ML16154A809
| ML16154A809 | |
| Person / Time | |
|---|---|
| Site: | Oconee  |
| Issue date: | 07/12/1995 |
| From: | Steven Rudisail, Shymlock M NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML16154A810 | List: |
| References | |
| 50-269-95-10, 50-270-95-10, 50-287-95-10, NUDOCS 9507240382 | |
| Download: ML16154A809 (16) | |
See also: IR 05000269/1995010
Text
pft REG&
UNITED STATES
o~
NUCLEAR REGULATORY COMMISSION
REGION II
0
101 MARIETTA STREET, N.W., SUITE 2900
ATLANTA, GEORGIA 30323-0199
Report Nos.:
50-269/95-10, 50-270/95-10 and 50-287/95-10
Licensee:
Duke Power Company
422 South Church Street
Charlotte, NC 28242
Docket Nos.:
50-269, 50-270, and 50-287
License Nos.:
and DPR-55
Facility Name:
Oconee Nuclear Station Units 1, 2 and 3
Inspection Conducted:
June 5 -
16, 1995
Inspector: J
7 7 95
S.' Rudisail
Date Signed
Accompanying Personnel: V. Beaston, NRR
Approved by:
_
_
_ _
_
7-__
_
_
M. Shymlock, C6Aef
Date Signed
Plant System Section
Engineering Branch
Division of Reactor Safety
SUMMARY
Scope:
This routine, announced inspection was conducted in the areas of electrical
design to review the licensee action in response to Electrical Distribution
System Functional Inspection (EDSFI) findings and other EDSFI issues
identified in NRC Inspection Report 50-269, 270, 287/93-02. These items were
being resolved as part of the licensee's Power Upgrade Project (PUP).
Results:
In the areas inspected, violations or deviations were not identified.
The inspector reviewed the licensee response and corrective actions for
violation 94-26-01, Failure to Perform Procedure as Corrective Action. This
violation was closed.
Deviation 93-02-02 was also closed. The licensee had completed the remaining
corrective actions for the deviation.
Enclosure
9507240382 950714
ADOCK 05000269
0
2
The inspectors reviewed various PUP items completed in response to EDSFI
findings. These findings were identified as Inspector Follow-up Item (IFI)
93-02-03.
Overall, the calculations were of good quality with no problems identified
during the calculation reviews. The licensee corrective actions for these
items were technically sound and through.
REPORT DETAILS
1.
Persons Contacted
Licensee Employees
L. Azzarello, Mechanical Systems Engineering
- K. Burchfield, Regulatory Compliance
P. Colaianni, Nuclear General Office, License Renewal
D. Coyle, Mechanical Systems Engineering
- J. Davis, Engineering
T. Grant, Electrical Systems Engineering
- J. Hampton, Site Vice-president
T. Ledford, Electrical Systems Engineering
- C. Little, Electrical Systems Engineering
D. Patterson, Regulatory Compliance
- B. Peele, Plant Manager
J. Stevens, Electrical Systems Engineering
Other licensee employee contacted during this inspection included
craftsmen, engineers, technicians, and administrative personnel.
NRC Employees:
P. Harmon, Senior Resident Inspector
- L. Keller, Resident Inspector
K. Poertner, Resident Inspector
- Attended exit interview
Acronyms and abbreviations used throughout this report are identified in
the last paragraph.
2.
Background
During January 25 through March 5, 1993, the NRC conducted the
Electrical System Distribution Functional Inspection (EDSFI).
The
purpose of this inspection was to assess the capability of the Oconee
Electrical Distribution System to perform it functions during normal
operations and accident conditions. The conclusion of the EDSFI team
was that the electrical distribution system would perform its intended
function pending further analysis and testing by the licensee. During
the EDSFI inspection a violation and several deviations were identified.
Additionally, findings identified during the EDSFI were collectively
identified as IFI 93-02-03.
3.
Review of Corrective Action for Violation and Deviations (IP 92702)
The inspector reviewed a violation identified during a previous EDSFI
follow-up inspection and three deviations identified at Oconee during
the EDSFI. These items are discussed in the following paragraphs.
2
3.1
(Closed) Violation 50-269, 270, 287/94-26-01, Failure to Perform
Procedure as Corrective Action was closed. This violation was
identified after it was determined that Keowee procedure OP/A/2000/45,
Lock Verification had not been performed. This procedure was issued as
part of the corrective action for deviation 93-02-02. During a review
of the corrective actions for this deviation it was determined that this
procedure had not been performed until the results were requested for
review by the inspector. This procedure was required to be performed
semi-annually. During this inspection the corrective actions for this
violation were reviewed. The inspector verified that this procedure had
been included in the procedure tracking system and reviewed the results
of the most recently completed procedure. The results of this review
were adequate. This violation was closed.
