megger check was performed. The cable lugs and stress cones were then installed. After installation of the cable connection hardware, a de hi-pot test at 35 kV de was performed.
This is the installation test de voltage prescribed in IEEE 400-1980. Cables already in service were tested with the lower voltage (10 kV) de hi-pot testing. This testing identified the degradation of 3 recirculation pump motor-generator (MG) set supply cables that were subsequently replaced with new cable.
A review of the past cable testing data at Oyster Creek showed that most of the cables which have failed or been degraded were those exposed to water, such as the recirculation pump MG set cables. When a cable fails the inservice test, it is sent by the licensee to a testing laboratory for analysis. However, since the start of the Oyster Creek cable testing program, the laboratory analyses for those cables which were found to be failed or degraded have found the root cause to be a manufacturing defect, not water intrusion.
The cable testing program has recently been augmented with what is referred to as characterization testing. Characterization testing includes a combination of non-destructive cable tests intended to establish a performance baseline for each specific cable. Future testing results are then compared to the baseline to determine the performance trend of the cable.
The characterization testing includes low voltage impedance testing to check cable electrical parameters of resistance, inductance, and capacitance. Cable insulation resistance is measured by a 1,000 Vdc insulation resistance test. Megger testing results are also included in the baseline data, as well as the current and past results of de hi-pot testing performed on the Oyster Creek 5 kV cables. Characterization testing also includes a newer test known as time domain reflectometry (TDR). TDR is performed by applying an impulse signal to the cable and analyzing the signal which is reflected back from the cable. This test is intended to provide a cable characteristic signature which can be directly compared with past and future test data to identify performance trends. TDR can also supposedly be used to find the specific location of a cable fault. Characterization testing was performed on 14 of the 47 5 kV power cables at Oyster Creek during the 14R outage. The remainder of the cables will receive their characterization baselines during future outages. GPUN also noted that it was their intention to eventually place the cable characterization data on the plant equipment computer data base (GMS2).
,
'.
The cable testing program currently requires cables that are exposed to water are tested'every four years or second refueling outage. Other 5 kV cables are tested every six years or third refueling outage. The inspectors concluded that the 5 kV cable testing program at Oyster Creek was effective and that the removal of the higher voltage inservice de hi-pot testing requirement did not adversely effect the ability of the program to detect cable degradation.
The program also provided an effective means of trending 5 kV cable performance.
5.0 SECURITY (71707)
During routine tours, inspectors verified that access controls were in accordance with the Security Plan, security posts were properly manned, protected area gates were locked or guarded, and isolation zones were free of obstructions. Inspectors examined vital area access points and verified that they were properly locked or guarded and that access control was in accordance with the Security Plan.
6.0 SAFETY ASSESSMENT / QUALITY VERIFICATION (90700)
6.1 Review of Licensee Event Reports NRC inspectors reviewed the following LERs for appropriateness of reporting, timeliness, event description, cause identification, and completeness of information. In addition, the need for onsite review was assessed.
LER NO.
DESCRIPTION LER 92-13 local Leak Rate Test Results in Excess of Limits Due to Valve Degradation LER 92-14 Voluntary Report - Service Water Piping Failure May Have Affected Secondary Containment Integrity LER 92-15 Containment Isolation and Standby Gas Treatment System Initiation Caused by Blown Fuse Due to Personnel Error -
LER 93-01 Standby Gas Treatment System Initiation Caused by Lightning Arrestor Failure Special Report 93-001 Control Room HVAC System B Inoperable Greater Than 7 Days
-
...
l
LER 92-13 documented the failure of the local leak rate test (LLRT) for the outboard main steam isolation valve (MSIV) on the B steam line (NSO4B). At the time the licensee could l
not quantify the leakage. During the 14R refueling outage, NSO4B was rebuilt and successfully passed the LLRT. In the LER the licensee stated a supplemental LER would be issued to document the cause, corrective action, and safety significance of the LLRT failure.
By the end of this inspection period, the licensee had not submitted the supplemental LER.
l The licensee submitted LER 92-14 to document the potential for a hole found in the service water piping downstream of the reactor building closed cooling water (RBCCW) heat exchangers to provide a ground level release pathway. The location of the leak was on the 51 foot elevation of the reactor building. The location of the leak was at a low pressure point in the piping causing air to be drawn into the pipe. From the location of the leak. the service water pipe is routed to the discharge canal. After identifying the condition, the licensee installed a temporary patch to prevent air inleakage. The licensee examined other locations in the service water piping and replaced the effected elbow and several other elbows in similar configurations. This item was reviewed previously by the NRC and is documented in NRC inspection report 50-219/92-25. The licensee is evaluating the effects the hole in the service water had on secondary containment. The licensee indicated the results of this evaluation would be submitted in a supplemental report.
