Text

LOCA Aspects of C-E Advanced LWR - Sys 80+
	ML20094E943
Person / Time
Site:	05200002
Issue date:	02/11/1992
From:	Ivany R, Kapinos J, Rosen S; ABB COMBUSTION ENGINEERING NUCLEAR FUEL (FORMERLY
To:	;
Shared Package
ML20094E941	List: LD-92-011, Forwards LOCA Aspects of C-E Advanced LWR - Sys 80+, in Response to NRC Request for Addl Info; ML20094E943;
References
	NUDOCS 9202110335
	Download: ML20094E943 (108)
	v • d • e

___ -_ _ _ - -

LOCA Aspects of the. _,

Combustion Engineering

, Advanced Light Water .

. Reactor -- System 80 +

e?n}W 4

S. Rosen R. D. Ivany J. Kapinos S. Sim Nuclear Fuel Engineering Nuclear Power Systems Combustion Engineering, Inc.

Windsor, Connecticut J &

TIS-8328

.) c..-

'i --

~<

} .

s

~

T 9202n 0333 ,201. ,3-Wyl40!:B D ..E M

- [DR ADOCK 05200002

...3'.r 1 W-

"xm u. . L ' ;,5 , . . . .

O '

. J +' - S '.: t 1 J .

. -.a ' % ?%

L@@A AcpO3to cf tho Combustian Engineering Advanced Light Water Reactor - System 80 +

Abstract Introduction pumps are no longer needed in The C.E ADVR is an upgrade of the The paper provides analytical the ECCS. A comparison of the proven System 80 NSSS standard results of the response of the ECCS for System 80 and for design and so is referred to as "Sys- Combustion Engineering System System 80 + is 1;iven in Table 1.

tem 80+ ." floth plants are rated at 80 + NSSS to the Loss-of-Coolant Accident (LOCA). The purpose of LOCA Analyses 3817 Mwt, but System 80+ incor-these analyses is to show that: 1. The LOCA analyses performed to

3. orates a number of modihcations date includes investigations of to the design, for realistic pipe breaks (s 10 in.

dia.) using best estimate methods, both small break cnd large break A best estimate small bre.:k study the reactor core remains covered behavior of the teactor coolant throughout the transient; and 2, system. These studies are dis-addressed the economic concern of how large a break size can be toler. for large pipe breaks using licene cussed below.

ated without the two phase fluid ing methods, low pressure safety level falling below the top oi the injection pumps are not required smal! break study core, With a best-estimate analyti- for the new ECCS design with The purpose of the small break cal procedure and no single failure direct vessel injection, LOCA study was to determine if the core was shown to remain cov- the core could remain covered cred for breaks up to, and includ- Plant Changes Which Affoct with two phase fluid for the dura-ing, that of a 10 in. dia. break. LOCA Response tion of the transient following a The C-E System 80 + NSSS is an realistic size break, in order to A large break licensing analysis enhanced version of the NRC ap. evaluate the expected plant per-confirmed that adequate reflood proved System 80 NSSS standard formance in the unlikely event exists following the end of SlT design. Iloth plants are rated at that a LOCA should occur, best discharge without a LpSi system. 3817 Mwt. Some of the System estimate analytical procedures 80 + design changes that mfluence weie employed without a worst the response to a postulated LOCA single failure assumption.

include a larger heat transfer area in the steam generators and a Realistic breaks would, more larger pressurizer volume. The likely, occur in the long lengths emergency core cooling system of tributary piping than in rela-(ECCS) employs four (4) trains of tively short main cooling piping.

high pressure safety injection llowever, as a conservative simpli-(lips 1) pumps and safety iniection fication the breaks in this study, tanks (SITS) which iniect directly were located at the bottom of a into the vessel annulus instead of cold leg adjacent to the reactor the cold leg iniection for System vessel inlet nozzle, lireak sizes

80. Low pressure safety inicetion analy:cd are given in Table 2.

Table 1 Emergency Core Cooling System Features Icature Spirm 60 + System 80 High pressure safety injection 4 2 pumps (same characteristics)

Type of iniection Direct to Redundant headers vessel annulus to each cold leg Low pressure safcty injection none 2 purnps Safety injection tanks 4 4 Table 2 Small Break has investigated 0.05 ft2 (worst small break for System 80 licensing analysis) 0 20 ft2 (6 in. dia.)

0 55 ft2 (10 in. dia.)

1 y_

The analyses were performed rigore 1s. 0 06 ft? cold t eg nreak Large Dreak Study

- r~ ' {

with Combustion Engineering's 30 '

The pmpose of the large b.

improved small break LOCA '~] licensing i OCA study was to - .

computer program, CEFLAS114 AS 32 termine whether modifications t (fil). This program incorporates a l the safety inicetion system cook number of analytical models permit mw pressure safety inice-g #"

[

which arc more realistic than '

tion pumps to be climinated liom .

i those used in the cunent, approved a N---*-w-e the 1 CCS (equivalent puinps licensing methodology. The single f24 would still be required for derav most important model improve- I heat removalh These pumi s were ment in the small break analysis --

previously shown to be necessaty is the ANS 5.1 decay heat functie.a 5 20 for a bnef period of time in the with a two standard deviation un- current generation NSSS 1.OCA celtainty in place of the Appendix 16 l Active Core licensing (Appendix K) analybes.