3.2
(Closed) Deviation 93-02-02, Item 1, Deviation from FSAR Color Coding
Requirements for Cables was one of three examples of this deviation
identified during the EDSFI.
The first example was identified as power
cables 1XS2 and 1XS48 not being correctly color coded. These cables run
from the motor control centers 1XS1 and 1XS2 to transformer CT4 cooling
fans. Final Safety Analysis Report (FSAR) section 8.3.1.3 states that
motor control center (MCC) 1XS1 power and control system cable are color
coded gray and that MCC 1XS2 cables are color coded yellow. Contrary to
this requirement, both cables were color coded black.
The licensee responded to this deviation and agreed that it was a
deviation from the requirements of the FSAR and an oversight during
initial installation.
The licensee considered the color coding was
intended to ensure separation during installation. The separation
requirements are for mutually redundant cables to be run in separate
trays. The cables are routed in different cable trays until they reach
the block house of the CT4 transformer. The cable is interlocked armor
type cable. The licensee determined the impact of failure of both
cables and concluded that should they fail the transformer would
continue to operate as designed with only a small impact on transformer
life. Alarms are available to indicate loss of power to the fans and
high oil and winding temperature in the transformer. Operator action
would be required upon receipt of alarms. The licensee concluded that
the impact on safety due to the cable being color coded black was
insignificant. The inspector agreed with this conclusion. The licensee
proposed additional corrective action to perform an inspection of all
plant, switchyard, and Keowee cables fed from safety related busses not
previously identified as having mutually redundant safety related
functions.
The inspector reviewed the results of the cable separation study. This
was identified as item 13 of the licensee's PUP. This study identified
several cables of mutually redundant systems at Keowee which were routed
in the same cable tray. The licensee had determined that these cables
were routed such that the distance between the cables was sufficient to
preclude the need for re-routing the cables for the purpose of cable
tray separation. The inspector performed a walkdown of the affected
cables to determine if the licensee evaluation of their safety
II
3
significance was appropriate. These cables were interlocked armor type
cables and were adequately separated. The inspector agreed that these
cables were sufficiently separated to preclude the mutual failure of
both cables. This deviation item was closed.
3.3
(Closed) Deviation 93-02-02, Item 2, identified mutually redundant
cables for Unit 2 emergency core cooling isolation valves 2LP-19 and
2LP-20 being routed in the same cable tray. This was contrary to the
requirements of FSAR section 8.3.1.4.6.2. The licensee agreed that this
was a deviation.
The corrective action for this deviation was to perform a modification
to separate the cables. Minor modification OE-5990 was prepared to
separate the cables. The inspector reviewed this modification and the
scope of work accomplished during its performance. This modification
accomplished the required corrective action. Inspection of Unit 1 and
Unit 3 cable for valves LP-19 and LP-20 were performed and the
separation problem did not exist. The licensee proposed additional
corrective action to perform an inspection of all plant, switchyard, and
Keowee cables fed from safety related busses not previously identified
as having mutually redundant safety related functions. This study was
discussed in item 1 of the deviation.
Item 2 of the deviation was
closed. This closes deviation 93-02-02.
Item 3 of the deviation was
closed in inspection report 50-269,270,287/94-26.
4.
Review of Inspector Follow-up Items (IP
92903, TI 2515/111)
The inspector reviewed the items identified at Oconee as Findings.
IFI 93-02-03, EDSFI Findings consisted of six findings with several
issues identified in each finding. This finding was closed in NRC
inspection report 50-269, 270, 287/94-26 to document the findings which
were closed in that report.
IFI 94-26-02 was opened to identify the
EDSFI finding items which remain opened.
4.1.
Finding 2 was an identification of analyses, studies and calculations
which were not complete or had not been performed. There were ten
elements identified in this finding. Item 2.j was closed in IR 50-269,
270, 287/94-26. The other nine items identified in finding 2 remained
open. Some of these items were reviewed during this inspection and are
discussed below:
4.1.1 Finding 2.a identified that calculation OSC-2059, Oconee Unit 1 Voltage
and Load Study may not have bound the worst case. During the EDSFI
review of the Oconee short circuit and voltage studies, the team
identified deficiencies in calculation OSC-2059. The Unit 2 and Unit 3
calculation were being revised during the EDSFI and were not available
for review. The team noted during the review of OSC-2059 the following.
