ML20077R712
| ML20077R712 | |
| Person / Time | |
|---|---|
| Site: | Catawba  |
| Issue date: | 10/13/1994 |
| From: | Abraham P, Boyer R, Fanish P DUKE POWER CO. |
| To: | |
| Shared Package | |
| ML20077R694 | List: |
| References | |
| CNC-1535.00-, CNC-1535.00--7, CNC-1535.00-00, CNC-1535.00-00-0007, NUDOCS 9501230090 | |
| Download: ML20077R712 (19) | |
Text
_ _ _ _ - - - -_ _
. 10-14-1994 10:47AM FROM DPC NUCLEAR ENGINEERING TO 88313400 P.02 ATTAC190DIT 3 PRA Fo~r EPL and EFE Systrams For 10 CFR 50.59 (IEEE 308 - 1974) FORM 101.1 REVISION 14 Fom: 01777 fR7-92) )
CERTIFICATION OF ENGINEERING CALCULATION l STATION AND UNIT NUMBER Catawba Nuclear Station. Units 1 & 2 TTTLE OF CALCULATION Br-@~ Coordination Evaluation for the 125V de Vital I & C Power Svstem EPL) and the 600V ac Essential Auxiliary Power 5vstem EPE)
CALCULATION NUMBER CNC 1535.00-00-0007 ORIGINALLY CONSISTING OF:
PAGES 1 THROUGH t TOTAL MICROFICHE ATTACHMENTS 0 TOTAL ATTACHMENTS 3 TOTAL VOLUMES 1 . TYPEI CAlfULAT79N/ ANALYSIS YES O NO W
- 'IYPE I REVIEW FREQUENCY N/A THESE ENGINEERING CALCULATIONS COVER QA CONDITION _I_ TIT.MS IN ACCORDANCE WITH ESTABLISHED PROCEDURES. THE QUALITY HAS BEEN ASSURED AND I CERTIFY THAT THE ABOVE CAlfULATION HAS BEEN ORIGINATED.C OR APPROVED AS NOTED BE.OW:
ORIGINATED BY ,,N /d - s DATE /c //o !44 CHECKED BY R.h. [ [ DATE /d//ah94 APPROVED BY [M kh ed.Aw _DATE lo//3/99 ISSUED TO DOCUMENT CONTROL DATE RECEIVED BY DOCUMENT CONTROL DATE MICROFICHE ATTACHMENT LIST O Yes 5 No SEE FORM 101.4
- .tracww.wrs nou votuu.s omas cass .ry.
c.t.cuurm eno.s wou
.. .. .. . ,_ . -,m. . .
9501230090 941229 PDR ADOCK 05000413 P_ _. _.
,PDR
,10-14-1994 10840AM FROM DPC NUCLEAR ENGIEERING TO 88313400 P.03 CNC 1535.00 00-0007 By M! Date /*//M96_
'I pg.' 1 of.5 NUCLEAR ENGINEERING GROUP ENGINEERING CALCULATION PROCEDURE APPLICABILDT CHECKLIST Descripiloo of Analysis The Severe Accident Analysis Section (SAAS) received a request from Catawba Electrical Engineering to evaluate the significance of the absence of breaker coordination on portions of the 125V de Vital 1 & C System (EPL) and the 600V ac Essential Auxiliary Power System (EPE) distribution centers. The increase in the core melt frequency resulting from system failure should be determined along with any appropriate sensitivity studies.
?
e Determination of QA Condition 1 Applicability YM NO Does this analysis determine the presence or absence of an X ""'*"i'**d 5*'**Y 4"*5'i "?
Does this analysis justify a change in a Technical Specification ~
Hmit r verify the acceptabiUty, of a current Technical X Specification limit?
Does this analysis justify a design or a change in the performance or design of safety-related structures, systems, or components?
y Does this analysis modify or justify the licensing basis safety
- ^^i Y5i57 X
Is this analysis intended to provide the basis for, or input to, other y safety-related analyses?
If the answer to any of the above questions is yes. then this analysis is safety related and must be -
classified as a QA Condition 1 item. As such. it must satisfy the requirements of NE-103.
i Fonn NE-103.1 Revision 0 l
6e-14-ina ze:4t>rl FROM DPC NUCLEAR ENGINEER NG 8831y,00 P.04 ,
, i CNC-1535.00-00-0007 By /N,f, Date i=lrol44 i P
NOT YES APPLICABLE j
- A description of the analysis has been entered on Form NE-K 103.1.
The QA Condition of the calculation has been determined X on Form NE-103.1 and entered on Form PR-101.1.
Design rnethods and procedures have been referenced.
Design criteria have been identified.
Input data and assumptions are valid and properly documented.
K All computer programs are properly identified, documented.
and executed consistently with their derivation.
K All computer programs have been certified in accordance ,
with NE-103.
X The calculation has been presented in sufficient detail to g permit an adequate review.
Calculation and analytical methodologies are consistent with approved methodologies and numerical results have been r N verified.
i X Ccinclusions and results are consistent with the calculations.
f 4
P
..a, , . _ _ . . _ . . _ . _ ..., . _
a0-14-1994 10849AM FROM DPC NUCLEAR ENGDEERING TO 88313400 P.05 l CNC-1535.00-00-0007 By M ! Date /a//./94_
'/' pg. U of_3.__
BREAKER COORDINATION EVALUATION FOR THE 125V DC VITAL I & C POWER SYSTEM (EPL) AND THE 600V Ac ESSENTIAL AUXILIARY POWER SYSTEM (EPE) 1.0 Statement Of Problem The Severe Accident Analysis Section (SAAS) received a request from Catawba Electrical Engineering to evaluate the significance of the absence of breaker coordination on portions of the 125V de Vital I & C System (EPL) and the 600V ac Essential Auxiliary Power System (EPE) distribution centers. The increase in the core melt frequency-resulting from system failure should be determined along with any appropriate sensitivity studies (Ref. 9.1).
