IR 05000352/1993026
| ML20058E485 | |
| Person / Time | |
|---|---|
| Site: | Limerick  |
| Issue date: | 11/19/1993 |
| From: | Greca A, Kay L, Ruland W NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML20058E475 | List: |
| References | |
| 50-352-93-26, 50-353-93-26, NUDOCS 9312070030 | |
| Download: ML20058E485 (11) | |
Text
.
.
.
i
,
.
U.S. NUCLEAR REGULATORY COMMISSION
REGION I
r REPORT / DOCKET NOS. 50-352/93-26
50-353/93-26 i
LICENSE NOS.
NPF-39 i
NPF-85 LICENSEE:
Philadelphia Electric Company
.
FACILITY NAME:
Limerick Generating Station, Units 1 and 2
!
INSPECTION DATES:
October 4-8,1993
!
INSPECTORS:
@
D Of3)
dnne M. KayAteactor Engineer Date
,
Electrical Sectihal EB, DRS
<
!
d A t _W 38 ! ! 9 9b Lib r'
Aniello L. Della Greca, SyIleactor Engineer Dhte
Electrical Section, EB, DRS
'/!/fh APPROVED BY:
/
-
.'
.
'Date
' William H. Ruland,' Chief ~
'
Electrical Section, EB, DRS l
!
!
.
$
9312070030 931124 PDR ADOCK 05000352 G
PDR_;
-
.
.
- -.
,
-
.
t
Area Inspected: This was an announced inspection to review the licensee's corrective actions of nine previously identified electrical distribution system functional inspection (EDSFI)
findings including an escalated enforcement item and two other violations.
Results: No violations or deviations were identified. Of the nine open unresolved items reviewed for Units 1 and 2, five unresolved items were closed including the escalated
,
enforcement item. The remaining four open items were updated. These items are discussed
!
in Section 2.0.
,
.
a i
The engineering evaluations and corrective actions for these issues were thorough and facilitated closure of several unresolved items. Procedures reviewed involving the testing of j
safety-related relay logic schemes were complete and design analysis studies were supponed
~;
adequately by test report information specific to Limerick.
.
b
'
!
l f
J f
i n
!
.
.
!
!
l i
i
.
I
!
!
!
.
_.. _
. _.,
.
,
DETAILS 1.0 PURPOSE The purpose of this inspection was to review and verify Philadelphia Electric Company's (PECo) corrective actions for previously identified NRC Electrical Distribution System Functional Inspection (EDSFI) findings.
2.0 FOLLOW-UP OF PREVIOUSLY IDENTIFIED FINDINGS
'
2.1 (Closed) Inspector Followup Item No. 50-352/92-81-01 (50-353/92-81-01)
,
periaining to the apparent lack of independence between onsite and offsite sources During a functional inspection of the electrical distribution system at the Limerick site, the inspection team concluded that, in the event of a loss of a coolant accident (LOCA)
-
'
concurrent with a loss of one offsite source, the design of the system may not be able to meet the independence requirements between onsite and offsite sources of 10 CFR 50, t
Appendix A, General Design Criterion (GDC) 17. The team's concern was that, under the postulated scenario, the addition of the accident loads to the remaining offsite source could i
temporarily degrade this source and initiate a periodic tripping of the emergency buses supply breakers, ultimately resulting in the inability to mitigate the accident.
[
I On January 12, 1993, an enforcement conference was held with PECo to discuss the apparent deugn deficiency. PECo stated at the conference that the addition of the emergency loads to the affected buses would have resulted in a temporary degradation of the remaining offsite soarce. However, breaker cycling would not have occurred because, after the first cycle, two of the four buses would have transferred, by design, to the diesel generators. The loads l
on the other two buses would not have been sufficient to drop the voltage of the remaining
source below the degraded grid relay setpoint. To address the concern regarding the initial i
level of the offsite source voltage, the licensee reduced the response time of the transformer l
automatic tap changers from 30 to 10 seconds and increased the actuation time of the offsite
.
source breaker from 0.25 seconds to one second. The latter change, after the first cycle, f
gave preference to the diesel generator source.
i Based upon PECo's clarifications and stated corrective actions, the NRC, with a letter, dated March 12,1993, informed the licensee that the issue was no longer considered to be an apparent violation of the NRC requirements.