(1) The momentary current on the 4.16 kV supply breakers under the
worst case scenario was 83 kiloamperes (kA) exceeding the breaker
rating of 80 kA. The team noted that some of the breakers had
been successfully tested at 83 kA by the manufacturer.
4
(2) There was only one input file for both short circuit and voltage
drop calculations which may not bound the worst case minimum and
maximum voltages.
(3) The calculation did not adequately model the transformer tap
position or the pre-fault bus voltage.
(4) When determining the total loads on transformers the study used 75
degree cable temperature, and constant motor efficiency and power
factor at full load.
(5) Only one 230 kV system impedance was modeled in the study.
However, the,EDSFI team previously concluded that even though the
calculation results may not be bounding that they were acceptable and
no operational problems were identified.
Item 22 was identified to address this EDSFI finding. The inspector
reviewed the following calculations revised to address this finding.
Calculation OSC-2059 Rev. 3, calculation OSC-2060 Rev. 3, Oconee Unit 2
Voltage and Load Study, and OSC-2061 Rev. 2, Oconee Unit 3 Voltage and
Load Study. The calculations were reviewed to ensure that adequate
voltage would be provided to the units auxiliary power system and that
the fault duty of the switchgear and protective devices were adequately
sized.
The concerns identified in the EDSFI were also reviewed to
ensure that they were addressed by the revisions to the calculations.
The methodology used to perform the calculations was reviewed to ensure
each item was addressed. The inspector determined from review of the
calculation that only one input file is needed as this file contains
manufacturers and test data and actual cable lengths. The revision
incorporated the actual transformer impedance for the tap setting and
modeled pre-fault bus voltages. The calculation uses actual motor KVA
and power factors. The worst case 230 kV system impedance was used to
calculate fault currents.
The results of the revised calculation
lowered the magnitude of the calculated faults thus addressing the
circuit breaker rating concern. This item was closed.
4.1.2 Item 2.c identified that a transient voltage study for the 4kV safety
loads when supplied by the Lee or Central substation was not available.
The licensee responded that a calculation to demonstrate dynamic
modeling of the Lee and Central power sources would be completed. In
EDSFI report section 2.6.2 a concern was expressed that during starting
of a unit's LOCA loads and two unit's shutdown loads, the transient
voltage dip could exceed 20 percent. The licensee agreed to prepare a
transient voltage study for the 4 kV safety load groups when they are
supplied by the Lee gas turbine or from Central substation. During this
inspection the revision to calculation OSC-3585, Voltage Study for the
Oconee Auxiliary System When Fed from the 100 kV System via CT5 Rev. 1
was reviewed. This revision computed the transient reactance of the 100
kV system (while being supplied from the central substation) to the
standby bus. The transient reactance when the Oconee auxiliary system
is supplied by the 100 kv system is 1.5 times less than when the Oconee
5
auxiliaries are supplied from Keowee. This results in a smaller
transient dip when supplied by the 100 kV system. The inspector
concluded that the calculation OSC-3585 demonstrated the ability of the
100 kV system (while being supplied from the central substation) to
supply adequate voltage to the Oconee auxiliaries and recover from the
postulated transient. This part of finding 2.c will be closed. The
review of the voltage study for Lee gas turbine supplying the Oconee
auxiliaries remains open.
4.1.3 Item 2.d identified that a study to review control cable length and the
size of fuses that protect them had not been conducted.
This item was
addressed by the licensee as PUP item 22.
The licensee completed three
calculations in response to this finding. The inspector reviewed
calculation OSC-5930 Rev. 0, Unit 1 QA1 Motor Starter Circuit and Fuse
Adequacy Calculation, calculation OSC-6143 Rev. 0, Unit 2 MCC Contactor
Voltage Adequacy and Fuse Adequacy Verification and OSC-6144 Rev. 0,
Unit 3 MCC Contactor Voltage Adequacy and Fuse Adequacy Verification.
The purpose of the calculations was to demonstrate that each units motor
control circuits would have adequate voltage during a LOCA degraded grid
scenario and that during an overvoltage condition the fuses would not
blow. The results of the calculations demonstrate that adequate voltage
is available for each of the motor starting circuits and fuses are
adequately sized for the circuit application. The inspector did not
identify any additional concerns during the calculation reviews. This
item was closed.
