ML20198D288
| ML20198D288 | |
| Person / Time | |
|---|---|
| Site: | Catawba  |
| Issue date: | 05/14/1992 |
| From: | Tuckman M DUKE POWER CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9205190301 | |
| Download: ML20198D288 (22) | |
Text
Il 1
nutwr,caran y s a, um f'afwta %:l<ct Gewution Ur;V:ine r:I Vu, e (% nde nt 4 4$ ('c u orJ In ad 7sy; \\1I.;c<:; otro e W L hr 3 i45 on tat) 3t?s in F
ouxepowEn May 14,1992 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D.C. 20555
Subject:
Catawba Nuclear Station Docket Nos. 50-413 and 50-414 Reply to a Inspector Follow-up item Electrical Distribution System FunctionalInspection Inspection Report Nos. 50-413/92-01 and 50-414/92-01 Attached is Duke Power's response to the Inspector Follow-up Item 413/414,92-01-07 (twelve findings) cited by the subject inspection Repoit dated March 18, 1992.
These fmdings were identified in the areas of design, procedural adequacy, maintenance, and testing.
Very truly yours, EW MMQ M.S. Tuckman JLUEDS Attachments 0
9205190301 920514
[g I PDR ADOCK 05000413 Q
PDR 6
-.,s a
U.S. Nuclear Regulatory Commision May 13,1992 Page 2 xc:
S.D. Ebneter Regional Administrator, Region 11 R.E. Martin, ONRR W.T. Orders Senior Resident Inspector i
i o
e b
>m-e
.,-,a
-..w-n v.
bxc:
R.L. Gill R.O. Sharpe M.E. Patrick A.V. Carr NCMPA-1 NCEMu SREC PMPA Group File: CN-815.01 (92 01)
Document Control: 801.01 l
DUKE 10WER COMPANY REPLY TO AN INSPECTOR FOI. LOW.UP ITEM IFI 413/414, 92-0107 Finding #1 - Short-Circuit Studies Were Not Available (paragraph 2.3.1).
RESPONSE
11ackground:
Calculation CNC-1381.06-5, Rev.1. "7500 KVA,6900/4160V Power Transformers Impedance Calculation" (reviewed by the EDSFI team) demonstrated that the 4160V essential switchgear fault-clearing capability will not be exceeded.
Neglecting all resistances and AlLCahlc inEnlarges, the calculation considered both maximum LOCA loading with the Diesel Generator unconnected and Normal Operation loading with the Diesel Generator in parallel with the system i
(test mode). Using the minimum 5.5% impedance of a 6% standard, the maximum available short-circuit MVA at the 4160V switchgear is 206.6MVA or 83% of the 250MVA rating.
However, this calculttion does not consider maximum switchgear voltage and does not compare actual calculated short-circuit currents at maximum voltage to recorded switchgear ratings.
1.
Corrective 4ctions Tnken nnd Results Achieved From the designed values shown above, no reasonable emount of overvoltage will jeopardize the 4160V switchgear capability. Even if the bus voltage should reach 110%
(which is greater than the worst-case result from voltage study CNC-1381.05-70, Rev.
2), the designed maximum short-circuit duty would become 227.3MVA or 91% of the switchgear rating, A preliminary computerized "as-built" calculation was performed on February 11,1992 during the audit. Applying the conservative 10% overvoltage to its results yields a maximum symmetrical fault current of 28.4KA or 89% of the switengear short-circuit rating at 110% of 4160V.
l=
l l
1
DUKE POWER COMPANY REPLY TO AN INSPECTOR FOI.I.OW.UP ITEM IFl 413/414. 92-0107 2.
_ Corrective Actiontig.he Tnken to Prevent Ftitth1T_lutliitt115 A formal computer based "as-built" calculation will be included in a comprehensive calculation for all switchgear and safety motor-control center busses. In the April 15, 1993 update of the Catawba FSAR, the rating of the 4160V switchgear will be corrected (to 35KA at 4.16KV) and the designed maximum short-circuit duty will be stated (from CNC-1381.06 5) as above,227.3MVA at i10% of 4160V.
3.
