ML19254E091

NRC Testimony Responding to Questions A1,A2,B1-B4,C & D of ALAB-537 on 790405.Provides Background Info for Grid Description & General Onsite & Offsite Power Sys.Supporting Documentation Encl

ML19254E091
Person / Time
Site:	Saint Lucie
Issue date:	10/29/1979
From:	Fitzpatrick R Office of Nuclear Reactor Regulation
To:
Shared Package
ML17266A109	List: ML17266A108 ML19254E087 ML19254E091 ML19254E093
References
NUDOCS 7910310082
Download: ML19254E091 (31)

Text

1 s

.c U:lITED STATES OF A" ERICA I:UCLEAR P.EGULATCRY CCa'"ISSIO:1 3EFORE T!:E l.TC'iIC Si1FETY l':2 LICE"SI'"I f ?"EAL ECARD In the Matter of

)

)

FLORIDA POWER & LIGHT COMPANY

)

Decket No. 50-3S9

)

(St. Lucie fluclear Power Plant,

)

Unit 2)

)

AFFIDA"IT OF RCSERT G. FITZPATRICK I, Robert G. Fitzpatrick, being duly swcra to depose and state:

1.

I am an electrical engineer and a senior member of the Power Systems Branch of the i;uclear Regulatory Commission.

2.

I have prepared a statement of professional qualifications which is attached to this affidavit.

s.

3.

I have pr$ pared the testimony attached to this. affidavit which addresses questions A1, A2, 81, 82, B3, 84, C and D as stated in ALA3-537.

I hereby certify t. Hat the above statements are true and correct to the best of my knowledge and belief.

i w pl d

D h,- m

!u_ - /

'W j-Robert G. Fitzpatrick //,%

Ssscriled and sworn to before me this., 7 day of Septetcer,1979 Ay Corrission Expires; S

t Noury fulic 7910310 1-r 1 )' r 3

I

i

=

i i,... T _, c...

e. n. e

..u v.u fiUCLEAR REGULATORY CC'.MISSIO:1 BEFORE lHE ATCMIC SAPEiY A';0 LICE:iSI':G AP.'C;L ECA23 In the Matter of

)

)

FLORIDA P0.!ER & LIGHT CCMPANY

)

Docket No. 50-389

)

(St. Lucie Nuclear Power Plant,

)

Unit 2)

)

NRC STAFF TESTI!:CNY OF ROBERT G. FITZPATRICK C

5 13%

I.

INTROD'JCTICS This testinony responds to questions Ai, A2, v1, 32, 33, 04, C cnd D as set forth in ALAB-537 dated April 5,1979.

Part two of ny testimony sets forth some background material to provide the Appeal Board a description of the Staff's view of the FP&L grid in general and the onsite and offsite power systems at St. Lucie in particular.

Part three addresses the specific ALAS-537 questions and includes Staff ccmments cn the Applicant's res:cnses.

Part four is a surrary of this testimony.

II. BACKGROUND Two documents have been previously submitted to the' Appeal Board whicn sumrarize the FP&L grid system and the St. Lucie power systems: joint testimony orepared by M. Srinivasan and D. M:Conald entitled "A Further

, Evaluation of the Florida Power and Light Ccmpany Electric Power System",

October 25, 1977 and my testimony of June 12, 1978. The f;rst document provides background material en power systems *n general, NRC recuirements 5

140

-m

f 2-7 concerning elet'.rital pc..er syste;r.3, a surcr.:ary of the major FP&L gric systen disturbances frca 1973 to 1977, and tre action the staff has taken in reccor.so to these major disturbances.

My testimony prepared in respense to the "speal Scard's March 10, 1973 ordar urdatos the former document with respect to the May 14, 1973 event and subsequent staff actions.

It also provides a discussion of NP.C requirements cencerning electrical power systems and documentation of additional staff positions imposed upon FP&L as a result of the engoing staff re-vicw in t.ke electrical pcwer systems area.

These two documents, when taken together, provide the background material upon which ny response to ALAB-537 has been constructed.

III. ALAB-537 CUESTIONS A.

Ger.eral Design Criterien (GN) 17b A1.

This critericn, entitled " Electric Pcwer Systems," requires in its third paragrapn (amphasis added):

Electric pcwer fr:m the transmissicn netucrk to the cosite electric distribution system shall be supplied by two physically :nd: pendent circuits (not necessarily on separate rights Of way) designed and located so as to minimize to the extent cractical the likeli coa of tneir simultaneous fatiure uncer operating and prtulated acc1 cent and environmental conditions. A switchyard com. ion to both circuits is acceptable.

All three transmission lines connecting the St. Lucie station to the applicant's grid originate at the Midaay substation.

The May 14. 1 973 incident, in which all pow r at that substation was lost despite redundr incoming sources, demonstrates that thesc circuits are indeed susceptib

to simultane:us failure.

The testi m y sh uid a? ress whether tne St. Lucie station ncnetheless m:.ets this GCC-17 req;1rement.

.,ee is

. L.

for Nuclear Power Plants").

5 m P00R ORIGNAL

~

"~

=w.

o SitJF A:.S'<4P TO 30?:D O'.TSTICi! Al The St. Lucie station is in full confori ance with the recuirs.ents cf GDC-17 including tha specific subsecticn addressed in this ques *. ion.

From the sta ff's point o f vie.9, ncne of cne three St. Lucie to ".idaay lines failea darir.g the My 14,1973 incident.

The May 14, 1978 inc; cant ws a lo:s of offsiM power to the St. Lucie station as was the May 16, 1977 incident.

I make a distinction between grid unavailability and circuit f:1!ure Mcause it is fundamental to the staff's interpretation of GCC-17 Grid unavailability (i.e., loss of offsite power) is recognized as an anticipated operational cccurrence. That is, it is an event that is-expected to occur one or more times during the life of the nuclear power station.

In this light, the regulations do not require a design which precludes the event but rather the capability o f the integrated nuclear power station design to ccpe with the event if and when it should occur. The St. Lucie to. Midway lines a

'were not tripped a: pert of the sequence of events that le'ad to the isolation 'of the Midway Substaticn and were intact and available to supply St. Lucie with offsite power had it been available. By the staff's interpretation of GCC-17, there have been no sir.ultaneous circuit failures on the St. Lucie to Midway transmissicn lines.

General Cesign Criterion 17 is specific in the requirements placed upon the physical configuration of the offsite power system in the cicsc proximity o f the nuclear generating unit. The Appeal Coard is correct that the "c =0n switchyard" referred to is the one electrically connected S

142

-w

-=w..%.e-,-..,~

. w - w m,

,,o,_-

di. uct:y tc t h unit c. > e." h i a n d o r. '; i t e d i : / i u

• ini eys':m.

U ust ca: 23 th i r.. ! '.ci j: rd i: 10 :2'. : ' c c '"?0 4-^

H a"0*-

!"ir 1: nct an MC recuire-

'mant. A minicun o f two tr:%c issica li".:.; rust 00"ect this s s itchyard to

=-

t ofisite power system and a minimua of t o line n.st cocoect this switcn-yard to the onsite distribution system.