4.1.4 Item 2.e of the EDSFI identified that the Keowee 600V Auxiliary Power
System Voltage Analysis did not evaluate maximum and minimum expected
voltages. PUP item 4.a was identified to address this EDSFI item. The
licensee completed calculation KC-UNIT-1-2-0095 Rev. 0 to address
voltage adequacy in the auxiliary power system. The calculation
documented and analyzed the adequacy of Keowee electrical distribution
system voltages when supplied by either the 230 kV switchyard or the
Keowee generators. The results of the calculation demonstrate that the
Keowee Station auxiliary electrical system will be within acceptable
limits to allow performance of their design functions under all system
configurations and at minimum and maximum predicted switchyard voltages.
The inspectors reviewed the calculation and found the results
demonstrated the adequacy of the voltage supplied to the Keowee
electrical auxiliaries. This item was closed.
4.1.5 Item 2.f, was a concern that no analysis existed to support that the
Keowee auxiliaries would not be damaged by overvoltage or overfrequency
when supplied by one Keowee unit.
Item 2.h identified that an analysis
to support the assumption that Oconee safety loads could properly
perform during an overfrequency transient lasting 40-50 seconds. The
licensee plans to complete an analysis to support this assumption.
During a previous inspection documented in NRC inspection report 50-269,
270, 287/94-26, the inspector reviewed the actions completed for these
finding. One calculation had been completed. The calculation to
demonstrate the effect of overvoltages and overfrequencies on the Keowee
- II6
auxiliaries was complete. The calculation OSC-5701 Rev. 0, Oconee
Keowee Overhead Path Analysis identified weaknesses that limit the
percent power the Keowee units can generate to the grid and the
alignment of the Keowee units. Additionally, calculation OSC-5952 Rev.
0, Oconee-Keowee Underground Path Analysis Using Cyme was reviewed
during this inspection and discussed in paragraph 4.2.2 of this report.
Corrective actions identified by these calculations are being addressed
in modification NSM 52966 currently being reviewed by the NRC. This
item will remain open pending the completion of this review and
implementation of the modification by the licensee.
4.1.6 Item 2.i identified that calculations for the Safe Shutdown Facility
were not complete. PUP items 16 and 17 was identified to address this
finding. The inspector reviewed the completed calculations for this
finding. The licensee had completed calculation OSC-5093 Rev. 0, SSF
Electrical Distribution System Load Flow, Voltage Adequacy, and Fault
Study. The inspector reviewed the results of this calculation.
The
results of the calculation for the SSF Diesel Static Analysis which was
attachment 5 of the calculation indicated a potential overload of the
feeder breaker for load center MCC XSF. The feeder breaker had a long
time setting of 660 amperes. The calculation demonstrated that this
breaker could be subjected to 863 amperes. This loading could result in
a breaker trip after 5 minutes. PIP 4-095-0400 was initiated based on
this calculation result. Immediate corrective actions were taken to
ensure SSF operability. Four duct heater breakers (for room heating)
were tagged out which would reduce loading sufficiently to correct the
problem. Long term corrective action included a minor modification to
replace the breaker trip element to a size sufficient to support the
calculated loading.
Other than this problem the results of the
calculation demonstrated the adequacy of the SSF electrical system
ratings. This finding was closed.
4.2
Finding 5 of the EDSFI was the identification that Keowee engineering
analyses were not sufficiently comprehensive and specific values had not
been established which would bound the design criteria. Four items were
identified in this finding. Finding 5.a identified that all credible
failure modes for the governor control system and voltage regulator had
not been considered. This item was not reviewed during this inspection
and remains open. The additional items for finding 5 are addressed in
the following paragraphs.
4.2.1 The inspectors reviewed PUP item 5 which addressed EDSFI finding 5.b.
Finding 5.b of the Oconee EDSFI report stated "The basis for bypassing
Keowee trip functions during emergency start of the unit was not fully
analyzed or documented."
Section 3.2.4.2 of the Oconee EDSFI inspection report, "Protective
Feature Issues," stated that "the control logic bypasses all of the
Keowee normal automatic electrical and mechanical protective trips on an
emergency start. The bypassed trips included generator and turbine
bearing overtemperatures, volts/hertz, overspeed, governor oil pressure,
7
generator field ground, and maximum excitation."
The EDSFI team had
concluded that the basis for bypassing Keowee trip functions during
emergency start of the units was not fully analyzed. As a result of the
concern, the licensee initiated PIP 0-093-0081 to request an analysis of
Keowee lockout relay trip signals.
The inspectors reviewed calculation/analysis KC-0107, Analysis of Keowee
Lockout Relay Trip Signals, dated May 22, 1995. This analysis
documented the bases which would be used to determine lockout trip
signals at Keowee.