2.0 Relation to OA Conditions Because this calculation is used to justify a design in safety-related systems and '
components, it is classified as OA Condition I.
3.0 Analvtient Model Emnioved The Catawba Probabilistic Risk Assessment (FRA) report serves as the basis for this-evaluation. All appropriate portions and corresponding data will be used.
4.0 Aeolicable Codes and Standards .
N/ A 5.0 Desien Inouts See Reference 9.2 6.0 FSAR Criteria .
t N/ A 7.0 Assumntions All assumptions made are included in the body of this calculation.
10-14-1994 ta:49AM FROM DPC NUCLEAR ENGINEERING TO' 88313400 P.06 CNC-1535.00-00-0007 By _ 84! Date / ole /44
'/' pg. # of 2_ l i
8.0 Eneineerine Judements l None :
9.0 Re;ferences 9.1 1.etter from R. E. Martin (USNRC) to D. L. Rehn (Vice President, Catawba Nuclear Station); 9/14S4 (Attachment 1) ;
i 9.2 Memo from Allen Dickard (Catawba Electrical Engineermg) to Paul Farish (Nuclear Engineering); 7/7S4 (Attachment 2) 9.3 Catawba Probabilistic Risk Assessment Report: Rev.1: File: CN-1535.00: 9/22N2 9.4 Duke Drawing CN 1705-01.01: One Line Diagram: 125 V de Vital I & C Power System (EPL); Rev. 8; dated 1/20S4 I
9.5 FAX transmission' from Allen Dickard (Catawba Electrical Engineering) to Gary l Cruzan (Nuclear Engineering); 9/30#4 (Attachment 3) l 10.0 Calcuhtions l Attachment 2 (Ref. 9.2) provides .much of the background information for this j investigation. The evaluations are carried out for Unit 1. Due to plant similarities, it is l assumed that the sesults for Unit 2 will be identical.
10.1 EPL System-I Per Ref. 9.2, since the EPL System is an ungrounded system, the postulated faults j require that either a simultaneous positive to-ground and negative-to-ground fault l occur or a double line (positive-to-negative-line) fault occur. He former . type of ;
fault requires that two failures occur, which is beyond the requirements of the i i
single failure criterion and :he design basis of the plant. Should a single line to-ground fault occur, the power system will remain unaffected: however, a ground. !
fault detector will alert the control room operator via both an annunciator and - l computer alarm.
From Ref. 9.2, the latter type of fault is extremely unlikely. To illustrate this, the site conducted a thorough study of both the Nuclear Plant Reliability Database System (NPRDS) for all U.S. nuclear plants a:: well as the Catawba PRA (Ref. 9.3). ,
Failure discoveries made at Catawba since 1985 and all U.S. plants since 1990 were analyzed. Only three reported cases were found where a double fault occurred on a de system. One case at Catawba involving a shorted lamp holder was attributed to improper installation during previous maintenance activities. The other two cases
$10-14-3994 10850Af1 FROM; DPC NUCLEAR ENGINEERING TO- '88313400 07
.. i CNC-1535.00-00-0007 By d -.- Date MsN4
- / pg. C'of _ f*
1 i
occurred at other utilities and involved component failures within battery chargers.
In both cases, plant status was reported to not have been affected. No reported cases
)
were found that involved double line faults attributed to cable failures. The risk of cable failures is further reduced at Catawba since 2KV-rated, interlocked armor cable is used throughout the plant. Finally, engineering personnel at other utilities were contacted regarding their experiences with faults on safety-related de power l systems. No faults capable of challenging the EPL System were reported. ]
)
Faults were postulated at several points on the EPL System (refer to Ref. 9.4 and Attachment 2). The worst case faults result in a loss of one of four load group distribution centers, lEDA. lEDB, IEDC, or lEDD. nese load group distribution centers are important to normal plant operation but.none are essential foreplanty e shutdown.1Aad Group Distribution Center 1 EDA is auctioneered with power from Diesel Generator Control Panel 1A to provide a highly reliable power . supply for train A Vital I & C Power Distribution Center lEDE. Train B VitalI & C power on '
IEDF is similarly powered from Distribution Center lEDD and Diesel Generator Control Panel IB. The Auctioneered Distribution Centers lEDE cnd IEDF are essential for plant shutdown. None of the faults examined caused the complete loss of Auctioneered Distribution Center lEDE, although power from one of the two ,
Auctioneered Diode Assemblies providing power to 1EDE would be lost when its ,
associated load group distribution center fails by fault. 'Ihe second of the auctioneered distribution center's power supplies is a train of the 125V de Diesel i Essential Auxiliary Power System which is unaffected by any of the documented l breaker coordination problems. Herefore, it follows that the loss of any one of the '
four load group distribution centers above poses no significant effect on a unit's core damage probability. l To confirm this conclusion, the Catawba PRA results were examined in light of l these potential faults. One of the initiators evaluated in the Catawba PRA is the '
Loss of Vital I & C Power Bus (T14) which, for Catawba, is Auctioneered Distribution Center IEDF (Train B control power). Per Ref. 9.3, the T14 initiating frequency calculated for Catawba is 5.0E-02/yr. The resulting total CDF, as ,
calculated with the Risk Management Query System'(RMQS) module of the CAFTA (fault tree analysis) software, is 7.76E-05/vr.
. Sensi'tivity Analysis - When the T14 event is assumed to occur (i.e., probability - :
= 1.0), RMQS calculates a total CDF of 7.88E-05/yr., an increase of LlE-06 or . ;
1.551. nis is considered to be an insignificant additional contribution. ,
herefore, it is evident that most of the cutsets p;rtaining to the load group i distribution centers are not of a high enough probability to significantly contribute to the overall CDF.