This is the extent of the physical complement of equipment required by GDC-17 for the o #fsite povier system.

Figure Al (UlPA) depicts, in block diacr:m form, this mininum physical cis'plement of alectrical ;;wer system ecui;rert requ'nd to meet the require -

h ments ci GCC-17 The additic".21 ;-hysical requirc sr: placed u;an this y,,

complement of electrical pcwer systen equips,ent is taat the circuits ce

" located so as to minimize to the extent practical the likelihood of their

[~

[idditaneous failure." Staf f review of this latter aspcct of the design (i.e location) is described in Standard Review Plan 8,2 anm those specific subsections which address this subject are attached as Enclosure 2.

fl_

l 2.

here is no,NRC requirement concerning how many switchyards out in the grid Must be directly connected to the station switchyard. The allowance of a ccmmon right-cf-way for the offsite power circuits implies the acceptability- -

of terminating at a common dis' tant switchyard subject to the ability of that

-switchyard to meet the same design criteria as required of the unit switch- --

yard.

~-

.The ability of the ccm on grid switchyard (in this case Midway) to meet the same criteria as the cor. mon plant switch ard (5t. Lucie) is necessary to

/

assure the ninimization of simultanecus failure of the grid transmission P00RORGNE

.U. l / 3

~~

-5

.,. d i s u t il i z e t beca u e 'n LcE\\rconteni

ne i.Mependanr.e

'2 syst m c'1 cssociated 00wer sooplies anc she physical separation of 'he busas upon

.hich the circuits terminata.

Apolicable criteria and noidance for the revie.1 of 3 com on switchy3rd ar2 included in Erclo;ure 2 section III.c.

The Midway Substation design meets all of these requirements The most important consideration for the required minimum o f two offs e

pcwer circuits is ' rat they not be the weak reliability link in the offsite powe# sup?ly systc..

As long as these offsite cover circuits have a reliability equc' to or greater than the offsite poxer system to ';hich the y

connect, where they connect to this system is of secondary concern The availability of offsite power to the nuclear unit can be no more than the lesser of the availabilities of the offsite power system or the connecti ng offsite pcuer circuits.

In other woids, no matter h]w many circuits connect a nuclear powar genersting station t3 the grid, and no matter how well they are designed and protected from postulated failures, a grid blackout renders them all useless.

The above availability considerations are exactly '.;hy GDC-17 includes the words "to the extent practical" as it is in recognition of this situation.

Based upon the above, it has been and continues to be the staff's conclusien that the St. Lucie design meets the physicci configuration requirements of GCC-17

}kk GJ

/

h m.e--

u:

..c.,:--

e. -_

..;..s c, :

The applicant state; that three separate events iM to the 147 14,1;73 loss of offsit: --.'er at St. Lucie.

The St3'f does rot agree.

The first event cited was the planned removal of the 200P,v hnch to Fr2tt &

Whitney transmissisn line for testing.

This actico can nat be counted as an independent event. The remcyal of this transmissica line was a premeditated approved actisn that changed tne sti.te of the grid system.

It is from this olarned net, state t' rat one cust start counting ind2 pendent events.

In this case the t.:., indepenlcn: events were a switching error and a previously undatected maintenance error.

Applicant provides an analysis which demonstrates the ability of the Midway Substation to withstand two independent bus failures and still maintain the ties between the grid and St. Lucie.2/ This acade;nic exercise shows some of the inherent flexibility incorporated into a breaker-and-a-half switchyard configuration. This point has very little bearing on the capability of the grid with respect to St. Lucie.

'r'i t hou t being quantitative, a bus failure is probably the least likely failure in a power system.

To postulate two such independent failures on equipment located so closely together without affecting any of the intervening electrical equipment is of Icw enough probability to be considered incredible for licensing purposes.

1/

-2/ N? 7 Acnlic w response to Ecard questi r ;1, 300RORGNAL 3

\\I

1. ')

g

.,'pplicer: al o U m :'r r-U:lysie n'

?

ch r,y ;

n relic bil i ty cc ;ld be gair,ec Ly bringing one o f :.5 > St, L;.ic :crsaiscion line: die 3. tiy t o the Parch Sui: station.1

The analy',10 app ars to be carc fi.lly constructed to demonstrate no reliability can be gained by a physical change in the crid system. The staff is not convinced that the modified desip, a; pre:.cnted, is tha deign trat t!'e Uplicant would choose if t!w 9:eil ' card should require a grid connection for St. Lucie at other th;n the :ti3;ay Substaci:n.

A sacmingly much moce practical arpecach.:culd have bien to - vestigate what lower soltage distribution systems in the area of St. Lucie could be tapped for a connection to the nuclear units.

This concept vould leave the three St. Lucie to Midway lines intact with ro reduction in present reliability.nd would provide a relatively low cost alternative way to supply St. Lucie a grid connection at other than the Midway Substation wit'h a definite increase in overall reliacility.

i.

The following is a hypcthetical example of a way of providing a separate connection to grid other than at '<idway.

(The viability and practicality of this or any other hypothetical enmple '.ould ultimately have to be determined by the applicant.) The source of this example is the grid configuration figure presented in attached M to the applicants tostimon',

on this question.

There is a 132Kv right-of-way that crosses the St. Lucie-Midway lines and goes betucen White City and Jensen.

This line could be t.cre d a t i ts in tarsecti:n '.:i t h the n. w. i e to i ' elly l i'm.

Aiditic al 5\\%

^

Pa ge $ Applicant response to Coard Question Al, 300RORGin

8-Ccc uce Oe ci-ruit bra 3kers a this end of the lina.. auld r:ct te n scassary, it, Lucie to ' tie ta y lines.;ould r.ct.. ai'.sred and the axisting desi;n could be run ca one of the al ready.r.cets GCC-17 requirements, this 13:<v lina This is particularly insortant three existing lines of transmission to..ers.

as there is only one right of way allcwed to the applicant for crossing fran the mainland to Hutchison Islar.d.

Ternination at St. Lucie would require a transformer and some 4Xv switchgear.

This line aculd not be required for bulk power output from the generating unit: 2nd therefore could te dedicated l to

..:, s with a r quired power trans fer capibility equa to "'e : erg;ccy the combined e..ergency loads of tha two units, This tr'nsf c:pability T

is well within the limits of a 133Kv line, I do not believe it is necessary to cake this or any similar modification to-the 5t, Lucie design, however, this is the type of enalysis (with further su; orting details) that I w;uid have expected fr m the a;plicant, 9

3 i47 r

9 300RORSN;L

.n.

e A

rigure al Casic R2cuirements Of GEC 17 i

t i

i_. -, _...<

c:. : ",

i e u -,. s ".

Q n %s 8 6 --

g i

-l P.

4 c n;m s

u.

s.

""1"*

• s.' C..,%,1 1 '.'*..?.,

p, s..e 4

C-Rio

- ; : ~. s.