The Keowee hydro units are equipped with two lockout relays (LOR),
designated as the Emergency LOR (86E) and the Normal LOR (86N). In
addition each unit has a shutdown relay (99SX) which causes the wicket
gates to close when deenergized and an Alarm LOR (30X) which will
prevent non-emergency automatic starts of the Keowee units. The 86E LOR
is designed to immediately separate a Keowee unit by tripping the
generator output breakers without unloading the unit. The 86E LOR is
not bypassed during an emergency start of the Keowee units. The 86N LOR
is designed to unload the generator by activating the 99SX relay first
and allowing the wicket gates to reach their speed-no-load positions
prior to tripping the generator output breakers. The 86N LOR is bypassed
during an emergency start of the Keowee units.
The Keowee lockout relay analysis concluded that the existing design was
adequate to ensure proper unit response to an emergency, however, it
also recognized that system enhancements could be achieved. One of the
recommendations made in the analysis was to move the volts/hertz and
generator overexcitation trips from the 86N LOR to the 86E LOR. This
recommendation will be addressed by Keowee Voltage Regulator
modification NSM ON-52965. The overspeed trip will be removed when NSM
ON-52966 is implemented and replaced by a time-delay trip input to the
86E LOR. The analysis also recommended that the mechanical systems
group review the bearing overtemperature trips setpoints to see if they
could be set just below the bearings' material melting points, thereby
allowing operators more time to take action.
The inspectors found the licensee's basis for bypassing Keowee trip
functions during emergency starts acceptable, and concluded that the
basis had been adequately analyzed and documented. The licensee was
tracking the recommendations of the analysis with PIP 4-095-0577. This
item was closed.
4.2.2 The inspectors reviewed PUP item 4.h which addressed EDSFI items 5.c and
5.d.
Findings 5.c of the Oconee EDSFI report stated "The effect of frequency
of the electrical power supplied by Keowee to the ECCS pump motors had
not been fully evaluated."
8
Finding 5.d of the Oconee EDSFI report stated "Acceptable voltage and
frequency limitations for the Keowee electrical auxiliaries and the
emergency power system should be defined. Additionally, acceptable
recovery times from voltage and frequency excursions should also be
identified."
Section 3.4.1 of the Oconee EDSFI inspection report, "Review of Voltage
and Frequency Analyses and Tests," identified concerns with the
licensee's analysis of the Keowee hydro units ability to supply adequate
power to Oconee auxiliaries during voltage and frequency transients.
The inspectors reviewed calculation OSC-5952, Oconee-Keowee Underground
Path Analysis Using Cyme, dated May 25, 1995. The purpose of this
calculation was to determine the adequacy of the Keowee units when used
to provide emergency power to the Oconee auxiliaries via the underground
path. The calculation contained voltage and frequency plots for five
different scenarios. Also provided in the calculation were the
transient start data for the ECCS pump motors. The five scenarios
modeled were:
(1) A Keowee unit running at steady state, no load, accepts Oconee
Unit 1 LOCA loads and Oconee Units 2 & 3 LOOP loads
simultaneously;
(2) A Keowee unit running at steady state, no load, accepts Oconee
Unit 1 LOCA loads followed 20 seconds later by Oconee Units 2 & 3
LOOP loads;
(3) A Keowee unit running at steady state, no load, accepts Oconee
Unit 1 LOCA loads along with one condensate booster pump;
(4) A Keowee unit supplying 75 MW to the grid receives an emergency
start demand, the Keowee unit accepts Oconee Unit 1 LOCA loads
plus a condensate booster pump once the governor reduces system
frequency to 110% of its rated value, Oconee Units 2 & 3 LOOP
loads are accepted by the Keowee unit 9 seconds later;
(5) A Keowee unit supplying 75 MW to the grid receives an emergency
start demand, the Keowee unit accepts Oconee Unit 1 LOCA loads,
Oconee Unit 2 LOOP loads, plus a condensate booster pump once the
governor reduces system frequency to 110% of its rated value,
Oconee Units 3 LOOP loads are accepted by the Keowee unit 9
seconds later.
The licensee compared the results of the transient analysis using the
following criteria to determine the operability of the Oconee auxiliary
system:
(1) At the 4 kV level, voltages on switchgear ITD (Bus 115) were
compared against the settings of the 27S relays to determine the
effects of transient voltages on the relays operation;
- I
9
(2) At the 600 V and lower level, voltages at MCC XS1, XS2, and XS3
were compared with pick-up and drop-out voltages of contactors to
determine the effects of transient voltages on contactor
operations. The licensee based the pick-up and drop-out voltages
for contactors on test values provided by GTE Sylvania;
(3) Motor starting currents were compared against the settings of
their associated overcurrent relays to determine the effects of
transient voltage on motor operation.