K requirement of ANS 5 0 with a That time period occurs immedi- j 20% uncertainty factor. 12 _. ._ m .a _ _ _ m _ ately following the end of salety --

0 300 630 900 1200 1500 iniection tank discharge during Small Dreak Results hme m see- which the fluid loss rate from the The most important result for the core exeteds the net imeetion spectum of small h eak LOCAs rate from a smgle operating lil'SI with rs.spect to the potential for haute Ib. 0.2 ti? cold Leg Dreak pump. Such a situation is charac-fuel damge is the transient two- 36

"- ' ~- "' ~ ~ ] teri:cd by a decrease .a the liquid _ _ _ . _ .

phase level inside the core sup- level in the vessel annulus. __

port barrel These results are 32

)

shown in Figure 1 a, b and c for The improved LCCS design for the cold leg breaks of 0.05 ft2 ,0.2 ft 2 28 .

System 80+ pisnt pmvales a and 0.55 h2, respectively. These e greater amount of liPSI flowrate figures show that the core re- g ~ than do the 1.CCs for the cunent mains fully covered for the dura. T24 generatmn plants. This is particu-tion of the transient for all break I larly important following the end ,

sizes up to and including the .j 20 '-~~~T -

of SlT discharge. Table 3 presents a largest one analy cd (0.55 f t2 ). companson of the ECCS for Sys-j tem 80+ to that for a cuncut gen-The core should also remain 16 Acuve cme eration C L NSSS. The improved covered for breaks smaller than l lil'S! flowrate for System 80+.

those analyzed (less than 0.05 12 a i , m relative to cunent generation C L ft 21. This conclusion is based on 0 200 400 coo 800 1000 plants,is a result of the direct results for tN C-E System 80 Time m sec. inlectmn into the vessel annulus standard plant. System 80 has a and the use of four llPSI trains.

somewhat smaller RCS volume that does the System 80+ design Houre 1c. 0.551:2 cold L.eg nreak The large break LOCA analyses of the ALWR, and also a lesser 36 ' ' were run with C.E's currently ap-number of IIPSI pumps of similar proved licensing methodology size (see Table 1). 32 (Ilef.1).

28 6 __

.g24 I

1 2 20 r- l 16 ~ Active Core 12 O 100 200 300 400 500 Time in sec.

Fig.1: Two Phase Mixture Levelinside Core Support Darrel for Various

$1:e Cold Leg Breaks

- _ _ _ _ ____ ___ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ . I

The e,uditions shown in Table 3 Tabh 3 (loss n 'ne diesel generator) pro. Es ccriebiesty :e ocubi>tnded C:Id Lc; trnk (with loss of one diesel generator l duce the lowest LCCS injectmn flowrates. Ilowever, these condi- SWon bu + SWnn *0 tions do not necessarily result in sifs dehvering to vewel (net) 4 3' the highest peak licensing break 11 PSI pumps dehvening to vessel (net) clad temperature. That results wah loss of one diesel generator 2 3/4' from full FCCs flow with max' LPSI sumps delivering to vessel (netl imum spillage which decreases wah bw of one diesel-generator NA 1/2' containtnent pressute and worsens steatu hinding dCring core Icflood. .I and i N debery nue at m. empty time with low of one thesel generator iapprod 2260 GPM 3%0GPM Table 3 shows that the System 80 + plant with ducet injection low HNuned lot budou (as inning Lore to the vessel has, two (2) (net) '""d

D " " U M

I I N #M IN #

'l IIPSI pumps dehVermg watch

A portion of injectiott flow spills to wHtaimnent through the whereas, system 80 has three- cold leg break.

quarters (netl of one llPSI pmnp -_-- -

delivelitig water. This increment pIovida the necessary iniection flowiate to snatch the fluid los'. Table 4 ECCS Ucensing Results for ocuble-rate from the cose- Ended (cuittotinel preek Large Dreck Results gyI3Ni,3 The large break study was focused h^k N U"dd"""

on the double ended break of a ."""""* #

cold leg at the vessel inlet nonle, System 80 This break size and locatmn pro- tralo Verde) 2169 13 1 duces '.sults (peak clad tempera. System 60 +

--~

216s

'~

13 3 tutel which are worst or very close to the worst. Peak clad tempera-ture and local ondation results are given in Table 4. The principal Oc;ble Ended coid t.eg n,cak 38 result from the large break study is that the LOCA licensing criteria F

l-.