Since the risk of core damage due to the loss of load group distribution centers is l even less than the risk due to the loss of the auctioneered distribution centers, and :
since a large change in the probability of the loss of the auctioneered distribution l centers is of little significance to the overall CDF, it is conduded that an l 1
4
I .
10-14-1994 10tSOAM FROM DPC HJCLEAR ENGINEERING TO 88313400 P.00 CNC-1535.00-00-0007 By N /'
Date /a//a/M pg. __/c., of f ,
inrovement of load strono distribution centers' brealer coordination will have a n_ eolicible effect on the risk of core dammee.
10.2 EPE System -
Per Ref. 9.2, the incoming breakers for motor control centers (MCCs) 1EMXA, 1EMXB, IEMXC 1EMXD, IEMXE,1EMXF,1EMXI,1EMXJ, IEMXK, and 1EMXL are coordinated for the worst case postulated fault at the first cable termination outside the MCC. MCC IEi4XG utilizes two incoming breakers because it can be powered from either Unit i load center IELXA or Unit 2 load center 2ELXA. It is understood that MCC 1EMXG is not coordinated becaut: it's.
incoming supply breakers (F03A from load center 1ELXA or F06A from :oad center 2ELXA) will open for a large instantaneous current before the individual load breaker to Unit 1 Control Room Air Handling Unit (AHU) #1, F05C. All other loads on MCC 1EMXG are fed from smaller breakers and cables with lower maximum fault current for which the incoming breakers are coordinated.
He loads supplied by 1EMXG are shown in Attachment 3 (Ref. 9.5) and consist of the following:
- various HVAC components
. the feeder to power panelboard transformer IEKTG, which supplies control room ventilation system control power
. valv:s RN54A, RN57A, & RN63A which are involved in the swapping of the Nuclear Service Water System from its normal intake source, Lake ,
Wylie, to its assured source, the Standby Nuclear Service Water Pond .
(SNSWP), during a low RN pumphouse pit level or high containment pressure signal i
. MCC 1EMXO,'which provides power to several other RN valves involved in the swap fromlake to pond 1
Note that all ot' these components are recuverable. l 1
The RMQS file (referenced in Section 10.1 above) was checked for the contribution to core melt of breaker F03A. This component (as well as other
- components associated with MCC IEMXG) was not in the RMQS file. Likewise, RMQS was inspected for the contribution of the RN valves and none were found.
Rus, given that these components are not included in the high CDF contribution cut sets, and given that a loss of the affected components is recoverable, it may be l' assumed that the" loss of IEMXG poses noisignificant' impact on the, core _damagey probability.
1 l
, l80-14-1994 10hR1 FROri DPC NU("LEfR ENGINEERING TO 88333400 P.09 CNC-1535.00-00 0007 By _ Date f' pg. 4 of 7 To verify this assumption, a review of the ' Essential Auxiliary Power System fault tree, as found in Catawba PRA Appendix A.16, was performed assuming coordinated conditions. Using the basic event probabilities found therein, the failure probability of 1EMXG will bg the following:
ITEM DESCRIPTION PROBABILITY P1ELXA loss of power on 600V ac Icad center IELXA 1.lE-04 PACEMXGCLT breaker IELXA-5B transfers position 1.2E-05 PACGF3ACLT breaker 1EMXG F03A transfers position 1.2E-05 -
4.8E-06 l PACEMXGBLP bus fault on IEMXG PACEMXGTRM unscheduled naaintenance on 1EMXG 3.9E-07 1AE-04 For an uncoordinared condition, the failure probability of IEMXO will thus be the above probability plus the sum of the faults to AHU #1 which will open breaker F03A before opening breaker F05C. To find an appropriate value for the frequency of qualifying AHU #1 faults, the NPRDS file was searched for industry fan faults which could have produced fault currents much :Jgher than the individual load breaker setting. Out of 235 fan failure events reported, only four are envisioned to have involved large instantaneous overcurrent or catastrophic occurrences which are necessary to trip breaker F03A before breaker F05C. The NPRDS file contains 1304 similar fan loads which have been in service approximately 100,000,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> over the past ten years. The failure rate is then about 4E-08/hr or approximately lE-06 for a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> raission time, or roughly two orders of magnitude lower than the coordinated failure probability for 1EMXG.
. Smitivity Analysis - If we increase the AHU #1 breaker fault probak / by an order of magnitude, the resulting failure probability for IEMXG v41! be 1.4E-04 + IE 05 = LSE-04; or an increase of about 2dE Therefore, since the AHU #1 breaker failure probability is relatively small (even when increased by an order of magnitude), it is concluded that the uncoorrW=ted situation for the EPE Systern hat a neglicible imoact on the PR A and does not g[fget the overall Catawba CRE.
11.0 Conclusions h 11.1 Based upon a review of the 125V de Vital I & C System and the Catawba PRA
[ CDF high contribution cutsets, it is concluded that an improvement of the load group distributien centers' breaker coordination will have a negligible effect on the risk of core damage.
11.2 When u: coordinated conditions are assumed, the failure probability for MCC IEMXO (or 2EMXH) is approximately the same as when coordinated conditions
J 10-16-1994 10252AM . FROM 'DPC NUC1. EAR ENGINEER 1NG TO 88313400 P.10 CNC-1535.00-00-0007 By N '
Date /*//=/94
'/ ' pg. T of L ;
are assumed (even when increased by an order of ma;nitude). Therefore, based upon this information and considering that the components affected by the loss of -
1EMXG are not included in the high contribution CDF cut sets, the uncoordinated situation for the EPE System has a negligible impact on the PRA and does not affect the overall Catawba CDF. .
h i
i e
f 4 L
i
._m . _ _ _ _ _ _ _ _ _ _ _ _ . _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ . _ _
10-14-1994 10:52AM FROM DPC NUCLEAR ENGINEERING TO 88313400 P.11 ATTACleElfr 1. FRA For EPL and EPE Systems
,. - [%.