The inspectors were concerned about how the initial voltage level of the
Keowee unit would affect the ability of the ECCS pump motors to start.
At the request of the inspectors, the licensee performed an analysis of
scenario one above for a case were the initial generator output voltage
of the Keowee unit was only 13.2 kV, instead of the 13.8 kV used in the
calculation. This analysis was requested to determine how sensitive the
computer simulations might be to small inaccuracies in the input data,
and to determine how much margin might exist between the nominal
operating voltage of the Keowee units and the minimum values needed to
ensure proper operation of the Oconee safety loads.
To address the inspectors concern about low generator output voltage,
the licensee plotted the starting currents for the low pressure service
water pump and the reactor building spray pump motors at Keowee output
voltages of 13.8 kV, 13.5 kV, and 13.2 kV along with the associated
overcurrent protection relay curves. These motors had been previously
identified by the licensee as the most limiting ECCS motors at Oconee
when started under degraded voltage or frequency conditions. From these
plots the inspectors were able to see how the motor start currents and
start times where changing with Keowee generator output voltage and
frequency, and gain some insight into what the computer simulation was
predicting and the actual trip settings of the overcurrent protection
relays. Based on these plots and the conservatism of the assumptions
used by the licensee in the analysis, the inspectors had no concerns
that the ECCS motors would not start.
The transient simulations conducted by the licensee for calculation OSC
5952 showed that underfrequency effects were not a problem when the
Keowee generator output voltage was within its currently specified
operating band of 13.5 kV to 14.1 kV prior to loading. The transient
analysis also showed that overfrequency effects were not a problem once
the licensee completes a modification of the Keowee units which will
ensure that the Oconee electrical distribution systems are not connected
to a Keowee generator output until after the system frequency has
returned to 110% of its rated value following a Keowee unit load
rejection. The.adequacy of this proposed modification of the Keowee
units is currently under review by the NRC's Office of Nuclear Reactor
Regulation.
Based on a review of calculation OSC-5952, which documents the adequacy
of the Keowee units to supply power to Oconee auxiliaries via the
underground path, prior NRC review of the adequacy of the Keowee units
- I
10
to supply the Oconee auxiliaries via the overhead path as analyzed in
calculation OSC-5701, Oconee Keowee Overhead Path Analysis and discussed
in NRC IR 50-269, 270, 287/94-26, and the licensee's proposed
modification to eliminate the Keowee overspeed/overfrequency concerns
currently under review by the Office of Nuclear Reactor Regulation, the
inspectors concluded that the issues identified as findings 5.c and 5.d
of Oconee Inspection Report 93-02 have been adequately addressed. These
items will remain open pending completion of a safety evaluation by the
Office of Nuclear Reactor Regulation documenting the adequacy of the
licensee's corrective actions for the Keowee hydro units
overspeed/overfrequency concerns.
5.
Review of Power Upgrade Project Items not Identified as EDSFI Findings
(IP 92903)
5.1
The inspectors reviewed PUP item 4.f, Keowee Voltage Regulator
Setpoints.
While performing an emergency start test of Keowee Unit 1 on
September 20, 1993, it was noted that the generator output voltage
leveled-off at 13.3 kV instead of its rated output voltage of 13.8 kV.
The low generator output voltage was later determined by the licensee to
have been due to the improper setting of the voltage regulator control
signal (See Problem Investigation Process (PIP) 4-093-0793). The root
cause of the problem was determined to be the lack of a documented
voltage setpoint. As a result, the licensee decided to formally
document the voltage regulator setpoints and incorporate the acceptable
generator output voltage operating band into the Keowee emergency start
test procedure.
The inspectors reviewed calculation/analysis KC-Unit 1 & 2-2023 dated
June 4, 1995. This calculation/analysis analyzed the operation of the
Keowee voltage regulator, and documented the bases for the Keowee
voltage regulator settings. The voltage regulator settings should
maintain the Keowee generator output voltage between 13.5 kV and 14.1
kV. This voltage range was based on dynamic computer simulations
performed by the licensee which showed the band to be adequate to ensure
operation of Oconee safety loads.
The licensee's analysis identified that the order in which certain
modules of the voltage regulator were calibrated was critical in
ensuring proper generator voltage during emergency operations. A new
calibration procedure for the voltage regulator was being written which
would specify the order. The analysis also recommended that the current
Keowee main step-up transformer tap setting be changed to tap three.
This change would allow the Keowee generators to operate in a voltage
band of 13.8 kV +/- 5 percent. Changing the transformer tap setting would
also increase the safety margin that currently exists between the Keowee
generator operating voltage and the minimum voltage needed to ensure
operation of Oconee safety loads.