- ' ~ ' - ~~~~~]

can be met with the LCCS de- 15 i cold teg scribed in Tabh. 3 {ahove), in par. g timnon ticular, for the System 80 +

(12 ECCS, without LPS1 pumps, the :: n fluid level in the vessel annulus $ !!

rernains at its maximum value g 9N following the end of SIT dis- o d charp This is shown in Iigure 1

2. It is the liquid level in the j6~

vessel annulus which provides y the static head to force reflood 33 water up into the cme. A maxi-mum liquid level in the annulus o __ _ _ _

assures maxilnum flow of cooling 0 120 240 360 480 600 water to the core following the Tune in we.

large break LOCA. n o. 2 Loveiin oowncomer l

Conclusions Fleference These LOCA analyses for the C E 1. CLNPD-132P, " Calculation System 80 4 plant show that: Methods for the C-L Large

1. No core uncovery occurs for lireak LOCA Evahtation realistic pipe breaks and best Model," Augun, ML estimate methods and 2 A LPSI pump is not required in the design to meet licensing criteria.

3

0"ention 720.6 In the logical expression of SGTR accident sequences provided by CE (floppy disk), five accident sequences (i.e., sequences 3, 4, 7, 9 and 10) were found to contain the complement of an event-tree top event, PMilG04 BX . Ilowever, no such top event !

can be found in the SGTR euent tree. Please explain this inconsistency.

Renp.pjlse 720.6 The fault tree, PMHG04BX, is defined as " Failure to Deliver Sufficient SI Flow to 4 of 4 Loops" (See page 6-277 in the System 80+ PRA Report). It is part of the " Failure of SIS for Medium LOCA" fault tree on page 6-274 of the System 80+ PRA Report. The complement of PMilG04BX was inadvertently used instead of the complement of the fault tree, PilOG01BX , which can be found on page 6-279 of the System 80+ PRA Report.

PilOG01BX in defined as " Failure to Deliver SI Flow to 4 of 4 Loops - No Break". The primary difference between PMilG04BX and PilOG01BX is that PilMG04BX does not include the common cause failures of the mechanical components in the Injection System (They are included at a higher level in tree PilMG02BX on page 6-275 of the System 80+ PRA Report). This difference doe not impact the analysis because the downstream faulted events in SGTR sequences 3, 4, 7,9 and 10 do not include any of the mechanical equipment in the Injection System.

I l

, - _ - . - . - . . . - . . . . - - - . . . . . . - - - . ~ . - . . . . . - . - . . - . . . . . . - - . ~ . ~ . . . - - . ~ . . . . - . .

?

Ouestion-720.9 In the event of a SG.'R, many operrtor actions are needed to prevent core damage, They include manual control of EFW to maintain proper le' e1 in the intacc SG; manual tripping of two out of 4 RCPs and after identifying the ruptured SG, isolation ' tne ruptured SG by c)osing MSIVs, ADVs, and main feedwater ' ? solation valves a n-l by isolating SG ~ blowdown, vents, drains, exhausts and bleedoffs. Are the required operator actions modeled i". the fault trees? If yes, in which fault trees?

Ecoponse 720,9 See the response to Question 721.1.

l-4 l

l l

I '

i 1

i H

l

. . ._ _ _ _ . m.- m _ .._ .... _ ._ _ _ __._ .-_._.-___ ___ _ .

Ouestion 72 M Please indicate where the accident sequence-involving a SGTR with cojncident loss of off-site power is modeled in the -pRA?

- If it -1s not modeled, please model it in the PRA or provide an

-explanation for why it need not be modeled. Note that, for this _ sequence, turbine bypass system, MFW system and SG Blowdown system will not be available. Moreover, all RCPs will trip and pressurizer spray will :not be available for lowering KCS pressure. Heat removal by SG will have to rely heavily on primary loop natural reeirculation and it may take a longer time to bring the reactor to stable cold shutdown.

Resnonse 720.20 All of-the frontline system fault tree-models in the SGTR event tree have all of tt1 appropriate support system models linked into them. The electrical distribution system fault l tree models include coincident loss-of off-site power. Thus, all- of - the SGTR sequences include consideration of; a coincident-loss of off-site power. With-a loss-of offsite power, the-RCPs trip and the main - pressurizer spray is not available for RCS pressure control. Heat removal via the-SGs relies on natural circulation in the primary loop. The I

reactor - head vent roubsystem portion of the safety depressurization system is available for RCS pressure control.

Extra time is required for the cooldown on natural-

, circulation. : This extra time was - assumed in the timing l analysis. The models for the Startup Feedwater System and the -

Steam-Generator-Blowdown System address the availability of offsite power. The Turbine Bypacs System was not credited in the analysis, with or without offsite power. In this analysis, a SGTR with a coincident full station blackout (loss of offsite power with failure of both diesel generators and the E Alternate AC -System) was assumed to lead to core melt because of the loss of RCS makeup capability.

l l

i c _ . . . - - _ _ . - _ . -. .- - --

l l

Dungtion 720.30 For an ATWS with a stuck open PSV, but with successful safety injection (aco ATWS sequence #24) cr for an ATWS with failure of boron delivery by the charging pumps, but with successfu;

der curization and safety injection (soo ATWS sequencu #7),

the -t load of the IRWST can be expected to be algnificantly highn_ than for ordinary transients due to high reactor power.