E UNITED STATES
. NUCLEAR REGULATORY COMMLSSION ,
wameworon. o.e. -
September 14, 1994
~
Mr. O. L Rehe Vica Presidaat - Catawba Site Duke reser Campany 4800 Concord Road '
York. SC 29745 .
suaJET: RE0ust FOR morTIons. nrosurica on EECTalcu. Bassa coomtmarlos, cATman nuctua sTarton, alt uns.1 AM 1, (TAC us. nesss7, messa) -
Dear Mr.'Asha:
The Electrical .Distriheties systes Functional Inspection (IDEFI)' performed at the Catauba Neelear Station Units 1 and I from Jammary 13 to Fahreary 14, 1992, identified a safety-significast deviation fres a writtes causttamat as follows: -
- '9E3 5-0800, standard Review Plas, states at page 8.3.1-5 thatF*.- t
, acceptance 'of a desigs] is based on meeting the spesific guideltaes fa nasalasary deida 1.32, etch endorses the lassitate of Electrical and' Electreates Engineers (IIIE) standard act. IEE Standard ass-1974,
' standard Critarfa for Class IE Peuer Systems W Nuclear Feuer Camerating statteas,* states is sostias 5.3.1 tfat protective deviens shall be provided ts Itait the degradstian of'C0 ass IE peuer systems.
l as page 8 75 states The that theliennene's system umsts FinaltheSafety reestrements Analysis of thisReport stan (F54R),dard. The PSIR -
Secttaa W.3.1.1.1.1 states in part, that protective devices os the 600-VAC Essastial Aus111ary Peuer syste are set' to achieve a salactive tripplag nehamn se that a slainst amount, of equismust is isolated for as adverse condittee sask as a fault.....' ,
The NRC staff concluded that the Duka Power Campany (DPC or licensee) had deviated f'res this commitammt is that, the incoming breakars te all the essential set voit alternating currest (V ac) antar-castrel ce= tars (IECs) are not esordinated with the ostgoing breakers free the IECs. Perther, tha 125 voit direct current (V de) wital tastrumentaties and central poser unide6-case breakers in the distributies centers are est caerdiasted for all faults.
In as attempt te determine the impact se" plant risk of thesa breakers that both the staff and DPC igreed unre uncoordinated, and ta get furt: hor assurance that the e of the lack of caerdinaties were insignificant, the IK .
staff, en Decembsr s.1953, sent a rupsest for additional inferustian (RAI) to
. DPC asking for (1) the locations of fastts of any kis1 that canid feed to
~ uncoordtaktad breakers in the 125-V de vital instrumastatism and contret poser ststas and ta the 406-V as assential aurtliary system.,(t) the idsstity or the breakers, (3) the leads served, and (4) ~ cassequestas of losing the safety leads affected. '
$ K W As r ii i V.) .) 't -
f,)j
10-14-1994 10:53An FROM DPC NUCLEAR ENGINEERING TO 88313400 P.12 vei as, e. -. .a - * -
g,4 4 Wff l "hYS+-
D. Raha September 14, 1994 j OpC responded.to the staff on February 7,1994 and on March 2.1994 des l copias of breaker coordination curves and system eme-line drawings shoutag locations for the inest probable worst-case fanits, the associated fault currants, and the breakers that wesid not coordinate la case of de double-11ae or three-phase faults were sabaitted. The licenses aise sabettted the '
calculations os dick the fault currents were based and a list of loads that could not be pouared if seen breakars were not coordinated. l 1
The NRC staff has deterstead that it caanst accept OpC's proposal ta change the F5AR causitmaats rather. than the breakers themselves solely on-the basis of the taformation saksitted up to this timer. The preessal shamid ha accompaaled by an appropeista evaluation sade persumst to the requiremmets of le CFR 50.59, "Chamens, Tests and Experissets'. Further the proposal is met .
consistant with tha following NRC staff guidanca, as stah in Generic Letter i 85-15. *Efestric Power Systems - Inadequate Castrol Over Design Processes:' '
~
- Lack of breakar caerdinatias can creata the potest'tal for ur !
, unacceptable level of r,quipmast less during fanit conditions. 1 Thus, the designs of these electrical systems were.aot fully te I confomance with CDC-17.' .
5
' In Informaties Hetics No. 38-45, ' Problems in protective Rel'ay and circutt i armaker Coordination, the staff's position is steiad as ;
"This infomattaa astice is belas provided to alert addressoas to a potaatially significant problem conceasing the possible lack of protective relay and circa,it breaker coordtaation."
Dpc has indicated that it caanst easily restare kroskar coordination by
' *racklag im* Peplacement breakars staca there are as appropriate replacamnet breakers available that will coordinata and fit la the esisting embicles.
In e y, the NRC staff requests that DFC provide a further submittal in accordrace with the provisions of le CFR 54.$3, that justifies a propesa,d amendmaat to the FIAR is lies of replacias the breakars. This semittal sa reference er faslude relevast infonmatten 21rendr saksitted on this.insaa. yIf probabilistic risk ========t (fRA) techniques are used te quantify the risk of Taavtag the breakers.uassordtaated, the description of the increase is risk should be sufficiently datalled and ripous ta show that the failure'to coordinate musld met unacceptably tacrease the magnitude of a risk snesere such as the care dange accidnat frequency. If available, Orc should insinde taforestion about the f)equency of cabis and IEC failures so that the staff can daterwise the frequency of the initiating events and the russitant caange i
to the core-damage Gwy. If practical, DFC should aise sher by a
' sensitivity analysis, that order.ofwssonitude changes in the assum,ed paramatar values used in the Phk would not invaltdata the coesiastens of the PRA.
e 6 9%
e 1
___.___l
1C-14-1994 10:53AN FROM DPC NUCLEAR ENGIEERING TO 88313400 P.13 ev w as ...** .--. - -
349 W/W
~
- 0. Rahn ,
September 14, t994 i
DPC's sutaitta'l should also 1* scribe the extent of plant modiffcations that would be necessary in order k restore breaker coordination, including the l number and type of.-breakars at d cubicles to be replaced and the extent to l which cables would need to be redrawn to avoid splices.