., e, c,. m...

e i

i

• g 9

')

I e

i es

,,.9..-

l c Ni i C..r mr s s

1 Lt N C UN2 No.1 N o. 2

'k f

'/

j

(

N

\\

\\

\\'

,.,s unti ~

+

(2~~. '.' 'E O AIC P

~

e e m"e

~. e, -e,. \\.'

V N c s. - e.

) 1 "_.

p O

'"s t.,,,

S Y S TD.8 b h 7 '".T '#

b'[EIY, f

(A g

-4 c

.,,i

. p. ~ s. 1.,,. -

Q; sy"

's e..y. _ 7...n

,4

./

--- -~

~ ---*

c u. z.. (. c y., r s,l

.a_.

i 3"*

P00R ORBEL

_ )-

l i

TC.

For its part, +4 fi nt par ';rach of CCC-17 mc.rsl to as"blish an unattain ile :et of candii.i.n3 i;r cle;t.ical pc.. r systc : generally.

It reads a; folic.as ': 9 a;is a: M ):

An ensite el.:ctric scu.ee sys m, anc an of Tsite elactr.c pc. er sp.a Gail a scoviwa L

,.w.lt functiuaing af strucures, 9:" s, and c^mrin *.s P orm nt to sifM.y.

The safety functica for each sjste, (::n 'rc the ct er syster is not f';nctieninc) shall te to arovice sufricient capacity ard ca.' ability to assure that (1) specified acceptabla f uel design limits and d23icn concitions of the reactor coolant pressure boundary are not exceeded as a result of anticirated 00eraticral cccurr ences and (2) the core i; cooled ana containfent 1stegrii9 bd other vital functions are ;..aintainea in Lne event of postulatad accidents.

This paragra;'h requires that an a.szass. rent of tPc sufficiency cf the o f fsi te ;.0.;e r sy s tcm ; tart. i th tc ::s c,;; bn tnet tra cnsite sy3 N1 is not func-ion f rg.

That ;dssess: Sat "_st then consid3r t'e eff3ct cf

" anticipated op' rational occurrences. ' But loss of the offsite power system itself ray reasonably be considered to be such an occurrence.

The parties sneuld, therefore, explain hoes the St. Lucie plant can com-ply '.-tith the literal requirements of this peragrapji as written.

If it cannot, they should attempt to justify the situation in ten:s of the purpose of the requirement.

STAFF RE370?;3E TO 502RD QUESTIO:: A2 T,he suggested literal interpretation of GCC-17 must. be., reconciled. tith the other regulatory requirements of Part 50. Appendix A of 10 CF2 50 pro-vides a definition of anticipated operational occurrences.

One of the examples given is loss of all offsite power. Clearly if all offsite pcwer is lost, offsite pcwer is not available to bring the reactor to a safe shutdown conditicn.

This is true not only of the St. Lucie design but of all nuclear power generating stations. The fact that this particular anticipated operaticr:'

occurrence can happen is one of the fundmental bases for the requirement of GCC-17 that a single-failure-proof cnsite pcwer systen, irdependent of the offsite system, be included in the design.

I believe the Staff's interpretation and enforcement of this aspect of GCC-17 is censistent ei'.I L ic

.a.

...r, I'

3 9

pm ORNML

-> ro m n

W.. r p r-r"5 c TO fo'_ I C ";T 's "N.! r R ,

The gpiicint's respense to this questica provides icTe very pertin.

observaticns concerning the proposa of the criterion.

Tha referencas to GDC-24 and 39 which were issued on July 10, l967 as the forerunners of the present day GDC-17 proviae an excellent vehicle for demcnstrating the evolution of our present day recuire erts.

Hes,ever, I believe the evolution of GCC-17 is not quite a; strLightfsruard as che cculd infer frcm tr.e apolicant's testim:ny.

I have,:rovided Tabla A2 (attschcd) as a dirent ccroariscn of 9:-l'. cd de previcus C:C-24 and 39 cf 19C7.

Refer ing to Table A2, I have ideati Fiat *.. 0 major changes and ene ma_ior acdition from the 1957 GDC.

The first : ajor change noted in Table A2 concerns the onsite po.cer system.

General Design Criterien 2; (1957) applied a single active failure critericn to the ensita power system.

The prcblem with this requirement was that an active or passive component in an electrical system is no where near as easily, defined as an active car'0nant in a mechanical system.

The cost notable analogy is probably piping in the ::achanical system and cabling in the electrical system.

Piping is assigned an American Scciety of M2chanical Engineers (ASW) code classification based upon what system in which it uill be u:ed.

The end result of this classification process is a fullf documentec qu:lification of the ::iping.

Once qualified by this process, the piping is considered a passive element in the rechanical system.

There is no analogous classification peccess for cabling.

Realizing this problem, the staff subsequently amandc'i the definitico cf single failure in 10 CFF. 50 App 2ndix A to reflect the posi; ion that both actise and mssi"c c m re'tc. c'w il d N i n cl u bd i n t h ' e i n ~' ? failur9 ar al ec il o* el ectric bj,..;.s.

,1.T' a i s ;,, r.... ; ;

v.

._.'.a:

v.' s. t..

.J

.w

~C..

FCQuircTent for a fully single-failur9-pr00 f nnSite ?!UOr Systcm.

This char.ge p ' '. ; e f ' 'r0 30n M ev3

' ? Critcri t 09 l"c J:if" p'

' itC "

300RORCN;'

5m

-\\2-Tr. 3 c econd..:a jsr shn,:.s t s d i n 7;:.i e 1.2 i s

'ij tae Sr.ca evolut V.

.. w...;

?[ r'.Guir; v.: 3

e f ',3 o!! i t e f.s '.,2 r s j, r. ;r

.t 3 3 ; ', C s siM i... J. c T0r

a t-g, '. a --., e r q. e * -

.;: ;n -, : ) +, j,-~,e-.,,

t, _3 3 ;,31,e

.o-and of fsite power systers, tne a'<olved requirm ents foc these two s/snas are quite different. Again, the single-active-failure regiirement did not fit electrical systems. Whereas the evolution of requirem!nts fo: the ensite power system are contained in the last four werds of paragraph 2 of GCC-17 (i.e. "assuning a singl e failure"). +Ee e"ol utiso of racufec " rte fer the ofisite cceer system cc :rf se the entirs thir ? ^ ire gra :h c f G' C-17, A single-failure-proof offsite power system is neither attainable nor within the purview of the iiRC, Faced.<ith these realities and the a.?biguity of the single-active-fatiora requirer.ent; the staff provided an explicit description o f v; hat is required as a minimum o f the offsite pcwer syste' (i.e. GCC-17, paragrcph 3).

This evolutien of r7cuire-ents for the offsite power sys cal is

~

much more~ realistic tilan the origir.al single-active-failure ~ requirea.cnt.

The major addition noted in Table A2 is paragraph 4,of GDC-17.

This paragraph introduces the requirements for transient and steady-state stability studies which de.Tonstrate the ability of the cnsite and offsite power systems to withstand system pertabaticr.s and requirements for the ninimun of two offsite pcaer lines to react cr.d ;erfcra intependently of one another.