What is your basis for aruuming that one containment spray j pump in adequate to successfully cool the IRWST?

Rennonso 720.,19 ;

i The primary purpose for cooling the IRWST is to comove tho >

residual heat from containment to provont a containment overpressurn failure. The containment spray system in !

designed such that one containment spray pump and its

, associated containment spray heat exchanger can removo sufficient energy from containment following a design basis large LOCA or Steam Lino Break to provent excooding the !

containmont design pressuro For an ATWS with an MTC that is ,

not adverso, the reactor power initially increanos. However, as the RCS temperature and pressure, the reactor power rapidly begins to decreano duo to the moderator temperaturo reactivity ,

foodback. The ATWS pressuro peak is past with approximately 1 or 2 minutos with reactor power dropping to about 5% shortly thoroafter. During this porlod, energy is being discharged from the RCS to the IRWST via the Primary Safety Valves (PSVs). Subsequent energy transfer would be via oither the stuck open safety valve or the depressurization valvo (s) .

Thus, tho-initial rato at which snorgy is transferred into containment from the RCS is less than for t. design basis LOCA or Steam Lino Break. With the boron add' on associated with the successful safety injection,- react ot power will continuo to decrease to decay heat levels. Thus, while the. total amount of energy added to containment (IRWST) following an !

ATWS with a stuck open PSV or successful depressurization may be slightly greator than for a design basis LOCA due to the higher initial power, this energy while be transferred to-containment at a lower rato and over a greator period of time.

Based on this, it was concluded that one containment spray i pump anu its. associated heat exchanger could provido sufficient heat removal from the IRWST to provent containment overpressurization.

4

_ . - - - , _ . , ~, ..-.,..~n._.__.2.. _.,._._ _ ._ _ . ___._.-._ _. ._._._... ,.._ --- _. - _ ~ _ . . - ~ . -

r Question 720.33 Describe in detail how reactor coolant pump (RCP) soal LOCAs are treated in the CESSAR 80+ PRA? ,

Responso 720.33 RCP seal I4CAs are not troated explicitly in the System 80+ l PRA. As an initiating ovent, RCP soal LOCAs are considered to 1 be covered by the small LOCA event troo. The gonoric !

' initiating ovont frequency presented in the EPRI ALWR PRA Key

Assumptions and groundrules was based on all operating events which had a leak rate or loah sizo consistent with small LOCAs. Thus, RCP soal LOCAs can be considorad to be covered by.

this initiating ovent frequency. The system responses nooded to mitigato an RCP seal 14CA are the sa'a as for any othor- 1 small 14CA.

1 Based on operating experienco and test data, au presented in CE HPSD-340, "A Combustion Engineering Review of NUREG-1032,1 Evaluation of Station Blackout Events at Nucioar Power Plants",. March, 1986, CE believos that the RCP seals used in CE plants will not develop oxcessive coal lonkage under total S*.ation Blac);out conditions (SBO) . Total S00 is defined here :

as a loss -- of offuito power combined with failure of both dies (1 generator and the alternato AC sourco. Thorofore, consequential RCP seal LOCAs following a station blackout were not modoled in the System 80+ PRA. .

f i

,.m .. y- -%i-w_

Question 720.35 List the top 100 accident sequences where recovery was credited. Was recovery credited before or after sequence truncation? for each of the accident sequences that has been corrected for recovery actions, please indicate which safety functions were corrected and what recovery factors were used. Was recovery credited at the cutset leveli If so, how many cutsets were involved?

Bu ngase 720.35 Table 720.35-1 (starting on next page) lists the accident sequences where the recovery was credited. The list comprises of 38 internal accident sequences and 11 seismically-induced accident sequences. Recovery was not credited in any tornado strike accident sequences because it was assumed that the offsite power could not be restored within the 24 hour2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months mission time and that the alternate AC )ower source, namely combustion turbine, would not be available as a baccup for the diesel generators. Table 720.35-1 indicates the safety functions that were corrected and/or recovery action (s) taken for a given accident secuence. The recovery factors that were used are to be found in terms of tie non-recovery probabilities in Table 5-7 on page 5-50 of the System 80+ PRA Report (DC1R-RS-02, Rev. O, Volume 1. January 1991). The recovery was credited at the cutset level. Recovery was applied to the point where the total contribution of the cutsets was at least 95% of the core damage frecuency attributed to that particular sequence and the contribution of the individual cutset was less than 0.1%. Accordingly, the number of cutsets involved in the recovery process varied from one accident sequence to another sequence.