Date of Full Conepliance Duke Power will be in full compliance by April 15, 1993.
t L
l-2 1
l
1 DUKE POWER COMPANY REPLY TO AN INSPECTOR FOLLOW UP IT131 IFl 413/414, 92-01-07 Finding #2 - Procedures Relative to Diesel Generator Ground Fault hiay be Misleading (paragraph 2.3.4).
1
RESPONSE
1.
Corrective Actions Taken nnd Results Achieved A procedure revision was completed to OP/l(2)/B/6100/10L (Annunciator Response For Panel 1(2)AD.ll). In this procedure, 'D/G A GND FAULT' alarm (page 6), Probable Cause #2 " GENERATOR NEUTRAL GROUND FAULT" was deleted since the 59DGN ground fault relay does not detect such a fault. The probable cause section was also revised to inform the operator that the alarm may be in response to a ground fault condition on the diesel generator or the 4 KV system. This directs operators to attempt to locate and isolate the fault by selectively opening breakers, conditions permitting.
The conditions which should exist prior to retransferring the essential busses to the offsite source following a concurrent LOCA and LOOP require factors, in addition to switchyard voltage, to be considered. Factors such as the amount of loading on the essential busses and the on-line generation and stability of the grid, should be assessed.
2.
Corrective Actiom to be Taken to Prevent Further Incidents Procedure EP/l&2/A/5000/03, " Loss of All AC Power", will be revised by July 31, 1992 to provide guidance to ensure appropriate conditions exist before re-transferring to the off-site source.
The recommendations from calculation CNC-1381.05-00-0070 for Reactor Coolant Pump Motor stading are currently being evaluated to identify plant operating modes, voltage level, etc. that may require restrictions prior to starting a Reactor Coolant Pump Motor.
Procedure OP/l&2/A/6150/02A " Reactor Coolant Pump Operation" will be revised to inenrporate the applicable recommendations by July 31,1992.
3.
Date of Full Compliante Duke Power will be in full compliance July 31,1992.
3
~
4 l
r DUKE POWER COMPANY REPLY TO AN INSPECTOR FOLLOW-UP ITEM IFI 413/414,92 0107 Finding #3 - Ground Fault Protection on 600V System (paragraph 2.3.8).
RESPONSE
1.
Corrective Actions Taken and Results Athleted A review of the original philosophy tojurper the solid state trip device ground function-on the load centers feeding 600VAC essential motor control centers was performed and revealed that the intent was to insure coordination between the load center and motor control center feeder breakers on the solidly grounded 600VAC Essential Auxiliary Power System. By defeating (jumpering) the ground protection, the motor control center feeder breaker would have ample opportunity to trip and clear ground faults without spurious trips which could de-energize an entire motor control center or even multiple motor control centers, if the load center incoming breaker's ground protection were activated.
Since the load center breaker ground function uses the same SCR output as all other breaker protective functions, the ground unit can not be used to alarm only. Even with the maximum pickup (1200 amps) of the ground function, there is no coordination with thermal magnetic circuit breakers above 70 amps continuous rating. Since n'ost essential motor control centers have a least one feeder of greater than 70 amps contintnus rating, application of the maximum ground fault setting would result in un-coordinate <i regions on the time-current curve.
2.
Corrective Actions to be Tnken to Prevent Further Incident; Studies will be conducted to evaluate 1) ground fault detection (alarm only) for the load i
center bm.akers; or 2) limiting available g ound current by resistance grounding.
3.
Ihte of Full Con 1pliance ite above described study will be completed by August 31,1992.
4 1
^
e
-n
,~---n,-m,-,
i DUKE POWER COMPANY REPLY TO AN INSPECTOR FOLLOW-UP IT131 IFI 413/414,92 01-07 Finding #4 - Overvoltages on Class IE 600V MCC's (paragraph 2.3.12).
1.
Corrective Actions Taken nndEcsvits Achieved I
Duke Power's corrective actions to date include the following:
- Examination of the effects of source voltage and bus loading on load center voltage:
Switchyard voltage and tap settings of transformers connected between the switchyard and load centers are set to ensure the minimum voltage necessary for equipment to operate during an accident, therefore, it is not practical to lower voltage by tap changes.