The at.c.e described changes and addition-feca the l?57 criteria to present criterie signi ficantly improved, clarificd, and stren-i.ened the require"ents for electric cower systems.

In summary, the Board's in;.errretation of GCC-17 100R ORGINAL

e 9 9

- : a-

, 1. 3 A [ i

[9*7 9 **

- 8 is es J.

i e'

p

• ,, =-, ' u

• Q,{i

, A 9 [b" - iM--

.s

• .A g.,3 e

V

- =

e u

a w

y COE 'u f ' ': 1f1 0n~ Oi~ I'3-I7 JJ i l'AS

'2 '.* * '; I ' ' ' ' ' C ',' ? i' ~ " ".'/ ' ' P S -

26, v

4 e

P00RORCNL

e 9

t-f y 9 r3 e - e9 i

• klat.

J '. f.,

Cro.wi ;an of Pr snn t CCC 17 '.'i th l c. : -

u. c re ;s u n,s.

O.

r.,a, e & ~,

.a; v

a --

3.,r

\\

\\

c-c e, n n.

i

-,3, l - n/ ;

s -,

a J L :0 i

(,. Een.,

u u,,

w 3

o 4

9 cnfeno, n.-rfeerne wtre int m.u

--h cnme e:retric por.- sr:'. a std

.f. e..:*e l?

(

e-etn m.-r reurn en:

.e ;.rw.we to

,,a uDC 24 (all) p urc.:

...c.:.; u are: re i, mm.s.

j.

Stra ight for'.ard 2ne corn;c,enu.m m..2nt a r.af et n.

r.afety fun:tm for e ch

-t e, w : 2...n, i

the other sy= tern is not function.-d ch.G be I !

to cronce suficent c.rv.ty n: cm;.ty

'o PJ.eut* tt4t tlk. ;* : 1: j '0cep.4t;,* fuel

{

?

dcs:rn 1:r.c ar.d der n e rc.a3rs ei the i

f re w or c e u.t pre.<.cre k m.. try u, ca.c

..- GDC 39 (2nd sen t::nce)

! Minor char.ces ez:e*:.t f a s a tt a;.; cf u.

epate:

8 0; e r.s.

at.:ns: w rences

=d i:, me m..

ec

-o

.1 c.nto.. ;:n ct t.1; :-,:y u d cther i

r.t.; ic. r:..tre unt er.:d in tse eie.t l

e: :=

n. :.u=t.s.

i-ne cnu : etectne noser sume:. t. iud.

tes e., t.uv.n 5. ana he,r.me ett:c d.s.

< > nC 30 0 n-

• e e n *m'..<'.. e )

S 'u ra D. f' "..7. c i-..>a rd r

c tribution :, a shr.'1 r_t.e r ff!:.mt inde.

l,

n derm. re-2.

• nen:y. n.d teca.:

to p n form ti.eu-re.lety funct;erts s saanz a l. GDC 39 (2nd senter.cc - ensite) ! Major cnar.ge sir.gle f aliure.

Tie:trte power ficm the tiv.riission ret-1 mort to t 'e ors te e' ectr'

L3tr:bti 10n i

l sy; ert sr.1:1 be sJ;'pl:ed by tt o ul:y ictUy I

l Indeper.arnt circasta (not r.rcerarGy (a rap-f arate r.?nts 3! A &> fdasped A.1' luct.'Or. 43

?

as to m'ini. E to the eatent p ;." t! the f

112c; h:K.J cf thor tsunmMus f*

2re j

under c;erstir e.nd postcILMd e :c dent ar.J em L cnn:cnta cend.u ;n:.. n m.c2.4 ud coacarcn ;o >:'.h circu;'s :s r.ccc0;.ab;e. Each

'l cf t hese c ~* v'.s t r:Jj be C

,3*d to t*

n.tstic in asif.cizet t:=e fed:nns a ;os.'

.'t GDC 39 ( 2nd ser'*"r qce - o f fs i +e)'

'"a I'c r c'qanc,a of.:1 c. Ate a:te nsur.e rcar-rat pow er rup-pil?S End t!'e other offp e;actn: Power cir-i et.:t. Lo s.r.ss e that spac I!ed acet p*.able f uel

{

Csagn lim.t3 f.nd des ;n ecnc.t;ons of L.he reactor cooJr.t pr*ssu e t>>an:Lary a.rt not i

ex etia d. Ore of the.~* *:.rcuG.: th:J1 ta da-i s.;ned tc be avt:1.t;e n'.'..n ta fes seccads i

fobesing a 'rse Of <coiant L dent to ensure that u.,r c coo;.ng, centA;.nmen, tn'c-;rity, I

aTd othet W Al 6&fety it.I.0tiona n.*e tr.Ala-tLned.

/

3 Prons:ca: ahi;I be triclud-d *.o re.tn!.: e 3

the pr00&c:..ty cf WJ.x electnc power trem Any Si the t 51. a.; su;pi.es As a result cf.

f or c:'rc;3 m. au tne ic.xs of power carar-it o

11ed by *. ' nL:.c?.' pow?r urut, the IM5 Of

'i

• e,-

a: 4+*

p:ur f- = the t:r.s::ua. n networa. cr l\\

the ;0cci e'

r a.;- fre:s the ca.1:te ela t: c i ',

pcserav;;;.cs. -

,,l,)

i e,- er" 300RORSNL

1.

0

,s.

.m

-e,--

j i.e

.n.3 i7<pa

..n.-

>rrai

.s..

aim g r.e e: -

a.

a T. c a.

..S T"?. ; '"

3.

4 u

. u c.

2,,c s _.-

J-

j j..';3

- J.

..,-4 ln.-ya

. J.

9 1

• e

-~7 L,,

t 4

. e..:.. s a..

3

- m....

n.

..,.o,-.,..

.,e 4

..,ca

.,17

-.3,s. a. J.- 4..,...

<. e.. ;

4..,,,, < - -, _

,. c;. rm.. r.

j, /.

.u,."i j...,,

.r.m....

a u.

.. ~

se...

4. - < n -,..

4.... l.,

3. ~a e..

.n.,...,.4..,-

4. c.

..c..-s.-.

r..m..

..u

~. -

probabil f seems b;:ed cn historical evants, to lie in tF2rarce 1.0 to 0.1

<v a >

r -

a,,;.,.,

y e r

=e.,.,

.n.. a c :..

..,o a n.,1.. e-

c. u -... u,.,

.... a

.a

...a.

. =.,..,.

.e.

w=

v syster.s (diesel ger.erators) will start.

The prob 2bility that any one diesel ceneratcr will fail to start on demard is taken by the staff to be cne per nundrad cemancs, i.e., 10 ~ 25/.

If these figures are accurate, then the cccbined crcbability for the " loss of all AC pcwsr" scenario is in the range 10 te ic per year. 2j/

in this

-2

(.,,3. a,

s..,a. :.