Question 720.36 Please provide dependency matrices showing major frontline system and t- ;1r .

dependence on all the relevant support systems at the train level. Provide a similar matrix for support system dependency at the train level.

Response 720.36 C-E is currently updating the System 80+ PRA. The dependency matrices requested above will be included as a part of this update.

l i

1 - _ - - . . . . - _ - -

Table 720.35-1 : Ascident Stquences were Recovery was Credited

=-

Table no. in DCTR-RS-02 Accident Vol.3 where Safety function (si Corrected and/or Sequence Description Recovery Action (s) taken ,

is given _ _ _ ,_

2-Large LOCA 8.2.1-1 Containment Cooling; Open manual discharge valve (s) 2-Hedium LOCA 8.2.2-1 Containment Cooling; Opes manual discharge valve (s) 3-Hedium LOCA 8.2.2-1 Start and Load Standby AC power or Restore Offsite power within 12 hours1.388889e-4 days 0.00333 hours 1.984127e-5 weeks 4.566e-6 months to provide motive power to Engineered Safety feature (ESF) pumps 7-Small LOCA 8.2.3-1 Hanually rackin equipment breakers to provide 125 VDC :ontrol power for the Engineered Safety feature (ESF) pumps 10-Small LOCA 8.2.3-1 Start and Load Standby AC power or Restore Offsite power within 12 hours1.388889e-4 days 0.00333 hours 1.984127e-5 weeks 4.566e-6 months to provide motive power to Engineered Safety feature (ESF) pumps ll-Small LOCA 8.2.3-1 Start and Load Standby AC power or Restore Offsite power within 12 hours1.388889e-4 days 0.00333 hours 1.984127e-5 weeks 4.566e-6 months to provide motive power to Engineered Safety feature (ESF) pumps 10-SGTR 8.2.4-1 Hanually rackin equipment breakers to provide 125 VDC control power for the Engineered Safety Feature (ESF) pumps 12-SGTR 8.7.4-1 Start and Load Standby AC power or Restore Offsite power within 12 hours1.388889e-4 days 0.00333 hours 1.984127e-5 weeks 4.566e-6 months to provide motive power to Engineered Safety Feature (ESF) pumps

(

8.2.4-1 Hanually rackin equipment breakers to 13-SGTR provide 125 VDC control power for the I

Engineered Safety Feature (ESF) pumps; Start and Load Standby AC power to provide motive power to Engineered ;

i Safety Feature (ESF) pumps ,

l 7-LSSB (Large 8.2.5-1 Start and Load Standby AC_ power or Secondary Restore Offsite power within I hour l Side Break) to provide motive power to Engineered

, Safety Feature (ESF) pumps l

Table 720.35-1 : Accident Seouences were Recovery was Credited ;roiA rt; 1 Table no. in DCTR-RS-02 Accident Vol.3_where Safety function (s) Corrected and/or Sequence Description Recovery Action (s) credited

_._Lsgiven ,

17-LSSB 8.2.5-1 Start and Load Standby AC power to provide motive power to Engineered Safety Feature (ESF) pumps 4-LOFW (Loss 8.2.6-1 Start and Load Standby AC power or of[ main) Restore Offsite power within 9 hours1.041667e-4 days 0.0025 hours 1.488095e-5 weeks 3.4245e-6 months Feedwater or Restore Offsite power within 16 Flow) hours to provide motive power to Engineered Safety Feature (ESF) pumps 8-LOFW 8.2.6-1 Start and Load Standby AC power or l Restore Offsite power within I hour '

or Restore Offsite power within 4 l hours to provide motive power to Engineered Safety Feature (ESF) pumps 9-LOFW 8.2.6-1 Hanually rackin equipment breakers to provide 125 VDC control power for the Engineered Safety Feature (ESF) pumps; Start and Load Standby AC power or Restore Offsite power within 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months to provide motive power to Engineered Safety feature (ESF) pumps 8-TOTH (Other 8.2.7-1 Start and Load Standby AC power or Transients) Restore Offsito power within 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months to provide motive power to Engineered Safety Feature (ESF) pumps 9-TOTH 8.2.7-1 Hanually rackin equipment breakers to provide 125 VDC control power for the Engineered Safety Feature (ESF) pumps; Start and load Standby AC power or Restore Offsite power within 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months to provide motive power to Engineered Safety Feature (ESF) pumps

Table 720.35-1 : Accident Sengentet_were Recovery yas Credited (cont.d)

Table no. in DCTR-RS-02 Accident Vol.3 where Sa'ety function (s) Corrected and/or Sequence Description Recivery Action (c) credited

.1s, _g.iven __

3-l00P (Loss 8.2.8-1 Start and Load Standby AC power or Of Offsite Resto.e Offsite power within I hour or Power) Restore Offsite power within 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months or Restore Offsite power within 16 hours1.851852e-4 days 0.00444 hours 2.645503e-5 weeks 6.088e-6 months to provide motive power to Engineered Safety Feature (EST) pumps 7-LOOP 8.2.8-1 Start and load Standby AC power or Restore Offsite power within I hour or Restore Offsite prwer within 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months to provide motive power to Engineered Safety Feature (ESF) pumps 9-LOOP 8.2.8-1 Start and Load Standby AC power or ,