However, it should be noted that bus voltage will decrease during an accident due to the additional loading.
- Review of the IE loads connected to these busses during various plant operatir.g modes:
During normal operations, the majority of the IE loads are in a standby mode, not connected to the bus, and thus are not exposed to overvoltage conditions, llowever, there are a few loads, such as ventilation system loads and other miscellaneous loads, that run for extended periods of time during most plant conditions. These loads do have a risk of degradation.
2.
Corrective Actionslo_he_Taken to Prevent Finihrr_hiddfuls Corrective actions to be taken include performing a study of the following:
- Identify all 600V loads potentially exposed to voltage levels exceeding their maximum rating for extended periods of time. The long term solution may be to replace these motors with new ones rated at a higher voltage. This will be determined during the study.
l Determine the maximum sustained voltage which can exist at the load centers without causing degradation to loads based on the voltage drop between the load centers and their L
loads. A " worst case" circuit will be identified and an associated maximum allowable voltage will be calculated.
3.
Date of Full Compliance The above described study will be completed by December 18, 1992.
5
l l
)
r DUKE POWER COh1PANY REPLY TO AN INSPECTOR FOLLOW-UP ITEh!
IFI 413/414, 92-0107 Finding #5 - Inadequate DG Dynamic Performance Analysis (paragraph 2.4.2).
RFSPONSE:
Background:
Calculation CNC-1381.05-0065 is the dynamic analysis for the Emergency Diesel Generator (EDG) at Catawba Nuclear Station, simulating a LOCA/ LOOP loading sequence. In this calculation, a computer program called DGh1P ( Diesel Generator hiodeling Program ) is used to simulate the loading sequences. The program uses manufacturer provided models for the i
generator's exciter-regulator and governor instead of typical models found in IEEE Standards.
hianufacturer provided data is also used to model motors and loads.
Because of the software limitation of DGhiP, cable and bus impedances are not included in the model. The geacrator output terminal voltage is assumed to be the load / motor input terminal voltage. *1his assumption results in the calculated load / motor starting inrush currents being conservatively higher and the generator transient voltage dips to be lower than would be the case if cable / bus impedances were included. The net effect of the higher inrush current / lower generator voltage essentially offsets the omission of the cable / bas impedances with regard to the total system response, as has been confirmed when the DGh1P results were found to be in good correlation,'as expected, with information measured during a diesel generator test at hicGuire Nuclear Station.
CNC 1381.05-00-0065 complies with Regulatory Guide 1.9 which sets limits for the generator voltage, frequency and speed. The results shown in this calculation indicate that throughout the LOCA/ LOOP loading sequence the EDG voltage, frequency, and speed are within the limits set forth by. Regulatory Guide 1.9, even though actual load / terminal voltages are not calculated.
6
. _ _. _ ~
DUKE IUWER COMPANY REPLY TO AN INSPECTOR FOLI.OW-UP ITEM IFI 413/414, 92-01-07 1.
Carr.cclive ActinnLIttken and Results Achievet!
N/A 2.
Corrective Actions to be Taken to prevent Ftirlhtr_litddtnis Although the use of the existing DOMP is satisfactory in determining compliance with Regulatory Guide 1.9, Duke will replace the existing DGMP with a new computer program that will model individual loads. Validation of this program will require a very-elaborate test to be performed on the diesel generator and its associated loads (to be completed by July 15,1993). The complete simulation of the EDG system under the LOCA/ LOOP loading sequence using the new program will be performed after the test is completed and the program is validated.
3.
Date or Eull Complianct The schedule for completion of the more detailed dynamic analysis is July 15, 1994.
7 o
I
~
D11KE POWER C051PANY REPLY TO AN INSPECTOR FOI.! OW-UP ITES1 IFI 41; A14, 92-01-07 i
i Finding #6 - 120 VAC and 125 VDC System (paragraph 2.5).
RESPONSE
1.
Carrective Actions Taken nnd Reinll3 Achieved An administrative control has been established to restrict placing the plant in a condguration where an 825 AH Vital Instrumentation and Control Battery simultaneously feeds two load channels. Should this configuration be necest.ary due to certain Technical Specification surveillance requirements, the 825 AH battery feeding the two load channels would be declared inoperable.