During the current review, the inspectors confirmed that the modifications had been implemented and design changes had received appropriate review in accordance with
10 CFR 50.59. Based on review of the applicable design and plant documents, the
'
inspectors found the modification acceptable. This issue is closed.
i
-
l
T
l
,
2.2 (Open) Violation No. 50-352/92-81-02 (50-353/92-81-02) regarding the exclusion of several de loads from the surveillance procedure, Technical Specifications, and FSAR
During the September 1992, functional inspection of the Limerick electrical distribution
,
system, the team observed several discrepancies between the calculated Class IE battery
loads and the loads identified in the Technical Specifications, the FSAR, and the surveillance test procedure. In particular these discrepancies included:
,
a)
80 amperes identified in Appendix IX of calculation 6600.E.09 for spring charging
,
motor inrush current-
,
b)
28 amperes identified in calculation 6600.E.22, Revision 1, from a postulated undetected high impedance fault; and
,
c)
the ampere load resulting from the operation of the high pressure coolant injection l
(HPCI) pump discharge valve.
These loads had been evaluated by PECo but not included in subsequent revisions of the
!
Technical Specifications, FSAR, and surveillance testing procedure for the batteries. As a i
result, these documents provided non-conservative information. This rendered the battery service test non-conservative.
.
!
l In their letter, dated January 8,1993, PECo acknowledged the violation and stated that they
'
would revise the calculations to correct the inconsistencies. They also stated that a new calculation would be prepared to redefine the current component load profiles and the results
,
would be included in the licensing and operational documents, as needed.
!
During the current inspection, the NRC reviewed PECo's actions to address the above l
discrepancies and found that a new calculation had been prepared by a contractor. The calculation, LE-052, was prepared to determine the worst-case load duty cycle, the adequacy i
of the existing batteries to meet the load requirements, and the maximum load that could be
,
added to the batteries without exceeding their capacity. A review of this calculation by the (
i inspector determined that it had addressed all NRC concerns, but had not yet received approval from PECo engineering organization. Therefore, the calculation results had not j
been incorporated in the above documents, even though interim procedure changes were already in place. This item remains open pending completion of corrective actions by the f
licensee.
,
>
h i
a t
f
.
-
,
2.3 (Open) Violation No. 50-352/92-81-03 (50-353/92-81-03) related to the adequacy of the battery voltage requirements specified in the surveillance test procedure In conjunction with the review of the de electrical distribution system described in Section 2.2, above, the NRC team observed that calculation 6470.E.24, Revision 3, based the allowable circuit length on the assumption that the battery terminal voltage would not drop below 108 Vdc. However, this design requirement was not reflected in the Technical Specifications or the battery surveillance test procedure.
In their letter, dated January 8,1993, PECo acknowledged the violation and indicated that the revised calculation would define the required battery terminal voltage and the results of the calculation would be used as input to the applicable voltage drop calculations. In the l
interim, the surveillance procedures would be revised to include the loads currently not addressed by the test profile. During the current inspection, the NRC reviewed the status of the licensee's corrective actions pertaining to this issue and determined that the voltage drop calculation being performed by a PECo contractor was still incomplete. However, interim i
procedure changes had been in place to address the NRC concerns. This item continues to l
be unresolved pending completion of the calculation and appropriate procedure changes.
l 2.4 (Closed) Unresolved Item No. 50-352/92-81-05 (50-353/92-81-05) regarding transformer sizing and establishing load controls
The EDSFI team determined that under worst-case loading conditions, safeguard transformers and certain 4 kV load center transformers could be overloaded. These overloads are small and shown below.