To enable the continuance of our reytew of this matter, we request that DPC provfde its response to this letter at the earliest practical date. This reqeirament affects. fewer than ten nspondents, and therefore, it is not i subject to Office of Management and Budget review under,P.L.96-511. '
Sincerely, '
ff -
Robert E. Martin, Senior Pr.oject Manager Project Directorate II Division of Reactor Projects - I/II .
Office of Nuclear Ratctor Regulation *
. 4 Deckat Mos. 50-413 and 50-414
.cc : See next page -
i e
G 4
$ , O a
4 e
9%
p ,.
~
. ~.
~
l .
a e
s -
i 10-14-1994 1N54AM FROM DPC NUCLEAR Em NEER NG TO 88313400 P.14
' ggg.
- g. , .
{' .
Mr. David L. Raha ' /*//-l9l-Duke Pcasar Company Catsuba Nuclear Statica ec: -
Mr. Z. L. Taylor Hr. Marvin stakule, Chief' latory Como11ance stanager Project Branch #3 Power Campany . U. 5. Nuclear Regulatory Constission 4800 Concord Road 101 Marietta Street, lef. suite 2900 Yor*, South. Carolina 2W45 Atlanta, Georgia 30323 .
A. V. Carr, Esquire , North Carolina Dectric Membership !
Duka Power Corporation 422 south church treet P. D. sex 27306 ,
Charlotte, North Carolina 28242-0001 ,
Raleigh, North Carolina 27611 .
.J. Michaal McGarry, III,' Esquire *
. Seator Resident Insie/wr Winston and straun Resta 2, eos 179 N 1400 L Street, at . York, South Carolina 29745 Vashington,DC, 20005 Regional Adslett,trator, Regica 11 North.Carollaa Municipal Power U. 5. Natisar Regulatory Caumissism
. Agercy Number 1 101 Marietta street, lei. Suite 2900 1427 Maadeuwood Boulevard Atlanta, Georgia 16333 i P. D.*Bax 25513 Raleigh, North Carolina 27626-4513 Max Batwia, Cktef 5 sureau of Radiological Hamith '
Mr. T. Richard Purytar south Carollaa Departamat of Meclear Technical Services Manager Haalth and Environmental Centrol Westinghause Doctric C3cporation 2600 8s11 street Carolinas District Columbia, South Carolina 29201 2709 Mater Ridge Parkuay, Sutta 430 Charlotte, Nepth Careltaa tatly . Nr. G: A. :
. Licensing -
County Manager of York County Duka Peuer Company York Courty Courthemsa .P. 0. 8sa 1006 York, South Carolina 29745 Charlotte, North Carolina 23201-1006 1
, ~R ichani P. W11sen, Esquire Saluda River Dectric ;
' Assistaat Attorney 9eneral -
P. O. Bat $29 l South Carolina Attomey General's Laurens, South Carolina 29360 l Offics .
P. D. Sex 11549 .
Ms. Karma E. Long Columbia, South Caroling 29211 Assistaat Attorney General .
i . North Carolina Department of Justice Piedmont pharicipal Power Agency P.'O. Bas 629 111 Village Drive . . Raleigh, Marth Carlina 17602 -
l Erwer, south Caroli 6a 2965: -
Oalaa Wathna, Lead RIP Planner-
. Dayne H. Broma, Director Division of Easrgency Management
, Division of. Radistica Protection 116 West James Street 4
N.C. Depatement of Environment, - Ra1.eigh, North Carolina 27601-1335 Haalth and Natural Researces ~ '
P. O. Box E7547
. Raleigh, North Carolina 27611-7687 -
4
'1'0-14-1994 10:54fr1 FROM DPC NUCLEAR ENG!tEERING TO 88313400 P.15 b
'. ATTACIBG!Nr 2. FRA for EPL and EPE Systems QI ~
July 7,1994 m//#
,/,fg.
To: Paul Farish From: Allen Dickard
Subject:
Catawba breaker coordination, systems EPL and EPE
~
Attached is information from a package we sent the NRC 3/2/94 as a followup to a meeting we had with them in Washington in February. One of the questions they asked us was "what breakers are coordinated and not coordinated?". The attachment contains our response. As we discussed on the phone we do not yet know if we have reached any agreement with them on where we shouldpostulate credible faults, which impacts how much coordination we have. I will let you know as soon as we receive further information from them. You can contact me on any questions related to the EPE Jystem et 831-4086, RAD 83S2 or Jim Glasser for anything related to EPL system at 831-4068, JDG7537. Thanks.
s 9
4 4
e 4
. .~1e-14-1994 te:55AM- FPIPl DPC NUCLEAR ENGINEERING TO 88313400 P.16 >
- 4. - .
,'j . Responsa to Pcrt 1 cf Ettetrical Distributien Systrm Functi:nni- gg ,
Inspection (EDSFI) Deviati:n DEV 50-413,414/92-01-02 gg ,
125VDC Vital I and C Power (EPL) System Breaker coordination Study .
L Summary A. Where are our faults?
. v See Table 1.1 and attachment 1 for the postulated fault locations. Since the ~
125VDC Vital I and C Power (EPL) System is an ungrounded system, the postulated faults require that either a simultaneous positive-to-ground and j negative-to-ground fault occur or a double line (positive-to-negative-line) fault .
occur.
ne former type of falt requires that two failures occur which is beyond the requirements of the single failure criterion and the design basis of the plant.