The LCO time period for this condguration would be 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />.
2.
Corrective Actinn5.19 be Tnken tolinent Further bifldtu15 A comprehensive study will be conducted to evaluate the voltage concerns associated with the 120VAC and 125VDC Vital Instrumentation and Control Power Systems. This study will concentrate on the following:
- Eliminating unnecessary :onservatism in the values contained in the battery loading database.
Developing analytical models to calculate the available voltage at loads (in addition to the Vital Inverters) fed by the 120VAC and 125VDC Vital Instrumentation and Control Power Systems.
Performing additional battery loading tests under simulated worst case design basis conditions.
The plant conGguration at the time of the test will be extensively documented.
8 l
l l
l
_. ~ - _ =... _.
DUKE POWER COMPANY REPLY TO AN INSPECTOR FOLLOW-UP ITF31 IFI 413/414, 92-01-07
- For those cases where adequate voltage is a concern, appropriate corrective measures will be evaluated. Some corrective mee.sures which may be considered include:
- 1. Replacing the 825 AH Vital Instrumentation and Control Power Batteries with ones of a larger capacity.
- 2. Installation of larger cables.
3.
Date of Full Complhwrc The study will be complead by December 18, 1992.
9 l
l v.
,r..
.. -, r g
y
DUKE POWER COSIPANY REPLY TO AN INSPECTOR FOI. LOW-UP ITEh!
IFI 413/414, 92-01-07 Finding #7 - Battery Charger Capacity (paragraph 2.5.2).
RESPONSE
1.
Correcthe Acilons Tnktiuw1LResult1Atttlemi N/A 2.
Correcthe..Actlens to be Taken.lo PreveutFurther litrjdents j
l A comprehensive study of the 125VDC Vital Instrumentation and Control 11attery Charger capacity has been initiated and will include:
- Elimination of unnecessary conservatism in the values contained in the battery loading database.
- Performance of additional battery loading tests under simulated worst case design basis i
conditions. The plant configuration at the time of the test will be extensively documented.
3.
Date of Full Coinpliante The study will be completed by December 18,1992, b
1 r
=4 10
9 DUKE POWER COMPANY REPLY TO AN INSPECTOR FOLLOW-UP ITEM IFI 413/414,92 01-07 Finding #8 Procedure Deficiency which could result in Starting Air Flow Path to the DG being lost (paragraph 3.3).
RESPONSE
1.
Carnethe Actlplitl<en and ResultsAthlettd Procedores PT/l(2)/4350/16A & 16B have been revised to include the following sign off statemem to enclosures 13.1 and 13.2 " Valve Aligriment Checklist":
"If either valve is found in a closed position, notify the Unit Supervisor "
v The addition of this step to these procedures ensures that if the AS FOUND position for the valves is not in the correct "OPEN" position, proper notification is made. This will protect against irnproper valve positioning following the test.
2.
Correcti' e Actions to bc Takeat.to_frevent Furtlier inchlenis No further corrective actions are necessary.
3.
[htte of Full Complitutte Duke Power is now in full compliance.
I 11
I DUKE POWER COMPANY REPLY TO AN !NSPECTOR FOLLOW-UP ITES1 IFI 413/414, 92-01-07 Finding #9 -
Perform Waterhanmier Analysis of Nuclear Service Water System (paragraph 3.4)
RFSPONSE:
1.
forrective Actions TakUUmiRrmlLtAthitted PIR 0-C92-0058 was written February 14,1992 as a result of the concern over poiential for waterhammer induced damage at the Upper Containment Ventilation Unit (UCVU),
or in its connected piping, supports, or valves. Further interviews were conducted with personnel associated with original Nuclear Service Water (RN) System startup, flushing, hydrotesting, and performance testing in an effort to identify known or suspected problems. None were encountered.
Further calculations have now been completed for mot,. eperated valves upstream and downstream of the UCVUs in conjunction with NRC I&E Bulletin 89-10. These calculations iadicate the existence of negative pressure in the downstream piping under normal operation. These calculated values are being considered in a system waterhammer analysis currently underway.