Transformer Max Rating leading Demand Overload Safeguard 14 MVA 15 MVA 1MVA 10B201 1000 kVA 1034 kVA 34 kVA 10B202 1000 kVA 1079 kVA 79 kVA l
The licensee stated that these overloads were acceptable based on ANSI Standard C57.96 which allowed a greater overloading of the transformers without sacrificing their normal life expectancy. However, the licensee had not established operating instructions to ensure that the load carrying capability of the transformers as defined in the ANSI standard was not exceeded to prevent degradation of the transformers.
In response to the team's concerns, PECo engineering reviewed electrical loading calculation 6300E.18, " Load Study-Station Auxiliary Power System," to determine actual bus and transformer loading during emergency conditions. By securing the appropriate loads, transformer loading would remain below the maximum ratings. Based on this review, loads l
l l
l_
_ _ _
-
t
>
including various lighting panels and the turbine turning gear were determined not to be
,
required after 10 minutes into an emergency shutdown condition. To ensure the transformers
-
will remain within the design ratings, the licensee has revised plant operating procedure
,
!
SE-10. "LOCA," Revision 27, to secure these loads.
The inspectors reviewed the licensee's corrective actions discussed above and verified the transformer size was adequate based on expected loading requirements, including the securing of loads not required as discussed above. Based on this review, the inspectors concluded these actions were adequate to prevent the transformers from being overloaded.
This item is closed.
2.5 (Open) Unresolved Item No. 50-352/92-81-06 (50-353/92-81-06) regarding the 120 Vac voltage study
!
During the EDSFI, Limerick Generating Station (LGS) had not performed a voltage study for safety-related 120 Vac components. Based on this, the acceptability of the voltage adequacy was unresolved pending completion of the licensee's study.
A voltage calculation study / report has been completed by a contractor for LGS to verify that
all equipment supplied from 120 Vac safeguard distribution Y panels have the minimum l
voltage required for safe operation. The bases for the study were taken from the controlled as-built drawing data for determining the voltage drop to each component utilizing a voltage drop computer software program and worst-case motor control center bus voltages based on
,
the Limerick Voltage Regulation Study. However, this calculation contained many conservative assumptions invoiving resistive load values as verified by the inspectors. These load values under accident conditions would be less than the values presented in the study and included resistances for indicating lights, hydraulic dampers, and cable lengths. Based
'
on use of these assumptions, discrepancies were identified involving the voltage supplied to certain 120 Vac components. The inspectors reviewed these discrepancies and determined that the required voltage values for the existing system design would be determined once assumptions were validated with actual plant data.
At the time of this inspection, PECo was reviewing the voltage study and incorporating plant specific data and test results into the report for resolution of identified discrepancies. This
- '
item remains open pending PECo's resolutions of the voltage study.
2.6 (Closed) Unresolved Item No. 50-352/92-81-07 (50-353/92-81-07) pertaining
to the use of 35 A fuses in control circuits The EDSFI team review of ac system protection and coordination revealed that closing, tripping, and control circuits of 4.16 kV breakers were protected by a single 35 A fuse. The
-
team noted that the use of 35 A fuses is appropriate in trip circuits due to the high current drawn by the trip coil, but seldom necessary in control circuits. For control circuits, a fault
.
.
.
generated at the end of a cable may not generate a fault current high enough to be interrupted by the fuse and may cause damage to surrounding cables and prevent other equipment from performing their function.
,
,
In response to the team's concern, the licensee performed a design analysis to evaluate the acceptability of 35 A fuses in 125 Vdc control circuits for medium voltage (4.16 kV)
switchgear. This analysis stated the objective of the fuse was to provide dependable and reliable power for the switchgear control circuits including the 4 kV safety 4 elated switchgear to trip successfully or close the switchgear breakers when required. These breakers would l
I be required to trip during fr. ult conditions that were evaluated in this design analysis.
The licensee determined that the use of 35 A fuses was acceptable. Results of the evaluation showed that for those postulated faults where the short circuit current was less that the 35 A
'
required to open the fuse, the relay coil would continue to heat up until it burned open, thus separating it from the rest of the circuit. The analysis further stated this discontinuity was an acceptable design as it was consistent with the facility's single failure criterion.