Implementing this recommendation
however would require further analysis and coordination of the
electrical system as a whole.
11
The inspectors concluded that the licensee's analysis adequately
documented the bases for the Keowee voltage regulator settings. This
item was closed.
5.2
The inspectors reviewed PUP item 19 which was the Penetration
Overcurrent Protection Calculation.
Section 6.1.2 of the Oconee EDSFI inspection report; "Protection,
Coordination and Containment Electrical Penetration Protection;" stated
"the licensee is not committed to and does not comply with IEEE-317
requirements for penetration back-up protection. In addition, formal
calculations were not available to demonstrate the adequacy of primary
protection."
The inspectors review calculation OSC-4151, Revision 1; "A Review of
Penetration Overcurrent Protection;" dated May 16, 1995. The purpose of
the calculation was to demonstrate the adequacy of primary protection
devices for containment electrical penetrations. The calculation
analyzed the maximum fault currents available at each type of
penetration installed except for penetrations associated with low energy
thermocouple and instrumentation circuits. The fault current levels
were then compared to the field cable insulation thermal limits,
conductor fusing limits, and cable and penetration time-current
protection device settings associated with each type of penetration.
The results of the licensee's calculation showed that electrical
penetrations and their associated cables currently installed at Oconee
are adequately protected by primary overload and fault protection
devices. This protection should be adequate to ensure electrical
circuits passing through the reactor building will not degrade
containment integrity before, during or after a design basis event.
Section "J" of the analysis made recommendations to enhance penetration
overcurrent protection. These recommendations included: (1) adding a
note to design documents which clearly identify type D-12 penetration
load limits; (2) replacing 100 amp circuit breakers associated with type
C2 penetrations with 50 or 60 amp circuit breakers; and (3)
providing
operators with explicit response guidelines to manually shutdown a
reactor coolant pump upon detection of a sustained overcurrent. These
recommendations were being tracked by the licensee with PIP 0-095-0606.
Based on a review of calculation OSC-4151, Revision 1; A Review of
Penetration Overcurrent Protection; the inspectors concluded that the
license's primary protection was adequate. The licensee does not comply
with IEEE-317 requirements for penetration back-up protection, however
they are not committed to this requirement. This item was closed.
5.3
The inspectors reviewed PUP item 20 which identified a revision to the
degraded grid voltage relay setpoints and a change to the degraded grid
voltage detection configuration.
12
In Section 2.3.2 of the Oconee EDSFI inspection report, "Degraded Grid
Protection System," team members noted the following concerns with the
system design:
(1) All three single voltage measurements were monitoring the same "Z"
phase of the 230 kV bus;
(2) The three-relay scheme which only monitored the "Z" phase of the
bus was insensitive to phase voltage unbalance;
(3) The licensee did not have any operating procedure or surveillance
test to monitor the phase voltage unbalance.
The inspectors reviewed calculation OSC-5579, "Design Inputs and 10 CFR
50.49 Evaluation for NSM-ON-52950." Modification NSM-ON-52950 installs
three new QA-1 Capacitor Coupled Voltage Transformers (CCVTs) in the
Oconee 230 kV switchyard, and provides a voltage input to the degraded
grid voltage protection logic. Additionally, the modification adds
three new QA-1 undervoltage relays to the protection logic, and replaces
three Yellow Bus and three Red Bus CCVTs (a total of six CCVTs).
The new CCVTs and undervoltage relays will add a second set of 2-out-of
3 protection logic to the Oconee degraded grid protection system. The
three new CCVTs will monitor different phases of the Yellow Bus and add
diversity to the Oconee degraded grid protection system. Because the
new CCVTs will be connected to three different phases of the Yellow Bus,
modification NSM-ON-5579 addresses the phase voltage unbalance concerns
noted during the EDSFI inspection.
The inspectors had no concerns as a result of their review of
calculation OSC-52950. This item was closed.
5.4
The inspectors reviewed PUP item 92 which analyzed the voltage adequacy
of 208/120 VAC Power Panel Voltage.
The inspectors reviewed calculation OSC-6135, "208/120 VAC Power Panel
Voltage Accuracy," dated June 1, 1995. This calculation performed a
evaluation of each 208/120 VAC station auxiliary panel circuit to
determine if each end device required to operate during a LOCA would
operate at the steady state LOCA reduced voltage conditions.