.z.,..a. r. r.,/1../ r i c.4 T s,

. i a.. e : r.z,v. u,l,

. i c.,.. a s e-. 2

, a

r.1

...c a

.v ic

c.. ~.. n. r 6.., ]

s j

3. } 4.. e.. e. - (.a.

r. m. j.. 4..,. -,,,.., -

.s..

".r e a

1.. j.n.,

e. r..~. '. ". n.

u..

e.

m,,

ri 4

i o.

presence of ha 2rccus cateriais or activities :q the vicini:y of the ;1 ant" a >a, J ].A.

e,.... :.. =. e. s.

e. r. a..s na..m.... c.

.c._-

p fj.=.,

..c a.

a

a.. -,. e.

}.

4. a-

"..e;

.a 2

a a

w basis" even-). 27/

'Jncer thsse cuid2ii tet, events wi:n a. realistically calcu-latedprobabilii7valueofa leisc lo ' per year (or lo' par year for a conservative calcula:icn) must be so censidcred.

The "10:3 of all AC pcwer" secuence is not precisely within the categcry of events centsmalated by the Standard Review Plan.

Mcwever, its ulti. rate result -- assuming tna: ccwer is not timely restcred -- is an unprotected loss of cco,iant accicent, the consequences of which are,41,e,iy to cv-o=d

ne guidelinec of 10 CF.R ?:rt 100.

We co nct anderstand '.ehy this sequence.cf c es-.-

(i.e., less c: otrsite pcwer ccmb1r.e,. witn failure or c.iese;s to starti. ~u c.-

appears to have a prctability well accve the guideline values, should ret be tak,en into consideration in the desic.n of the plant. --28/

The parties are :c addrer: this pcir, setting forth their reasons for achering (if they cc) ::

I centrary pcsiticn.

--25/

Fit:catrick Affidavit of June 12,197', p. 4.

Also see Regula: cry Gu'de 1.le,so,

e. e a. i, r.

a..

u 25/ This conclusion further assuT.es ba: the failure of two diesel ganeratc :

to start would be statistically ince:enden: events, an assumcticn '..hich leads to the lowest likelinced of ccmtined failure, and..hich micnt be ncncenservative if there exists the pctential for ccm.T.on failure' modes for the ansite sys: Ems.

9. 7_7,p..e g 7:/ce7,.e a r.

4r n 2. ?. 3, e.> r.ar,.=,-h.

.r..

r o

3. o.,f

. w h.= ". a.

a c c o '

• a. a t h a. <**

4 4

r c a,.. : v' 'i =

. c... d a -d

.O..' v 4. 2. i P 'a u'..

"s. u" u, 't... n "o> 's o a-c.e a.

2_

3 c

in ar.ather

,^.se.

p!!blic 3Srvic9 ~lectric i d

't F N r-(Fore Cr:eg 1%izz T nes

\\.

2*

- ~ '.

5 154 9$ $bb-

-l$-

.g inq r

.3 n ;ve

--3.---ced norcri:n! raiiarility coais e

e.c....s.e to the static ; blac?c; :::c.aric, "hia ::ec:ri b: reca^tly Le:n id:n.if'?d as Task Acticr. ?];n (T.:.5) n--,

and it is expected that nu~crical reliabil :y goals will be established as a result of this progran in much the sace.anner as was done for anticipated transients without scraa.

Because this tssk action plan is in its initial stages of development, no criteria hava been established, i:e have therefore at:cted the 10-7/jea goal referenced abova by the pppeal Beard i Or,ticip;t on of the result: cf Ip: A

• 4 and have recuired the a;clicar.t i

to demonste::e that this goal can be m.et by the St. Lucie Unit 2 design.2',

c.oc; ::c. o. m...c r r. * : :L r C.r;.m.. S r...., :,

en.,s,,n Cn:-iaC,.i,.

a,..m t,. w s...

...iu The applicant's answer provides an analysis of probability of ac power restoratica verses tine folicwing a station balckout.

In cur revie't of this analysis,

,e found an apparent discrepancy between the Florida pcwer & Light historical data and,the ccrstant used in the probability equaticn specifically, in resconse to Board Questicn,83, the applicant demcnstrates an average duration of 25 ninutes for its loss of o ffsite power events.

The applicant then proceeds to use 1.6 hr-I as the constant in the probability ecuation.

This constcnt represents an average duratian of 37.5 minutes which is conservative based upcn the historp.a1 data.

The 26 -inutes would yield a constant equal to 2.3 he~

'de were infcered by tha applicant that the 37.5 minute figure used represented one star.dard devia-tion on the data base for conservatism.

1*/

Siegel Testincr,y, September 21, 1979.

5 155

^

e..

,.e

-I

'J:e ct *he 2.3 cc ti:, :enstant et ces the-applicaat's figure of 3.6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />

~7 to coet the 10 '/ye,r pal

• .0

' 7 n;urs.

I f..: u;a Oc a.:v /:;i.. Osti -t:s : f

-Y diesei <jenerator unreliability used in the Reactor S2f ety Stac/ (i.e. 3 x 10 ')

ar.d tha 2.3 hr restcration tito constant, the tim ceriod becc~es 3.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />

' essentially tha figure presented by the applicant).

The significance of these tima periods is that if it can be demenstrated tnat natural circulation of the pri.rary systan can be raintained for at least the ti-e period calculated, the pro-bability of cure damage due to the station bicckout scenario fill remain belo..

10~

Frcm tha aboie calculaticns, if the ;ppli:1n; can de.nonstrata ori-per year.

mary systam integrity (p i.e. natural circulation with no excessive leakage) for four hours of s'taticn blackout conditicns, the probability of core damage is well 7

belcw the 10 * / year goal for the St. Lucie 2 design.

B2.

In line '. tith the above discussion, the testimony is to analy 2 events that would cccur between the " loss of all AC pc;iar" and the violation of either the fuel design limits or the design conditions of tha r? actor coolant pressure boundary (or any portion ther:of).

In particular, the parties should, if possible, re-concil e,+hmi r <!' '#es ir.1 raererses to question C.l(b) of our March ~10,1973 crder,bm ar, i f n:t. poi":,ug :recisat wnere the disagreerents lie.

22/

(References fn 24 reproduced belew:)

The staff's judgmcot is that the first faiiure would be that of a primary purp seal, at about One hour af ter the icss of AC p '. er -- resulting in a small loss of cociar.: accidea:.

(Fitzpatrick Affida vit of June 12, l.,, o, p. I..s).

si-m ~ a..r -.i r..?

-nc-T T. App r..ee--

r Tr c.at:,

My affidavit filed in response to the Apccal Soard's order of Itarch 10, 1973 gave tha apolicant credit for ene hour fellc..ing the Icss of credit was taken frcm the applicant's PSAR section 9.2.2.3.1 which demonstrates S*

?003 D M R

^

.m r i t.y far an car o f ep. ra. ;'

j l',-F

.an..

s corpcnem tool ing '<mer.

Ta sta f f r+cacn iz ed tr > conservatis:.1 invai,cd in equ-ting tho dy mic candi*iens of tra c o '.'.y.nire vorces tna sts+ic condi'icn3 of the curo stcpped.