Supporting information to determine the limit at which the cable used at Limerick will fuse
.
open and affect the surrounding cables was provided in " Design Verification Test Report l
48503." This was an independent test conducted to support Appendix R separation criteria i
whereby cables were exposed to sustained overcurrents to determine cable limits and effects.
'
Results of these tests demonstrated that the sustained overcurrents were within the rating of l
the fuses and would not cause damage to the cables in question.
-
%
The inspector reviewed the design analysis and test report 48503 discussed above. Based on this review the inspector concluded that the 35 A fuses were adequate to provide reasonable
,
'
protection for the control circuits. This item is closed.
2.7 (Open) Unresolved Item No. 50-352/92-81-08 (50-535/92-81-0F) involving the overloading of emergency buses i
The EDSFI team identified that nonsafety-related loads were connected to safety-related buses. These nonsafety-related loads were the drywell chiller and control rod drive pumps.
>
Power to the control circuits for these loads comes from nonsafety-related motor control j
centers that are reenergized a few minutes into an event via operator action. The team noted j
that although these loads are shed from their respective buses during a loss of coolant accident, a failure of the nonsafety-related components could cause the loads to be applied to
the safety-related buses in an uncontrolled manner resulting in an overload of the buses.
,
In response to this concern, PECo had an evaluation of this scenario performed by a
,
contractor. At the time of this inspection, PECo was reviewing the recommendations presented in the evaluation. This item remains open pending PECo's completion of actions f
to resolve the possibility of overloading the emergency buses.
i i
)
,
~
,
'
2.8 (Closed) Unresolved Item No. 50-352/92-81-09 (50-353/92-81-09) regarding f
RHR and Core Spray pumps The EDSFI team's review of residual heat removal (RHR) and core spray pump curves for j
determining horsepower requirements revealed that pump horsepower was dependent on
,
backpressure at the outlet of a pump. The need for maintaining a backpressure on the pump was explained by the pump vendor. However, the licensee was unaware of this requirement.
During this inspection, the inspector reviewed licensee actions to resolve the need for and i
acceptability of a minimum backpressure for the RHR and core spray pumps. The licensee contacted the RHR and core spray pump manufacturer (Ingersoll-Rand) and performed an evaluation of the pump backpressure as it relates to the system con 6gurations. The
'
evaluation noted, based on the manufacturers response, that with proper net positive suction head (NPSH) available and a fully primed discharge nozzle, a pump would perform as
,
designed. The evaluation further stated that the required NPSH is ensured by maintaining the suppression pool level within technical specification limits, and the requirement for a l
fully primed discharge nozzle is ensured by the pumps being installed at an elevation below the suppression pool. Additionally, the pump discharge lines are maintained full by either
'
the condensate transfer system or the safeguard piping 511 system. The evaluation concluded that minimum backpressure is not an issue for the RHR and core spray pumps in the as-
,
installed system con 6gurations. The inspector reviewed the evaluation and supporting documentation and determined the licensee's conclusions were acceptable. This item is
closed.
.
2.9 (Closed) Unresolved Item No. 50-352/92-81-10 (50-353/92-81-10) involving relay response times and testing The EDSFI team identined that surveillance test results involving a combination of two relays associated with emergency bus D14 showed that the total time delay exceeded the acceptance criteria limit established by the respective test procedure. Although the actual timing for relays 127-11702 and 102-11802 was within the Technical Speci6 cation (TS) limit of 79.0 to 87.0 volts and 1.5 seconds, the team was concerned that the test procedure did not have provisions to measure the individual response times of the undervoltage relay and its associated Agastat time-delay relay. Additionally, review by the NRC team of LGS relay test results for the above General Electric NGV undervoltage relay (127-11702) indicated that response times at differing voltage levels varied and led the team to question the validity of l
the test.