2 2.
forrective Actions 19_br_Taken to Prevent FurtherjydddttS Due to the complexity involved in waterhammer analysis, and the specific problems associated with the potential modifications to resolve service water vacuum conditions inside containment, no quick resolution is advisable. Rather the following actions will be comp:eted as indicated:
l L
12
,,.u c,
e
=,-ag 2
..-_g
., -,+.,
m..m.r-
+,, -
-.c
DUKE POWER CON 1PANY REPIX TO AN INSPECTOR FOLLOW-UP ITEM IFI 413/414, 92-01 07 A.
Research documentation and write a Work Request to inspect UCVUs, piply, and hangers for damage during upcoming Unit 2 outage Completion: 2-28-93 E.
Gather pressure data during an upcoming stroke test of UCVU RN isolation valves Completion: 2-28-93 C.
Perform waterhammer analysis and compare results to inspection and test results Completion: 6-30-93 D.
Incorporate any changes warranted to this piping during design and implementation of NSMs CN-11248 and CN-20639 Completion: 12-31-96 3.
Date omill Compliente Duke Power will be in full compliance by 12-31-96.
k l
l 13 v
I
,o a,--.
,.,.n_-
.s
,,w..
-~r e,
1 L) UKE LOWER COMPANY REPLY TO AN INSPECTOR FOLLOW UP ITEM IFt 413/414,92 0107 Finding #10 -
RN Pump Structure Minimum and Maximum Temperatures Outside the Design Limits (paragraph 3.5)
RESPONSE
1.
intrettive Atlhos TnknuuulEnults Achievn]
The calculation associated with RN Pump Structure temperature (CNC-1211.00410-0019) i has been updated to include the eenect RN pump horsepower; a ventilation flow rate of 10,000 cim (nominal) minus 5% (i.e. 9,500 efm); and worst case, outside air temperature extremes. Under these new parameters, the temperatures within the pump house can exceed the original design envelope. However, as part of the calculational update, an analysis of safety related equipment was performed and showed that there are no detrimental effects due to the new temperature envelope.
2.
. Corrective Ac11ons to lic Trhutto Prevent Furtiter incidents The FSAR, Equipment Qualification Criteria Manual (EQCM), and the Design Basis Document (DBD) will all be revised to reflect the new design temperature envelope.
l 3.
Date of Full Comnthulte Revisions to the FSAR, EQCM taid DBD will be completed by April 1,1993.
s
~
14 l;.-
~
DUKE POWER COMPANY REPLY TO AN INSPECTOR FOLLOW-UP ITiti IFl 413/414, 92-01-07 i
Finding #11 - Housekeeping for Safety Related Electrical Equipment (paragraph 4.2)
RESIONSE:
1.
. Corrective Actions Taken.nnd Results_Achlutd A) 600V Essential Motor Control Centers Predefined Work Orders have been created for 1EOC6 Refueling Outage to coincide with other maintenance work which requires the B Train Essential Bus to Oc de-energ! zed.
. The Predefmed Work Orders will ensure that dust or dirt accumulations are removed and that each circuit breaker is exerciscd on/off a minimum of five times to a:d in lubrication and contact cleani"g of the trip mechanism.
B) 600V Essential Load Centers & 4.16KV Switchgear The 4.16KV switchgear and 600V load Centers are cleaned only when de-energized, during outages. Since the inspcetion, there have been no outages or opportunides to i
clean these breaker cubicles.
C) 125 VDC Distribution Panels -
During the NRC-EDSFI Inspection several 125 VDC distribuuon panels were found to L
be contaminated with dirt. Suc., an environment in cenjunction with smail amounts of l
moisture could accelerate deterioration cf electrical equipment. Preliminary inspection of the 125 VDC distribution panels revealed no degradation of this ecgipment.
l L
l i
l' 1
15 i
+. - ~
p:
o.
1 DUKE POWER COMPANY HEPLY TO AN INSPECTOR FOLLOW UP ITDI IFI 413/414, 92-01-07 l
2, Correctlic,Attlons to be Tnken to Prevent Further Incidentis A) 600V Essential Motor Control Centers The remaining Predefined Work Orders for Unit 1 A Train and Unit 2 A & 11 Train will be crcited prior to the next scheduled refueling outages. Under the current scheduling methodology, only one Essential Train is removed froni service during each refueling outage. These remaining Predefined Work Orders for both units will be completed by January 1,1996.