LER 92-15 documents a partial primary containment isolation and initiation of the standby gas treatment (SBGT) system. On December 31,1992, a contractor was working on a control panel in the control room. An energized lead was grounded during this work that caused the SBGT system to start and a partial isolation of primary plant containment. The grounded energized lead failed the power supply fuse for containment isolation and SBGT system. The reactor was shutdown and defueled at the time of the event. The failed fuse was replaced. The licensee reset the partial primary containment isolation and secured the SBGT system. Personnel error caused the grounding of the energized lead that caused the event. The actuation signal was not generated as the result of a change in primary or
secondary containment condition. No additional followup of this event is required.
i LER 93-01 documented the failure of a swiThyard lightning arrestor on an offsite power line on January 1,1993. Oyster Creek was shutdown and defueled during the event. The failure of the arrestor caused power perturbations in the plant electrical system. The power perturbation caused a high level trip signal to be generated on the reactor building ventilation radiation monitor. With the high level trip signal, the normal reactor building ventilation system tripped and the SBGT system started. Normal reactor building ventilation was restored within ten minutes of the initiating event. Repairs to the lightning arrestor were completed forty minutes following the failure. No other safety system actuations occurred.
No additional followup of this event is required.
i
v
,
7.
'. -
Technical specification (TS) 3.17.B(2) requires the licensee to submit a special report'within 14 days after one train of the control room heating, ventilation, and air conditioning (CR-HVAC) system is inoperable for greater than seven days. Special report 93-001 satisfied this requirement. On January 28,1993, the B train of CR HVAC did not pass the acceptance criteria for the flow and differential pressure test being conducted. tThe system was declared inoperable. A walkdown by the licensee identified a broken access hatch on the supply duct '
downstream of the B CR HVAC fan. The access hatch was repaired, the system retested, and the system declared operable on February 11,1993.
The inspector reviewed LERs 92-13, 92-14, 92-15, and 93-01, and Special Report 93-001.-
Each report accurately described the events as they occurred. TS requirements for one train of CR HVAC being inoperable were satisfied. All of the reports were is_ sued within the-.
.
required time. Review of the LER cover pages found that the licensee had not provided the:
component failure information required in section 13 of NRC Form 366. The information'.
relating the involved components were discussed in the text of the LERs. ' Except for the -
j MSIV failure cause, the causes of the events were adequately described in the LERs. The'
j inspectors discussed the missing information in section 13 of the LER cover sheet with the j
licensee. The licensee indicated they would evaluate actions to correct'the lack of. component
!
failure information on the cover sheets to LERs.
!
7.0 REVIEW OF PREVIOUSLY OPENED ITEMS (92702)
.
.
.
d (Update) Violation 50-219/91-39-01. This violation involved the performance of an
.!
equipment lineup with the parts of the system tagged out of service.. This was contrary to licensee procedural requirements for equipment lineups. Specifically, the containment -
spray / emergency service water (CS/ESW) system breakers for valves V-3-87 and V-3-88
!
were improperly verified while tagged out of service. Also, when the inspector questioned j
the licensee on what the correct position was, the licensee determined that the electrical ~
~
component lineup was incorrect.
In response to the violation, the licensee changed procedure 310, " Containment Spray System.
Operation," to reflect the correct position of the valve breakers. Operations management developed a training lecture that provided expectations related to procedural compliance with
,
attention to detail using this event as an example. The lecture was given by operations management to licensed operators during requalification training. ~ As part of the Process i
Reengineering Program (PREP), the licensee revised procedure 108,7 Equipment Control," to
,
refine the process controlling switching and tagging, and equipment alignment and i
verification. To ensure operations department personnel were aware of this event, the
!
'
violation and portions of the revision to procedure 108 that applied at the time, were made '
required reading for operations department personnel.
!
l
-
!
l
'!
'.
o
-
The inspector reviewed the procedure developed because of the revision to procedure 108.
The new procedure,108.9, Revision 1, " Equipment Alignment and Verification," provides the instructions for performing equipment alignments and ' verifications. - Responsibilities for performing initial alignments, followed by independent verifications have been provided.
This included detailed instructions on how lineups were to be conducted when restoring a-system to service following maintenance or modification.
Review of procedure 310, Revision 46, " Containment Spray System Operation," by the inspector found that the licensee had incorporated the correct position of the breakers for valves V-3-87 and V-3-88. The inspector also reviewed the training lecture developed by operations management for use during licensed operator requalification training. The licensee had adequately described the events associated with the errors made during the alignment of the CS/ESW system breakers. The inspector reviewed the list of operations department personnel that had completed the required reading of the violation and applicable ponions of procedure 108. A number oflicensed and non-licensed operators had not completed the-required reading by May 31,1992. The inspector discussed the completion of the required reading with the Manager, Operations Suppon at the end of this inspection period. The Manager, Operations Support was reviewing the cause for personnel not completing the required reading. This item will remain open pending the licensee review of the failure of personnel to complete the required reading in a timely manner and NRC review of the actions
'
to correct this condition.
(Undate) Unresolved Item 50-219/92-18-01. Reactor water level mismatch for cold reference leg instrumentation. This item was updated previously in inspection report 50-219/92-25.
The item remained open pending the licensee addressing the recommendations of General Electric Service Information Letter (GE SIL) No. 470, concerns with the installation of a thermocouple bracket on the B GEMAC sensing line, and the lack of insulation on the sensing lines to the condensing chambers.