Restore Offsite power within 1 hour1.157407e-5 days 2.777778e-4 hours 1.653439e-6 weeks 3.805e-7 months or Restore Offsite power within 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months to provide motive power to Engineered Safety Feature (ESF) pumps ll-LOOP 8.2.8-1 Start and Load Standby AC power or Restore Offsite powe) within I hour to provide motive power to Engineered Safety feature (ESF) pumps 12-LOOP 8.2.8-1 Start and Load Standby AC power or Restore Offsite power within 12 hour1.388889e-4 days 0.00333 hours 1.984127e-5 weeks 4.566e-6 months s-to provide motive power to Engineered Safety feature (ESF) pumps 3-CCWB (Loss 8.2.9-1 Open mant&1 discharge valve (s) of Component Cooling Water Div.)

4-CCWB 8.2.9-1 Start and Load Standby AC power or Restore Offsite power within 16 hours1.851852e-4 days 0.00444 hours 2.645503e-5 weeks 6.088e-6 months to provide motive power to Engineered Safety Feature (ESF) pumps

_,---...m... . - - . - - ,,x,.., .-., . , _ .....y.- r._,_, , , ,.-.<rn. .,_,.,_..,n, --#w._--.,r , , .,r ,- vr

Table 720.35-1 : AccMenLituenttLwen R e cay er_Y_wn_Cr.e dhed ( c o n t . d )

Table no in DCTR-RS-02 Accident Vol.3 where Safety function (s) Corrected and/or Sequence Description Recovery Action (s) credited isgiven_.,,_ , . . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

5-CCWB 8.2.9-1 Open manual discharge valve (s);

Start and Load Standby AC power or Restore Offsite power within 16 hours1.851852e-4 days 0.00444 hours 2.645503e-5 weeks 6.088e-6 months to provide niolive power to Engineered Safety feature (ESF) pumps 8-CCWB 8.2.9-1 Start and Load Standby AC power or Restore Offsite power within 1 hour1.157407e-5 days 2.777778e-4 hours 1.653439e-6 weeks 3.805e-7 months or Restore Offsite power within 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months to provide motive power to Engineered Safety Feature (ESF) pumps 8-125VB (Loss 8.2.10-1 Start and Load Standby AC power or of 125VDC Restore Offsite power within 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months Bus B) to provide motive power to Engineered Safety feature (ESF) pumps 9-125VB 8.2.10-1 Manually rackin equipment breakers to provide 125 VDC control power for the Engineered Safety Feature (ESF) pumps 8-416KB (Loss 8.2.11-1 Start and load Standby AC power or I

of 4.16KV Restore Offsite power within 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months Bus B) to provide motive power to Engineered Safety feature (ESF) pumps 9-416KB 8.2.11-1 Hanually rackin equipment breakers to provide 125 VDC control power for the Engineered Safety feature (ESF) pumps; Start and Load Standby AC power or Restore Offsite power within 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months to provide motive power to Engineered Safety Feature (ESF) pumps 7-ATWS 8.2.12-1 Start and Load Standby AC power to provide rptive power to Enginecred Safety Feature (ESF; ,purops 8-ATWS 8.2.12-1 Start and Load Star $y AC power; i

Start and Load StanLy AC power or Restore Offsite power within 9 hours1.041667e-4 days 0.0025 hours 1.488095e-5 weeks 3.4245e-6 months to provide motive power to Engineered Safety Feature _[Lgpumps

Table 720.35-1 : Accident Secuences were Retoyery was credited (cont.d)

Table no. In DCTR-RS-02 Accident Vol.3 where Safety function (s) Corrected and/or Sequence De;.cription Recovery Action (s) credited is given 4

9-ATWS 8.2.12-1 Start and Load Standby AC power to provide motive power to Engineered Safety Feature (ESF) pumps 18-ATWS 8.2.12-1 Start and Load Standby AC power; I Start and Load Standby AC power or ;

Restore Offsite power within 12 hours1.388889e-4 days 0.00333 hours 1.984127e-5 weeks 4.566e-6 months to provide motive power to Engineered Safety Feature (ESF) pumps l 25-ATWS 8.2.12-1 Start and Load Standby AC power or i Restore Offsite power within 12 hours1.388889e-4 days 0.00333 hours 1.984127e-5 weeks 4.566e-6 months to provide motive power to Engineered i Safety Feature (ESF) pumps '

3-LHVAC (Loss 8.2.14-1 Open manual discharge valve (s) of I HVAC division) '

l 4-LHVAC 8.2.14-1 Start and Load Standby AC power or r Restore Offsite power within 12 hours1.388889e-4 days 0.00333 hours 1.984127e-5 weeks 4.566e-6 months j to provide motive power to Engineered