Mc'. lever, in the absence of direct test results for the static condition encountered during station blackout, the staff.ias unwilling to atte:ot to extrap0 lata from the applicart's analysis.

This rcnains the sta#f c o s ; ti c o.1:

.- i, - -...7...-.,.

.,.,e,,..

..e.,..,

7,.~..,,

The applicant notes that Unit I diesel generator can be aligned to Unit 2 safety busses and estimates that t.ic ten could effact realign:cnt of cr.e dias 01 generator in about one hour.

The applicant has not taken any credit for this capability in his aralysis of probability verses tire for restoration of power-to 'Jnit 2 followirg a station blackout. Hoaever, prior to the completion of TAP A-4J,. it is noc clear what credit the staff could give for\\this design feature.

Station blackout at multi-unit sites should Le analyzed in depth prior to determining the criteria governing reassignment of onsita pcwer scurces.

Such an urdertaking must be part of a comprehensive effcet wcich in this casa is TAP A 44 Beycr.d this, the site specific situaticn of requiring two,T.an hours of effort to effect tha trcasfer intrcduces opcrator error as a factor in the probability of failure to restore poser. Also, the applicant's crcbcbility calculations assure th:2 the t.co Unit ? diesel generators and rest:ratien o f offsite ccwer are being verked cn simul.;neously The additional

.10 men r^^uired the available ensi!0 marra cr.

G 1GJ

- ^

J lJ M

i c,

U 2/

hic. el tcs tin JPy, heptc.

'r,l

,-i-j j'

L

- e *- <-

I, m'3r/, t ;.e e n.dil i tj o f tnr3 ':r-im fi =: el pnr-no rs be%ccn un it is a sery d ;irable d -: iga 6:ature e;p:!al'y for tnc Unit 2 bluciout scenario.

hmver orier to co?pletion o r TA? ; 'a, it i: not cl w hou,uch crecit the sta ff could give for tnis design feature.

83.

The testirony should c:ntain a discussicn, supo ted b such data as is 3eli;able, rel<. tad ta :ne time t N3 *,i at to recuired ta s'act a liesel tenera tor as;um'ng it failed to respend r.o tra in tial, auto-sti ~ si:ral-

,, y-- a n 3 3.,J a

.s v.

The staff does not have cn independent data base in order to calculate a r.e2n-tinc-to rc; air (MITR) for the diesel generators in nuclea,r service.

The Licensing Event ReporM (LEls) submitted in accordance with the guidelines of Regulatory Cuide 1.15 " Reporting of Operating In formation - Appendix A Technical Sp:cifications"have not raquired MTTE data for diesel generat.or failure reports.

Therefort. if MTTR data had been included in any LER cn diesel generator failure,

\\

it would have been because the licensee volunteered suc:) in forma tion.

Regulatcry Guida 1.103 "Pariodic Testing a f Diesel Generator Units Used as Onsite Electric Power Systems at Nuclear Power Plants" (October 1975) established the requirement to report duration o ? outages from. hich.'iTTR can be calculated.

This regulatory guide applies to all construction permit applications following its data of issuance.

No operating nuclear plcnts fall into this cate;0ry 5

i5B

_-)O_

T' 2 Nul a tc r/ R: viem nts Ravie.-i Co m *. tee (" C) revi..ed o?cala'.ory N de 1.'.N and decid M ti.at it s! culd te connd~>i f 7 b?cPfitting t0 "Orliini reacter; on a case by case bas's.

TN svaff has since folic'..ed this Policy decision ar.d scre operating plants have been required to meet the rea. ire.cnts of this guice.

Tna nuccer of plants iv/olved are not sufficient to yield a sta*.istically meaningful data case and additicnally, the cutage tire reporting rac,;ir.mant applies to current failuras cr.ly r.at previcusly r: port 2d failures.

-...pe

.;. 7... c ~

. ~ ~. * -

n~

~q r

- ~... _. _. _.

J_...

The Ocplican*'s deletion of certain diesel genentor fsilures due to ccreective design measures is appropriate and there is m. erit to the argoent that use of historical data for returning a dio:el generator to service is censerv3tive as r.o u.1due tire-pressure constraints e.(isted.

G4 fir.aily, in the licht cf the discussic, of points 2 and 3 ata te, the parties are to revicu ocssible m2asures for decreasing the likeliheed of exceecing desir limits en the reactor fuel and pressure boundary under the asst..r.ption that there is s;.ne time available to activate an auxiliary po<ier source sub-sequent to a total loss of AC p0wer.

c n e :: r. - -....:.s -c.

3 s.._--rn". o,,

s.

m_

2 s.t..

s.

Tr.: limiting even: failcain; station blackout is loss c; national circulation in the primary coolant systes(i.e., loss of ccre ccoling capability).

Loss of natural circulation is brought about by a significant loss of primary coolant which. auld escap2 through the reactor coolant pump seals.

If the leak rates through the seals can be th:c.n to provide at least four neurs before sufficient inv0r.tcry is lost to stop natural circulatico; the probability of P00R ORB M 5159

i ou ;F ~ m r.

Jrc.

a?

, nty fc';r a 9 :. u.. ;, f 4 - mt i.. 'nterv is lost, the promii tty o f loss of n :tural circ !imen is negligiM e and tr.e first depen:cnt ":ilure (3:su. ming ac r?.icr b:s not been rest: red) :.uld then becena thL loss of capacity in the station batteries.

This failure would take away the operator's ability to control the auxiliary feedxater systen and all intelligence on the plant status.

... r 3-c-r.-c.

cr.,tCc.~.,e

,,c

o. n,,:, u,hai,0 i ;,

n,,;.. :. o.

aan.

, a

c..a c ;s

i..

s

... a Upoa damonstration of at least four hcurs of natural circulation follo. ing staticn blackout, tt.a staf f cor. cars with tre applicant's anc.:er t"at tre potential for exceeding design limits on the reactor fuel and pressure boundary is accr.ptably icw.

C.

S_l, e :., 0.. L....., -., u t. rt t. e u n,_ L.< i c....-

r..-...-

s

. t ar01LL1I

..q. 'a 4

Accordir.g to the staff, the applicant is beinc req 7 ired to d2 fine condition:

in which it will put it; power distribution systcn in an " alert status".5$/

Rt such tiaes, loss of offsite pcwer would presumably be m. ore likely tnan normal.

We wisn to be advised as to the existence of r:usuces that might be taken to' assure, er at least to increlse, the reliabili:y of the ensite power sy-tems during an " alert status" period.

2,i

--uvi Fitz;atrick Affidavit of v'cne 12, 1973, Encio:ure 3.

S T n'.c ' ~. ~..~u u"F2c T.o P o "s t " t t ', I r..". C_

S o

r s.

o-The staff concurs with the applicants' answer to this cuestion in that the diesel generator could be manually started iq anticipation of a los of of f-sito pcuar.