SI ould a single line-to-ground fault occur, the power system will remain unaffe' cted; however, a ground fault detector will alert the control room operator via both an annunciator and computer alarm. Any ground faults are promptly addressed through a program which seeks to maintain a dark control room annunciator board. Control room annunciator status is reviewed on a weekly basis. A sample status report is shown in attachment 5. ,
The latter type of fault is extremely unlikely. To illustrate this, a thorough study of both the Nuclear Plant Reliability Database System (NPRDS) for all U.S.
nuclear plants and the Catawba Probabilistic Risk Assessment (PRA) was
. conducted. I'ailure disenveties made at Catawba since 1985 and all U.S. plants since 1990 were analyzed. Only three reported cases were found where a double line fault occurred on a DC system. One case which occurred at Catawba :
involved a shorted lamp holder.which was attributed to improper iw=Hadon during previous maintenance activities. The two other cases which occurred at ;
nuclear plants operated by other utilities involved component failures within , l battery chargers. In both cases, plant status was reported to not have been i affected. No reported cases were found that involved double.line faults attributed ,
to cable failure:;. ne risk of cable failu es is further reduced at Catawba since 2KV-rated, interlocked armor cable is used throughout the plant. Finally, engineering personnel at other utilities were eqntacted regarding their experiences
. 'l with faults on safety-related DC pwer systems. No faults of the types that could ;
challenge the EPL System were reported.
B. What breakers are coordinated and not coordinated?
Attachments 3 and 6 contain the fault current calculations. The worst-case fault !
currents and resultant breaker currents are plotted against the breaker time !
cu Tent characteristic curves in attachment 4. Inspection of attachment 4 yields ,
the follow'mg conclusions concerning coordination:
Page l'of 9
' 19-14-1994 10:55AM
~
FROM DPC NUCLEAR ENGIEERING TO 88313400 P.17 L.
"~',
Respsusa t's Part 1 ef Electrical Distributi:n System Functien:1. .a W@; ;
Inspection (EDSFI) Deviation DEV 50-413,414/92-01-02 , , i./,./w 125VDC Vital I and C Power (EPL) System Breaker Coordination Study jFault "BC" (Reference Pnee 1 ofAttachment 41 No coordination. The fault current falls within the magnetic regions of both the ' !
125VDC Vital I and C Battery Charger breaker (cornpartment F03 A) and the 125VDC Vital I and C Battery breaker (compartment F02A',. ,
Fault "AD" (hference Pawe 2 of Attmehment 4)
Fully coordinated. A fault at the. input terminals of the Auctioneering Diode j p Assembly would trip the Auctioneering Diode Assembly feeder breaker j
, (compartment F01D) without operating either the 125VDC Vital l and C Battery breaker (compartment F02A) or the 125VDC Distribution Center incoming breaker (compartment F02B). The 125VDC Vital I and C Battery Cha:ger breaker (cosiiphnent F03 A) would not operate since the battery d-i.e4 is ,
current limited to a value less than the continuous rating of the breaker.
Fault "EDE" (Reference Pnee 3 ef Attachment di Fully coordinated. A fault on the 125VDC Auctioneered Distribution Center bus would trip either the Auctioneering Diode Assembly feeder breaker (compartment F01D) or the 125VDC Auctioneered Distribution Center incoming bmaker (compartment Fot A) without operating either the 125VDC Vital I and C -
Battery breaker (compartment F02A) or the 125VDC Distribution Center !
incoming breaker (comphoer.t F02B). The 125VDC Vitall and C Battery
, Charger breaker (compartment F03 A) would not operate since the baucry -
char 6er is current limited to a value less thanthe continuous rating of the breaker. ;
Fault "EDE1" (Reference Pace 4 of Attachment 4) y/ Partially coordinated whether the fault occurs at the load-side terminal ofthe breaker or at the first termination point of the feeder cable. The fault would trip ,
either the 4.16KV Essentir.1 Switchgear control power feeder breaker -
i (co~mpartment F01C), the Auctioneering Diode Assembly feeder breaker I (compartment F01D), or the 125VDC Auctioneered Distribution Center incoming breaker (compartment FOI A) without operating either the 125VDC Vital I and C . .
Battery breaker (compartment F02A) or the 125VDC Distribution Center !
incoming breaker (compartment F02B). The 125VDC Vital I and C Battery Cha ger breaker (compartment F03A) would not operate since the battery )
charger is current limited to a value less than the continuous rating of the breaker.
Full coordination is possible depending on the location of the fault. ,
I I
Pag- 2 of 9 .
I F c w ----e, a -
-3
.<- . 10-14-1994 10:56AM. FROM DPC NUCLEAR ENGIbEERING -
TO 88313400 -P.18
. .. .v Respsnst to Pcrt 1 of Electricel Distributien System Functirnet.
Inspectisn (EDSFI) Deviatisu DEV 50-413,414/92-01-02 g/[
,./,./q.
125VDC Vital I and C Power (EPL) System Breaker Coordint LStudy Fault "EDE2" (Reference Pawe 5 of Attachment 4)
Partially coordinated whether the fault occurs at the load-side terminal of the breaker or at the first termination point of the feeder cable. The fault would trip either the 600VAC Essential Load Center control power feeder breaker . l (compartment F01E), the Auctioneering Diode Assembly feeder breaker (compartment FotD), or the 125VDC Auctioneered Distribution Center incoming !
breaker (compartment Fot A) without operating either the 125VDC Vital I and C -
Battery breaker (compartment F02A) or the 125VDC Distribution Center incoming breaker (compartment F02B). The 125VDC Vital I and C Battery Charger breaker (compartment F03 A) would not operate since the battery charger is current limited to a value less than the continuous rating of the breaker. -
Full coordination is possible depending on the location of the fault. -
Fault "EDE3" (Reference Pare 6 of Attachment di e Partially coordinated whether the f4 ult occurs at the load-side terminal of the breaker or at the first termination point of the feeder cable. The fault would trip either the Diesel Generator Load Sequencer feeder breaker (compartment F01F),
the Auctioneering Diode Assembly feeder breaker (compartment F01D), or the 125VDC Auctioneered Distribution Center incoming breaker (compartment '
Fo1 A) without operating either the 125VDC VitalI and C Battery breaker -
, (compartment F02A) or the 125VDC Distribution Center incoming breaker (compartment F02B). The 125VDC VitalI and C Battery Charger breaker ,
(compartment F03 A) would not operate since the battery chargeris current limited to a value less than the. continuous radng of the b reaker. Full coordination is possible depending on the location of the fault.