During a drywell tour on January 30,1993, the inspector noted that the licensee had installed insulation on the sensing lines from the reactor vessel to the condensing chambers for GEMACs IA15A and IA15B. In addition, the inspector noted that the licensee had installed thermocouples to the sensing lines between the reactor vessel and condensing chambers, and on the condensing chambers for both GEMAC level instruments. The item remains open pending completion of NRC review of the licensee's actions in response to GE SIL No. 470 and the licensee's analysis of the effects the thermocouple bracket has on GEMAC IA15B
>
response.
(Closed) Violation 50-219/92-044)1. This violation was the result of the licensee's failure to update the electrical checkoff list _ for core spray system 1 as equipment and component power supplies changed. As a result, the position of several breakers required to place core spray system I in a standby condition were not verified to be in the proper position on May 2, 1991. Also, during the equipment verification, the component labels were not checked as required by station procedure 108, Revision 54, " Equipment Control."
- .
.
l
In response to the violation, the licensee changed procedure 308, " Emergency Core Cooling l
System Operation," to reflect the proper breaker descriptions in electrical checkoff list 308-2.
.
Also, review of other electrical lineups by the licensee identified three more procedures that i
required revisions to reflect the proper breaker descriptions required to support normal equipment lineups. The licensee stated that existing controls in procedure 124, " Plant Modification Control," were adequate to prevent futum modifications from causing similar problems. In each case identified by the licensee, the improper equipment description contained in the procedure were the result of modifications made before the controls implemented by procedure 124 were in place. To aid the operators with identification of the electrical components, the licensee also stated that an effort was underway to upgrade the
-
breaker schedules to an operator aid status. Operator aids assist personnel in operating the plant and are controlled procedurally.
The inspector reviewed procedure 124, Revision 15, " Plant Configuration Change Control,"
,
'
(the title of the procedure was changed from the previous version). Detailed instructions have been provided for procedure reviews and assignment of responsibility for upgrading procedures because of modifications. Instructions for updating plant drawings were also included, with specific responsibilities assigned to update the drawings. Equipment labeling requirements were included in the requirements of procedure 124. If permanent plant labels were not available, instructions were given for the use of temporary labels. Procedure 124 clearly identifies the process to be used to ensure that modifications have been appropriately incorporated into the procedures and drawings, and the components were properly labeled before turnover of the modified system to the plant.
,
The inspector discussed the upgrading of the breaker schedule with operations department
,
management. The breaker schedule upgrade to an operator aid was still ongoing at the end of the inspection period. Delay in implementation of the upgrade resulted from activities in support of the 14R refueling outage. Operations management indicated that with the completion of the 14R refueling outage, upgrading of the breaker schedule to an operator aid would be implemented.
Adequate controls were in place in procedure 124 to ensure that future modifications were
'
incorporated in plant procedures and drawings, and that components were labeled appropriately. Instructions in procedure 108 at the time the violation occurred, provided the licensee the opportunity to identify this event when the system was restored during the 13R refueling outage. Based on the inspector's review, the licensee's actions in correcting the description of the component to suppon placing the core spray system in the standby condition were adequate. For these reasons, violation 50-219/92-04-01 is closed.
t During the review of procedure 108.9, Revision 1, " Equipment Alignment and Verification,"
for followup of violation 50-219/91-39-01 (discussed previously in this report), the inspector noted there were no instructions on how to verify that the components being aligned were the correct components. Before the revision of procedure 108 made in response to the Process Reengineering Program (PREP) initiative, a person verifying the position of a component was
,
s~o
required by the existing 108 procedure to check the label to ensure that it was correct. Had the operators performed the label check properly in May 1991, the errors noted in violation 50-219/92-04-01 would probably have been identified by the licensee. With the removal of the procedure 108 instruction to check component labeling during the verification of equipment alignment, it was not apparent to the inspectors how this guidance was being transmitted to personnel performing equipment control functions. The licensee indicated that this guidance had been removed during the PREP update of procedure 108 because it was felt that appropriate guidance related to plant label verification was being implemented through other administrative means. By the end of the inspection period, the inspectors had not completed the review of the licensee's other administrative controls to ensure proper component identification during equipment alignment. This is an unresolved item (URI 50-219/93-03-02) pending further review of this issue.
8.0 EXIT MEETINGS (40500, 71707)
8.1 Preliminary Inspection Findings A verbal summary of preliminary findings was provided to the senior licensee management -
on February 25,1993. During the inspection, licensee management was periodically notified verbally of the preliminary findings by the resident inspectors. No written inspection material was provided to the licensee during the inspection. No proprietary information is included in this report. The inspection consisted of normal, backshift and deep backshift inspection; 115 of the direct inspection hours were performed during backshift periods, and.
87 of the hours were deep backshift hours.
8.2 Attendance at Management Meetings Conducted by Other NRC Inspectors The resident inspectors attended exit meetings for other inspections conducted as follows:
February 8,1993 Report No. 50-219/93-80, Degraded Shutdown Cooling Event At this meeting the lead inspector discussed preliminary findings with senior GPUN management.