'n fact, the staff has enforced such a rquirement in the tech m-requires the wei f;im] of many t;; tors ;,* :J;t be arri/ed at on a casa t; 300RORSNR

,m b

IDU by case basis.

f32ny diesel generator desigr.s including those at St. Lucie cannot be run unicac d (i.e. at idle) #n< cv.tenced ceriods of time.

!;c lead or light laad cr.aration <ill cause inco plete cc hustion of fuel resulting in the for..a:icn of gum and varnish deposits on tnc cylir. der walls, intake and exhaust valves, pistons, and piston rings, etc., and accumulation of unburned fuel in the turbocharger and exhaust systen.

The consecuence of no lead or light load oceration are pctential ecuirment failura due to t"e gum and varnisn d2p? sits cai fire in the engic.e e <::aust e.ysram.

Imc;ner factor to be considcred is that " alert sta'.es" on a ;0uer system ccule happen relatively frequently.

l'ost often the alert state '<tuld r.ot be of icng duraticn anc could he terminated, fcr examp12 by startup of ansti;er generating unit or units thus reverting the system back to a "nocF11 sta te".

Once in a while the alert state will be terminated by anotner contingency event (such as loss of a critical transr.ission lina) and the state of tha sy: tem will go to an "erergency state" where at least part of the system has been lost.

The above situation could place an undue number of challenges to the diesel generators where in retrospect most would have been unr,ccessary.

In the instances where the staff has required the manual startir.g of cnsite diesel generators, tha underlying cause has usually been a specific iten of concern such as an impending tornado.

The staff also concurs with the applicants conclusions that sucn measures are not required for St. Lucie.. The bases for this conclusion are that no-lcid running of the diessi generators for every alert state that the grid systea

~i S t

.'n + e r co'ild on,acessaril" ha ror their re rfor~' c e in 3 rell cc:rrer,..

5 \\6\\

. - ~ - -

a m

23_

it.. !d c:n;c cnt;;2,: ' e r :rs tnt ct,ulJ br

.o :W er incortar.t frc-t i c n,, c ri d t,b.: <;,1 :

(, 2 S.,

sy s.

.1. i s 2.ec i fic il/

2 2s i' r".d La G 2..

. t.C3il start and perfc.m its inter!ad mr.ction n. :e cre._ance of a sincic 12 icca.

D.

Od^0ING I:MOV6ENT CF SYSTE:1 ?ELI ACILITY The testimony should provide a concise, ua-to-dita discussion of existing naasures, or those clanned for tha near future, by '.shica the reliabilit-/

of the applicant's sjsten ?.ay be ennar.ced. Particular attenti0n should be ocid to the se.ir;ly excessive e.=rber of cer:c1'el errors..Sich appear to h3vo led to toe '.'aj il. 1975 cut 2g2 and to n..'s c;c.;ricuRa to tn:

,.3y lC, 2.

i.

IJ// Li3IsICanCO.

q l.3 a c..c e s e r 1 p-

.., c, r r+J J.im.

LJ cT..

m s.

The applicant has and is continuing to upgrade anJ the reliability of the offsite power system in three major areas.

These areas are 1) the strengthen-ing of the pcuer system (i.e. added gcneration, transnission, and syster.1 protec-tica), 2) po..er systaa field personnel training and guidance, and 3) a centrali ad monitoring and control facility.

t The ECO kv system additions addressed in the ap? icant's testimony provide i

a significant improvecent in reliability, capability, and performance not cnly for St. Lucie via "idway Substation Sut for the entira soutbaastarn Flcrida quadrant.

The additional transfer capability afforded by these lines.ill allcu the ganeration at St. Lucie, "artin, and the remainder of the system to ficu into the !;iami area upon loss of ger,eration within this heavy locd area.

The lack of this transfer ctpability has been a signi icant contributor to a large percsntaga of t: e cascading grid disturbances that have originated in southern Florida.

~

i62

.a..

I t

i

'b2 a nii an.'s res:unec :c.his ruestion :lva m eesses reasurn i. ot hava teco t-:mn t o A a + i w s,,i *-M m' w e: nel u rors.

The t<.o n:

oc cbl s ri

--er,. c

.,..]

ce,

..w i._.._.,,u,n,.

ca, jp.,,

. _ - +..37

+;..

.y.

3..,. 3 -.,

...,~+

nr,

.s :

... - as m.

under car.'.imyncy conditior. neice to alic ti.,1 t'e s..itcn.n] a'u fici; p;r-

~

sonnel are equipped with as,nroved written peccadares.

The fic t featura is a result of the positien taken by the staf# requiring same.6'/ The second feature ccccerning approved written pr cet.25 carks wnat a.0p;c.cs to be a maior change in FP;L philos:nh/. At tha ' me ~,-1373 meeting bet' eer.

.~.a.

c

-s

.c n..n..,....,~... le.,,

,,,J,

.-4=..

n,,. a.a:.,

er<4, c...-

...., r.

u.

~

.s.-

Fr'OL sta ted tha t. ri tte.1 peccedarts f;r field pa,;;nnci..are nct ;;;j cnd r.:t consider:d of ber.cfit.

Ecth of these featuras (i.e. analysis and uritten pro-cedures) are.rajcr iTpro',s ents in FPit operations.

The tnird c.ajor contributor to overall pcuer syM.cm relicbility is the rcu cperaticn;l system dispatch and control cente?. As described in the appit-cant's, testimony, this center provides tha system dispatcher with a tre en-deusly powerful tool to aid in the optir.um operation and, co occasion, restoration of the pcwer system.

In ceder to utilize this control center to its full potential, extensive operator training is required.

The appli-cant has purchased a training simulator and is conducting trairing programs for the operating staff.

This ccrplex ar.d pcwarful tool can only be as effective as those who operate it.

With properly traincd persennel, the dis-patcher confusion and lack of system status informaticn that cor.tribu;ed to the l'ay 16,1977 grid dist'.rbance are eliminated.

" IJ..

g,

% [f' j-J e q 30 907"

.. gl e%

)

f,e OO fe -

J. Ien 8 =

O W

P00RORSNAl.

~

.,,we.

-e--e.

---e-e-m-

--e.-

- J ';-

! *i.

~.-.,,er,

.w-n.,.

T, S. L::c: : tjni" 2 n sign cer:r.

gy.,, '.,, re :u f r.2 rents or rg 7,

,,,, cucnsten' ;C. D/. dctua] + t3 r,. rcLu're

+s,+

nm e -.

o;.

u:e e.,u t.,

coolan t *"" d,*'e_ DIS ca" endure s ta tico biar;c,.+

r, it can be concluded that the probabilit y o, c Ya da~'a9" d'a to a wta tica blackout is be t'-.,,he 5tandard Revie.., plan 2?.

nLIericc! r2]ig3f) j.Y 7011 of IO' t"?r yea r.

Th,'

o<<* 3ite pcs.;3 r g>un g;"I FIIII nave undgr70n A

~

(prige tg i g p

~

7 Drur? tion) ;q<",r

. 0S95 to sign'ejC"2D,7 increar,+ igg e

'3 UV'rili r 3 ] 1

',,7 f.,,

y ', - a a, - c..