Fault "VI" (Reference Pave 7 of Attnehment 4)
No coordination. The fault current falls within the magnetic regions of the 120V4 C Vital I and C 1nverter breaku (compartment F03C), the 125VDC Distribution Center incoming breaker (compartment F02B), and the 125VDC Vital I and C Battery breaker (compartment F02A). -
t Fault " EPA" (Reference Page 8 of Affachmettr 4) '
No coordination. The fault current falls within the magnetic regions of the ,
- 125VDC Distribution Center breaker (compartment F01C) feeding the 125VDC l Vital I and C Power Panciboard, the 125VDC Distribution Center incoming
! . . breaker (compartmerc 702B), and the 125VDC Vital I r.nd C Battery breaker (compartment F02A).
l \
Page 3 of 9 .
l '.
4 10-14-1994 10:57Af'F FROl1 DPb NJCLEAR ENGDEERING TO 88313400' P.19
- ' ' I'
' nesp nse to Part 1 ef Electrical Distributisu Systcm Functienni-N Inspecti:n (EDSFI) Dsvistien DEV 50-413,414/92-01-02 *M
/.4./y 125VDC Vital I and C Power (EPL) System Breaker Coordination Study Eapit " EPA 4" (Reference Pave 9 of Attachment 4) '
Partially Coordinated for a fault that occurs at the first termination point of the feeder cable. The fault would trip either 125VDC Vital I and C Power Panelboard breaker #4 (which feeds a Component Cooling System solenoid valve) or the 125VDC Distribution Center breaker (compartment Fo1C) feeding the 125VDC Vital I and C Power Panelboard without operating either the 125VDC Vitall and C Battery breaker (compartment F02A) or the 125VDC Distribution Center incoming breaker (F02B). The 125VDC VitalI and C Battery Charger breaker (compartment P03 A) would not operate since the battery charger is current limited to a value less than the continuous rating of the breaker. Full coordination is possible depending on the location of the fault. For faults where the resultant current flowing through the 125VDC Distribution Center breaker (compartment '
Fo1C) feeding the 125VDC Vital I and C Power Panelboard is less than approxirnately 480ADC Sill coordination would be achieved. From Table IJof >
attachment 6, the load current value for 2 EPA is 65.31 ADC. Therefore, from page 29 of attachment 3, the fault current (Igpy) must be less than: .
hm s 480-65.31 5 414.69ADC for full coordination to be achieved. Refer to attachment 6 for more information.
No coordination for a fault that occurs at the load-side terminal of the breaker.
The fault current falls within the mqnsic regions of 125VDC Vital I and C Power Panelboard break er #4,.the 125VDC Distribution Center breaker (compartment F01C) feeding the 125VDC Vital I and C Power Panelboud, the !
125VDC Distribution Center incoming breaker (compartment F02B), and the 125VDC Vital I and C Battery breaker (compartment F02A). '
C. What is the impact of the upstream breaker clearing? '
Should an upstream breaker such as the 125VDC Distribution Centerincoming breaker (compartment F02B) operate prior to the breaker nearest the fault clearing the fault, one load group of the 125VDC Vital I and C Power (EPL) ,
System wou'A be lost; however, the redundant train ~of power would be available '
to supply safe y-related instrumentation and control loads.
i D. What is the safety significance of taking out a train?
Though the following study demonstrates that full circuit breaker coordination l will not be achieved for all postulated faults, there is no safety si[nificance to !
Catawba Nuclear Station nor the public because a completely independent and l
redundant 125VDC Vital I and C Power train would be available to supply safety-Page 4 of 9 .
- . 1@-1 -1 10:57AM FROM DPC NUCLEAR ENGINEERING TO 88313400 P.20 eAar 2 +/*
Catawba Nuclear Station -
600 VAC Essential Auxiliary Power' System (EPE) ,
I.- Summary 1.- Where are our faults?
The attached one liile diagrams of the EPE system show the fault locations. Included are drawings of Unit 1 train A and B motor control centers, the Unit 2 motor control centers are similar.
Based on the low probability of bus faults and/or breaker faults at '
Catawba and our experience that we have never had such a fault on the EPE system at Catawba we do not postulate a fault on the output of the feeder breaker. Also, because of the interlocked armor cable protection and our experience that we have never had a fault on any such cable in service at any of our nuclear plants we do not postulate a fault along the route of the cable. We do postulate a fault at the input terminals 6f the load or at the first cable ,
termination after the cable leaves the motor. control center. The loads on each motor control center were reviewed and the load with the highest potential fault current was selected for modeling coordination of breakers for that bus. Attachment D on fault ,
p locations for the EPE system provides further justification'for selection history andofprobability.
these fault locations with information on fault 2.