Safety Feature (ESF) pumps S-LHVAC 8.2.14-1 Open manual discharge valve (s);

Start and Load Standby AC power or Restore Offsite power within 16 hours1.851852e-4 days 0.00444 hours 2.645503e-5 weeks 6.088e-6 months to provide motive power to Engineered Safety Feature (ESF) pumps 8-LHVAC 8.2.14-1 S! art and Load Standby AC power or Restore Offsite power within I hour or Restore Offsite power within 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months to provide motive power to Engineered Safety Feature (ESF) pumps k *

(Seismically-4-SEIS 8.3.2-1 Isolate failed motor jacket cooling heat exchanger for motor-operated induce event) pump _ i l

l

Table 720.35-1 : Accident Seguences were Recovery was Credited (cont.d)

Table no. in DCTR-RS-02 Accident Vol.3 where Safety function (s) Corrected and/or Sequence Description Recovery Action (s) credited is given 5-SEls 8.3.2-1 1solate failed motor jacket cooling heat exchanger for motor-operated pump 8-SEl$ 8.3.2-1 Isolate failed motor jacket cooling heat exchanger for motor-operated pump t 9-SEIS 8.3.2-1 Isolate failed motor jacket cooling heat exchanger for motor-operated pump; Reclose breakers to provide power to the equipment 12-SEIS 8.3.2-1 Isolate failed motor jacket cooling heat exchanger for motor-operated pump; Reclose breakers to provide power to the equipment 15-SEIS 8.3.2-1 Isolate failed motor jacket cooling heat exchanger for motor-operated pump; Reclose breakers to provide power to the equipment 24-SEIS 8.3.2-1 Isolate failed motor jacket cooling heat exchanger for motor-operated pump 28-SEIS 8.3.2-1 Isolate failed motor jacket cooling

' heat exchanger for motor-operated pump 29-SEIS 8.3.2-1 Isolate failed motor jacket cooling heat exchanger for motor-operated pump; Reclose breakers to provide power to the equipment 32-SEIS 8.3.2-1 Isolate failed motor jacket cooling heat exchanger for motor-operated pump; Reciose breakers to provide power to the equipment Seismically- 8.3.2-1 Isolate failed motor jacket cooling

, induced SB0 heat exchanger for motor-operated with Battery ) ump; Tie non-vital batteries to vital

depletion satteries within 8 hours9.259259e-5 days 0.00222 hours 1.322751e-5 weeks 3.044e-6 months

Q nstion 720.38 By using the IRRAS input data provided by CE (on floppy disk), the staff requantified 65 out of the 67 zero-level fault trees. The two f ault trees that could not be quantified (gave error messages) are POLX0 LEX and POLX0lHX, both involving long-term jecay heat removal. Please provide us ,

a corrected floppy disk. ]

)

Response 720.38 !

The fault trees POLX01EX and POLX0lHX, involving long-term decay hett i removal with loss of 125 Vdc bus and 4.16 KV bus, respectively, were too- i large to run directly in IRRAS becat se of computer memory (RAM) limitation, t Thus, following procedure, which is illustrated by en example, was employed '

to quantify these fault trees. ;

The fault tree POLX02EX was run as a sequence consisting of the fault trces PJOB0lRE, failure to establish residual heat removal (RHR) or

~: ahutdown cooling flow for long-term heat removal with 125 VDC bus B unavailable, PA0G0lME, failure to deliver emergency feedwater (EfW) with 125 VDC bus B unavailable toeithersteamgenerator(SG)tartupfeedwater')eitherSGwland PMSA0lME, failure to deliver s th-125 VDC bus B unavailabic.

HR2,1258 PJOB0lRE 4 PA0G0lME & PMSA0lNE 4 The_cutsets for this sequence were obtained using 1RRAS.

Similarly, the fault tree POLX03EX was run as-a sequence consisting of the fault trees PJOB0lRE, failure to maintain RHR-or shutdown-cooling flow for long-term heat removal with 12S VDC bus B unavailable, and PA0G0lRE, failure to restart EfW system for long-term heat removal with 125 VDC bus B unavailable.

HR3_12SB - PJOB0lME 4 PA0G0lRE-4 The cutsets for this sequence were obtained using.IRRAS.

The cutsets obtained for the sequences HR212SB and HR312SB were then processed through a proprietary uttifty code to ellminate the duplicate and nonsense cutsets,=and then combined to obtain the cutsets for the fault tree POLX01EX. Finally, the fault tree POLX01EX was quantified by loading the cutsets for POLX01EX as a fault tree in IRRAS.

Similar procedure was followed to quantify the fault tree POLX0lHX.

The final cutsets for POLX01EX and POLX0lHX are provided on the flop >y disk as POLX01EX ' I and POLX0lHX CUT (in ASCil format as required by :RRAS).

l

I Question 72QJ.1 ,

Please provide us a floppy disk containing the IRRAS input data for calculating the core damage frequencies attributable to tornado strike events. '

l

, Reinante 720.41 !