L B) (00V Essential Load Centers & 4.16KV Switchgear The 4.16KV,6.9KV,13,8KV switchgear, and 600V load Center established Predefined I
Work Orders are being used to clean and upgrade the materiel condition of the breaker I
cubicles. All station switchgear will be cleaned using the established Predefined Work Orders during the next three refueling outages on each unit (currently January 1,,996 for both units). The materiel condition will then be maintained every third outage for each breaker cubicle when its associated preventitive maintenance (PM) of the breaker is performed. The bus area of the 4.16KV, 6.9KV,13.8KV and 600V Load Center switchgear will be inspected and cleaned when the bus teardown and ic;pection is
(
completed. The teardown and inspection will be completed within the next three outages for the Essential 4.16KV bus and Essential Load Center buses, and within the next six outages for the remainder of the plant's switchgear and loadcenter buses. The teardown and inspection will be performed every nine outages for each bus with an interim inspection of the accessible portions of the bus every three outages when the associated l
- breaker PM's are performed.
16 L
L
a e
DUKE POWER COMPANY REPLY TO AN INSPECTOR FOLLOW-UP ITEM l
IFl 413/414, 92-0107 C) 125 VDC Distribution Panels Work Orders have been written to clean, inspect insulation and connections, and to test l
the breakers in the 125 VDC distribution panels for Unit I during IEOC7, which is scheduled to end January 1,1994. Unit 2125 VDC distribution panel inspections and breaker testing will be completed during the next Unit 2 refueling outage (2EOC5),
which is scheduled to end May 30,1993.
In addition, Predefined Work Orders will be developed by July 1,1992 to include the-125 VDC distribution panels in the current Duke Power Company molded case circuit breaker inspection and testing program. The Predefined Work Orders will ensure periodic testing and inspection of this equipment.
3.
. Date of FulLComnllante Duke Power will be in full compliance by January 1,1996.
f I
1 */
DUKE VOWER COMPANY REI LY TO AN INSPECTOR FOLLOW-UP ITEM IFI 413/414, 92-01-07 Finding #12 -Undetectat)le Failure Mechanism on MOV's (dual fuses in the control circuits -
paragraph 4.2).
RESPONSE
l
-1.
Conective Actions Takenlundentlis Achieved A review of the subject MOV circuits and other circuits with potentially undetectable fuses was conducted and established the following:
The subject valves are part of those associated with the Auxiliary Shutdown Compler l
(ASC). 'Ine ASC was designed to support a main control room evacuation (such as a Dre in the main control room). The additional parallel fuse prevents such a fire from causing a fault that will make the valve inoperable from the ASC (the additional ff provides separation between the control room and the ASC). Since some of the va've.,
have control room indicatien but no control room controls, the control room indicating lights were on a separate fuse. Similar cases exist for the Standby Shutdown Facility (SSF). No cases of dual fuses wert identified where one fuse could blow without being detected and render the valve inopvable.
Modification of the undetectable circuits to make any fuse failures detectable would require additional circuit componenta not needed for any other functions.
These components could compromise the reliability of the circuits and are judged to be more detrimental than beneficial.
!n
+
18 l-
DUKE POWER COMPANY REPLY TO AN INSPECTOR FOLLOW-UP ITEM IFI 413/414,92 01-07 y
2.
CDntttive Actions to be TMfn to Prevent Fut1her Incide',u Due to the unique design of these circuits, Duke Power has o ' c.itted to test the SSF and the ASC (four (4) panels / unit-both trains of the Auxiliary Giutdown Panel and the Auxiliary Feedwater Control Panel & SSF Panel). One (1) panel will he tested during each of the upcoming RFOs (both units), therefore the completion date for initially testing all eight (8) panels (both units) is October 1,1996. These panels will be tested periodically thereafter.
3.
Date of Full Cotilpliance Duke Power Company is now in full compliance.
D l-i-
l l
l 19 l
i