What breakers. are coordinated and not coordinated? '
The incoming br'eakers for motor con [rol centers 1EMXA, 1EMXB, 1EMXC, 1EMXD, 1EMXE, 1EMXF, 1EMXI, 1EMXJ,'1EMXK and 1EMXL are coordinated for the worst case postulated fault at the first cable termination outside the motor control center. Motor control center 1EMXG utilizes two incoming breakers because it can be powered from either Unit 1 load center 1ELXA or Unit 2 load center 2ELXA. These incoming breakers are not fully coordinated for a fault at the worst handlingcase unitload
- 1. which is control room ventilation system air is 100 feet long. Th.is unit is connected with a 250MCM cable that All other loads on motor control center 1EMXG are fed from current smaller for which the breakers incomingand cablevalitith breakers lower maximum fault e coordinated. )
1
{
9 4
9 I
l 1
' ~
N P f d F.P S st' ens Motor Control Center 1DfXG Loads
..p.w*
V ,
Breaker Load HFB100 Notor control center 1RMKO NFB40 . Power panelboard transfomer 1BK'IG feeder NFB20 A header return to SNSWP valve 1RN63A HFB20 Station RN header disch to RL system valve 1RN57A HFB20 Station RN disch header crossover valve 1RNS4A j ABF". mini flow outlet valve 1VA29A KFB20 NFB20 ABFU-1B niini flow inlet valve 1VA31A HFB20 ABFU mini flow outlet valve 2VA29A HFB20 ABFU 25 mini flow inlet valve 2VA31A HFBGO Pump room heater desister section PREDS-1A RFB60 Pung room heater demister section PRNDs 2A -
MFM150 ** Auxiliary building filtered exhaust fan motor ABFXF 1A ;
HFB150 Auxiliary building filtered exhaust fan motor ABFIF-2A l RFB100 Control room air handling unit 1 -
HPB50 Control room area PFT-1 moisture separator heater LB225
- Control room area air handling unit 1 ,
HFB20 Battery room exhaust fan motor i '
HFB20 Pressure fans xcenn valve.2VCSA EFB50 Control room area'f11ter train 1 pressure fan motor -
RFB20 Channel A outside intake isolation valve 1VC6A RFB20 Channel A.outside intake isloation valve 2VCGA EFB50 AB groundwater drainage interior aump pump motor C1 RFB50 AB groundwater drainage sump pump motorLAl y 11tFB50 AB groundwater drainage sump pump sotor 31 EFB30 Control room area, chiller coup A cil punqp motor HPS20 Ctri area chilled wtr train A nakeup valve 1YC77A RFB100 Control room area chilled wat,ar pump motor A
- This feeder has ttu highest potential fault current for a fault at the load of all feeders on this motor control centw..
- This feeder has the second highest potential fault current for a fault at the load of all feeders on this motor control center.
4
?
l l
l l
U
O O
grrACIIMENT H
l
. 1 i
U.S. Nuclear Regulatory Commission December 29,1994 ATTACWIDrr "B" page1of2 Frequency of Catawba Failures:
EPL System 125 VDC Distribution Centers All PIP's (Problem Investigation Process Reports) on this system were reviewed and there were three cases of problems internal to battery chargers that blew charger fuses but no faults that tripped breakers or challenged breaker coordination.
EFE System 600 VAC Motor Control Centers ;
All PIP's on this system were reviewed and there were no cases of faults that tripped breakers or challenged breaker coordination.
CAhld Any plant problems with cables are reported under the plant system that the cable is associated with. All of the PIP's referred to above for the EPL and EPE systems were reviewed for any problems with cable degradation or cable faults and none were found.
Industry Failure Data:
An NPRDS search ofindustry failures of electrical conductors, buses, cables and wire from 1/84 to 11/89 (reporting criteria changed in 89) was conducted. Of the 267 failures in this report there was one fault of sufficient magnitude to cause a miscoordination event. l This occurred when technicians at Oyster Creek left a safety grounding jumper on the load side of a breaker after a maintenance activity. When the breaker was closed the short i circuit tripped the main bus breaker also. The 1987 McGuire event referred to in IN 88-45 was not included in this data because it involved a fault in a non safety instrument air .
compressor motor. Failures in this system are not typically included in the scope of ;
l i
l I
l J
r e
e l
I l
Page 2 of 2 ATTACHMENT "B,, )
Extent of Plant Modifications:
EPL System Coordination of protective devices in this system can be enhanced by replacing the battery l breaker and main breaker in each de distribution center with breakers capable of an instantaneous trip setting of 10,000 amps or higher. For Units 1 and 2 there are eight distribution centers and this would involve replacement of sixteen breakers. There is very limited expansion space available within these distribdon centers and it is likely that the new breakers would have to be mounted in new, stand-alone enclosures. The power cables that terminate to these sixteen breakers would have to replaced (250 MCM cable, 700 feet total length) and additional cables would have to be installed between the distribution centers and new enclosures. Although the eight distribution center tie breakers are not normally in senice, additional coordination could be achieved by replacing these breakers with some of the original breakers removed from the battery and main breaker compartments, in the event the battery and main breakers are replaced.
EPE System Coordination between the motor control center incoming breakers and individual load feeder breakers can be enhanced u; replacing the incoming breakers with breakus capable of an instantaneous trip settmg of 12,000 amps or higher. For Units 1 and 2 there are 30 motor control centers with a total of 34 incoming breakers. In this equipment there is also very limited expansion space available and it is likely that new breakers would have to be mounted in new, stand-alone enclosures. The power cables that termirate to these 34 breakers would have to be replaced (12,000 feet of 500 MCM cable and 26,000 feet of
- 2 cable) and additional cables would have to be installed between the motor control centers and new enclosures.
Another option for enhancing coordination in this system is removal of the incoming breakers and re-terminating the cable directly to the bus. Most of the incoming cables will not reach the bus because they enter the motor control center from the top where the incoming breaker compartment is located, whereas the main bus is in the center of the equipment. This option might require re-pulling as much as 30,000 feet of power cable.
The four motor control centers that each have two incoming breakers could not be modified in this manner because the incoming breakers are mechanically interlocked in each motor control center to provide separation between Unit I and Unit 2 circuits.