,IC,i :'d ? ' re 3,, ); p p g 2. ~~

-4 3

,e, G'

I'llS S iCn Sys t;;d Pih iCh C"eri nji e t'4 r'eVICJ5 2 ;0 h.v Sy3 ten fe"r-

~

" *": r. cicy g,,.,g ation South to w

9n1amf a-a,

1. #I O CG0r9f a and t.5erafore strons- -

w.

lea durina.ys+sem ci; turban er, c.,q' s

3 sophi s tic"w.<'

centralized dispatch and c"'It -1 ciu3 fen of tha acceptabiiit7 of

~

e9t. iucie Uni' 9 1 ric p Wer sy te.ms s

rcrains unchanged.

9 e

9

e

.C. * < :a^ to, C. f i e' ". }

,.p.-.w..

..g

-, n e..

e-.,..~.,n,

,,.,,eepsv,e. (

...L

f', s..)

? '. a

.t A e ss 4

s.,

. s s s or

..u

..u a.

c 1. _-. e n

. r.

.f

..s:

.a

__EDUCATIE:

3.5. Electrical Engincering,1371; :ortheastern University, Costen, l' ass.

l1.5. Electrical Engineering,1972; f:ortheastern University, Ecston, Mass.

• ajer:

Electrical Pce,ar Sys tems E,.giacering pr--- 3::" sL "% ! CIC'. TIC'!S Fcca 1972 - lie 7.; I c;orxed for Yankae A.tenic Electric Cor pany in West:cro, Mass:chuc2tcs.

I '..as assigned to the Electrical and Centec1 Enginsering Group and my duties included work on tha Y3akse cpera ting' nuclear plants and tha Seabrack Pr0 ject.

(Prior to this I spent 3 years with Yankee as a cocpera-tive educatien studtnt while attending ?;ortheastern University).

From 1974 to the present I have wor' sed for the fluclear Regulatory Ccmission involved in the technical review of electrical systeras (onsite and offsite power, and instrumentation and centrol.). Through 1976 I vias a trember of the Electrical Instrunentation and Control Systems Branch.

This Branch was split in January 1977 into an I"C branch and a pcxer ' ranch.

Since this split I have ' een a tr.em-c c

ber of the Power Systems Branch. l1y present title is Senior Reactor Systens Ergineer (Electrical).

Following the Three Mile Island accident, I have been assigned to che Three flile Island Site Support Group.

I an a.crter of the IEEE and also represent the l;RC as a rerter of IEEE *:acle:r Peaer Engineering Co:raittee Subccmit:ce 4 " Auxiliary Pc,;er Sjstems".

This 5 as P00R ORSK

~,i..,,., a n

,g7.. >

..2-c-

,or,cary

..an;e a f 0 c f oi L 70.mr Citcuii5 Co cernin; w,. ai

..4 I.

p.... e C F 2..V T.'.'!

"t etry suncards and T.e Offsit: ;;uar sy;te.: 1: referred t: 1:

S

.t iriclucas 5.:

the " pre.arred ;nt r syc:em."

regui r.ory ;uida: c3 dently or core idtatift:d 00,rer sc.:r:2s capable of cNrating indepen id, of the ensite or standby po, er s:urc s and cn:cmpasses the gr i

f on line tcwers, transnissicn lines (overhead or undargrcund), transn s:

itchyard transformers, swite.: yard c:=p:ntnte and control systems, sw idad b:tters systcts, tha main sencrator, and discernect switehts, p power to safety-related and other equipment.

to supply electri:

sten.

The P55 will ravt2w the following featurcs,of the preferred p revie.4 d to detarr,ine that The preferred p;wtr system arrangement 1:

h tran;cission 1.

the required minimum of two separate circuits fr:- t e dd In network to the standby p:wer distributten syste:a is provi e.

h determinf M ti:e adequacy of this system, the indarenden 5

166 a

.e.

re-w,

2_

I t.o (or.oro) circuits is c;.=i '. :: 5 a th: b:*.h ei. trical and phy:f tal naration exist: so as to F.inimize the chance of sinalt:n cus failar.

T?.i; inclats a r0vicu of tha 3::igr-' Int of pcuer sources from the crid, location of rights-of-way, trans-mission lines and towers, transformers, switchyard inter:ennections (breakers and bus arrangements), switchyard centrol s;";tems and power supplies, location of switchgear (in-plant), interconne:tiens between switchgetr, ceble routings, rain g rerator df s:enne::, an:tt' re:;.:.n. y : Pall bc :c:n U lt, cr pr;f: red t. ;,. o r systc? cpercti:n (a:strir.; :tanity power is no: eva11able), tr.e sy:t:r ::'cty fun:t!:n can b; ::::rpi t.:.21 a::uning i sin'i2 fail ure. To utili:e this requirenent, tha : ingle fatlure 13 as:umed to o::ur dr,mstream of the preferred p:ver fee 1 bre:kers at the safety buscs, i.e., in the saftty-related di:tributi:n systen.

The acceptattlity Of the preferred p:wer sy: ten cesi]n in this reg:rd is based en it: c:nf:r.ance with Ge12rsi Desi;n Criterien 17 and its capability to sur,-7y tha reda-M:nt safati cc. penents and systems required by these General Casign Critcria.

b.

General Design Criterion 17 requires two physically independent circuits from the offsita griJ, ene of which 10 designed to be available within a few scett:ns f:llowing a loss-o.f-ce:iant

. accident, III.

REVIEW PP.0CECUP.E3 To assure that the requirencnts of General Design Criterien 17 are

stisfied, the follcwing review steps :hould be taken (as applicabia for a CP or OL review):

a.

The electrical schematics should be exanined to assure that at least two separate circuits from the tr:nsmission net,trk to the standby pcuer distribution systt?. buscs are provided (a swit:hyard may be cart:n to these paths).

[] lbb

~~m

,e-

-c.

b.

The rcuting of tran:mi;:f on lla1s should v2 examin:d ca the staticn leyvut crauings and verified co'ing thc si:2 visit to assure t!.;t at least two incependant circuics from the offsite grid to the scfety-related di:tribution buses are phy:ically separate and independer.t.

Profarably these lines should enter the station on separate rights-of-way, ideally on cpposite sides of the switchyard, should leave the suitchyard on opposite sides, and should tarmin:te at tran orrors located on opp; site sides cf tha reacter or turbina butiding.

.No othce line should crcss these t'.:o circuits. As physical separation becomes less than the ideal, attention should be dir2cted ::r.*ards assuring that no single event such a: a tover falling or a line breaking can simultaneously affect both circuits in such a stay that neither can be returned to service in time to prevent fuel design limits or design conditions of the reactor coolant pre:sure boundary frem being exceeded, s

c.

As the switchyard may be common to both circuits frcm the offsite grid to the safety-related distribution buses, the electrical schematics of the s,titchyard breaker control system and power supply and the breaker arrancament it: elf should be exaained for the passibility of simultaneous failure of both circuits frca single events such as a breaker not operating during fault conditions, loss of a control circuit power supply, etc.

/

_-