,
In response to the team's concerns regarding the varied relay response times, LGS performed
<
a design analysis, "Ana'pc the Adequacy of the Existing NGV Undervoltage Relay Design
and Testing Methodology." Results of this analysis identified that mechanical toggle switches used in the test boxes (for surveillance testing) could introduce a timing error as large as 389 milliseconds. However, this tolerance was acceptable because the TS allowable values for the undervoltage relays range from 1.5 to 64 seconds and trip setpoints range from
!
.
.
~
,
.
,
1.0 to 61 seconds. Although this test box error is acceptable based on its value being insufficient to exceed TS limits, the licensee has determined that increased accuracy of the test boxes and relay dropout values are necessary. Discussions held between the inspector and relay system engineer provided information regarding licensee actions to install new test boxes to significantly reduce timing error.
The licensee's design analysis also reviewed the methodology for relay testing. This analysis included discussions with the relay vendor. Based on this review, LGS has determined that relay dropout time should be measured at 0 volts as opposed to the setpoint dropout voltage.
This difference adequately eliminates relay contact chatter and allows the measurement to be made with a firmly established contact make-up.
The inspector reviewed the varied levels of undervoltage protection at LGS including NGV
relay characteristics and calibration / functional test procedures. This review was performed j
to verify the adequacy of the relay testing schemes including logic systems for buses,
transformers, and motors. Relay testing is performed in-house by PEco personnel who also perform testing at Peach Bottom.md PECO's fossil plants. At Limerick, the Instrumentation and Controls (I&C) Technical Support Group have relay responsibility. Testing is performed in accordance with 1&C procedures organized under the Routine Test (RT-2) discipline of
,
procedures. Formerly, these procedures were under the RT-8 title, but have been rewritten to consolidate all logic system functional procedures. This change occurred when I&C took
ownership of relays from a previous contractor who had the RT-8 responsibility.
'
The inspector determined by review of the following procedures and discussions with site and corporate personnel that these procedures provided for thorough testing of logic schemes for
all emergency buses including overlap testing.
[
ST-2-092-321-1 "4 kV emergency D11 bus undervoltage channel calibration / functional test"
ST-1-092-101-1 "Div I 4 kV safeguard bus undervoltage LSF/SAA"
,
ST-1-092-Ill-1 "D11 Gen 4 kV safeguard LOP /SAA and outage testing"
The inspector verified these procedures adequately tested emergency buses Dil, D12, D13, and D14 and D21 D22, D23, and D24.
!
The licensee stated that bus relay testing is performed during outages, motor protection is done during system outages, and protective relays are tested while at power due to overlap protection. Protective relay testing was performed while at power under 10 CFR 50.59.
The inspector identified no concerns with this method of test.
-
Based on the licensee's corrective actions to measure individual relay response and delay times, design analysis study, and review of relay calibration procedures, the inspector
,
'
determined that the licensee adequately tested relays to ensure they perform their intended safety functions. This item is close.
.
,
3.0 UNRESOLVED ITEMS Unresolved items are matters about which additional information is necessary to determine whether they are acceptable, a deviation, or a violation. Several unresolved items are discussed in detail under Section 2.0.
4.0 EXIT MEETING The inspector met with the licensee's personnel denoted in Attachment 1 of this report at the conclusion of the inspection on October 8,1993. At that time, the scope of the inspection and inspection results were summarized. The licensee acknowledged the inspection findings as detailed in this report and had no additional comments regarding the results, i
e f
a J
I t
.
o b
+
,
b
-
- - - - - - - - - - -
..
.
.
.
.
.
ATTACILMENT 1 Persons Contacted Philadelohia Electric Company
- W. Boyer, Electrical Engr., NS A. Giancatarino, Quality Assurance
- S. Hutchins, Electrical Engr, NS
- R. Krich, Manager, Licensing T. Shannon, Electrical Engineer U.S. Nuclear Regulatory Commission
- B. Welling, DRP, Region I Denotes attendance at exit meeting
l i
!
i
,
$
,
!
.
...
.
,_. -
,_
-
.