The calculation used the panel board voltages calculated in OSC-2059,
OSC-2060, and OSC-2061; the voltage and load studies for Oconee Units 1,
2 & 3 respectively. From these steady state panel board voltages, end
device terminal voltages were calculated based on cable impedances and
the expected end devices' load currents. Results of the calculation
showed that in general the calculated end device voltage levels were
greater than or equal to the minimum operating voltages published by the
manufacturers. Several end device voltages however were slightly less
than the manufactures' published minimum operating voltages. In these
instances, the licensee relied upon other vendor documentation or test
data to conclude that the reduced voltages were acceptable.
II13
Due to the volume and details of the calculation, the inspectors only
reviewed a portion of the OSC-6135, to verify that all loads fed from
the 208/120 VAC station auxiliary panels would have adequate voltage
available to perform under a worst case voltage condition. This
included instances were an end device would be supplied with a terminal
voltage less than the published manufacturer's minimum operating
voltage. The inspectors noted during their review that there was
conservatism included in both calculation OSC-6135 and the voltage and
load study calculations used to determine the steady state panel board
voltages.
Based on a review of calculations OSC-6135 and the review of
calculations discussed in paragraph 4.1.1, the inspectors concluded
there was adequate assurance that loads fed from the 208/120 VAC panel
boards would be supplied with the minimum voltage needed to perform
during a design basis accident. The inspectors did not however review
or consider transient voltage conditions which might occur during a
design basis accident. This item was closed.
5.5
The inspectors reviewed PUP item 113. This item documented an
evaluation of the 230 kV Switchyard 125 VDC Power System for float
charging the switchyard batteries at a higher voltage. The Oconee 230
kV switchyard batteries were recently replaced with larger capacity
cells with a 60 cell battery bank. The previous batteries were 59 cell
batteries and were floated at 2.17 to 2.25 volts per cell (VPC)
resulting in a maximum system voltage of 132.75 Volts Direct Current
(VDC). With a 60 cell configuration and a maximum floating voltage of
2.25 VPC the maximum system voltage would be increased to 135 VDC. An
evaluation of the 125 VDC system components was needed to determine if
the increased voltage would exceed the allowable voltages of the devices
powered from the batteries and charger.
Calculation OSC-5976 Rev. 0, 230 kV Switchyard 125 VDC Power System
Overvoltage Evaluation was reviewed by the inspector. The inspector
reviewed each component identified within the calculation that would be
affected by the increased voltage. Documentation was included within
the calculation to document the acceptability of the increased voltage
for these components. All components identified within the calculation
were found to be adequate for application in a system operating at the
higher voltage with the following exceptions:
(1) Cooper Bussman type REN fuses are recommended for replacement by
Cooper Bussman with type KNW-R, KTN-R, and FRN-R type fuses.
(2) Cooper Bussman type NON-0.5 and NON-3 fuses should be replaced
with NON-15 type fuses.
(3) Gould Shawmut type OT-6 fuses rated 10 amperes or less should be
evaluated and replaced as necessary.
14
PIP No. 4-095-0522 was initiated to document the fuse replacements.
This PIP is still open. The 125 VDC System is being operated at the
original float voltage of 132.75 VDC pending completion of the
corrective actions required by this PIP.
No additional concerns were identified by the inspector during this
review. PUP item 113 was closed.
6.
Exit Meeting
The inspection scope and results were summarized on June 16, 1995, with
those individuals indicated in paragraph 1. The inspector described the
areas inspected and discussed in detail the inspection findings. There
was no dissenting comments received from the licensee. Proprietary
information is not contained in this report.
(Closed)
Violation 50-269, 270, 287/94-26-01, Failure to Perform
Procedure as Corrective Action
(Closed)
Deviation 50-269, 270, 287/93-02-02, Deviation from
Separation and Color Coding Requirements for Cable and
Overpressure Protection for Piping
7. Acronyms and Abbreviations
CFR
Code of Federal Regulations
EDSFI
Electrical Distribution System Functional Inspection
Final Safety Analysis Report
IEEE
Institute of Electrical and Electronics Engineers
IFI
Inspector Follow-up Item
IP
Inspection Procedure
IR
Inspection Report
kA
Kiloamperes
kV
Kilovolts
KVA
Kilo- Volt- Amperes
Loss of Coolant Accident
Motor Control Center
MVA
Mega Volt Amperes
NRC
Nuclear Regulatory Commission
Nuclear Reactor Regulation
Mega-Watts
Problem Investigation Process
PUP
Power Upgrade Project
SSF
Safe Shutdown Facility
TI
Temporary Instruction
V
Volts
VAC
Volts Alternating Current
VDC
Volts Direct Current
VPC
Volts Per Cell