IRRAS was not used to calculate the core damage frequencies attributable :

to tornado strike event sequences. However, the cutsets for tornado strike '

event-sequences from which the core damage frequencies were calculated are I presented in Tables 8.2.1-2 through 8.2.1 8. ;

)

I h

, e i

t i

I i

F i

1

'l 4

f u .

l 2,.. ._.,..___.._2._..___. . _ . . . _ . . . . _ . . . . . . . _ . _ . . _ . _ . _ . _ . _ _ . . _ . . _ . _ _ . _ . . = _ _ . _ _ . . _ _

Ouestion 720.45-Provido a comparison of the sequences in the CESSAR 80 PRA leading to (at least) 90 percent of the estimated coro damage f requency with the corresponding sequences from the CESSAR 80+

PRA. Discuss the reasons for the improvement /worsoning of the estimated coro damage frequency in the corresponding 80+ PRA sequences.

Reappnne__720.45 Tablo 8.1-4 in the System 804 PRA Report, DCTR-RS-02, Rev. O, January, 1991, providos a comparicon betwoon the vutimated !

core damage frequency for System 80 and the estimated coro i damage frequency for System 80+ at the initiator level. The l attached slidos present the equivalent comparison at the '

-initiator level with an assosement of the System 80+ design i enhancements that resulted in the reduction in the ostimated core damago frequoney.

' Question 720.46 Describe how loss off offolto power (LOOP) is modeled for soismio events (e.g., in the seismic event troos dm'oloped in figuro 7.3-3).

Eenponse 720.46 j It was assumed that all seismic ovents resulted in-a loss of power I

offsite lasting longer than 24 hours. Tho unavailability of the offaite power sources was reflected in '

the electrical distribution system models. Use of the Standby Alternate AC source was not credited because it is not seismically qualified.

- , ~ . _ __ _._.....__a.a . . ._ . _ . , . _ , , , . _ , . . , _ - . , . - _ , _ . _ _ . . . . , _ , _ , _ _ , - . . . . . = , . _ _ , . _ , _ . , _ . , _ _ . , , , _ , _ . , . _ , _ , , , -

03 t

)

RISK _REDEING_EEAIURES_ EOR _DMlIHMLSE0llENCE Ill I

.SE0l!ENCE TYPE -

LOSS OF 0FFSITE POWER (LOOP)

INCLUDING STATIO!1 BLACK 0uT WITil l BATTERY DEPLETIOil REEE ESERTAIIYLDQtillMNT .S E00EtLCE (LOOP) (FAILURE OF ER4) l FRE0_U_EEX l OLD 4E-5 NEW 7 EEATURES .

O ALTERi4 ATE AC POWER S0uRCE (GAS

, TURBIllE) o SEPARATE OFFSITE POWER S0uRCE T1!AT BYPASSES TIIE SWITCllYARD

, o DEDICATED BATTERY FOR EACil DIESEL GENERATOR o F0uR TRAIN Ef4ERGENCY FEEDWATER (TWO wITil TURBINE DRIVEN Puraps o TURBINE GENERATOR ABLE TO Rutt BACK TO HOTEL LOAD, m SYSTEM & # -

. 84 I

RISK REDUCING _EEAIURES FOR DMINANT SE0VENCE_32 SEMENCE TYPE - TRANSIENTS REPAESENIAUYE.10MINAHLSE011EHCE !

(LOFW) (FAILURE To DELIVER Ef4ERGEllCY Bl)

EREQUEUCY l OLD - it-s NEM - n-c -

EEATURES o FouR TRAIta Er4ERGEllCY FEEDWATER SYSTEl4 o REDUllDAriT SOURCES OF Er4ERGEllCY FEEDWATER

., - 2 ER4 TAf4KS

- CONDENSATE STORAGE tai 1KS '

o llIGN RELIABILITY Co!4PONENT Co0 LING ,

SYSTEl4

- Two PUl4PS PER IRAIN

- NORl4 ALLY RUNNIt1G o START-UP FEEDWATER SYSTEli

- FRori CONDENSATE STORAGE TAtlK

- ACTUATED BEFORE ER4 o Futt RUN-DACK CAPADILITY o Two ER4 ACTUATIot1 SYSTEf4S

- REDUNDANT

- DIVERSE SYSTEM &#

Y

05 RISLREDJLCIRG FEERES FOR _DOMItintlLSE00ERCE_/L3 SE0l!ERCETYPE - STEAM GENERATOR Tune RuptunE REPRESENTATI1FJJ0MINANT SEORENCE -

(SGTR) (FAILURE To DELIVER EFW)

(SGTR) (FAILURE OF SAFETY IllJECTIOll)

EREDEEf101 OLD - ir-s NEW - et-a FEATU_RES i o Foun TRAIN EMERGENCY FEEDWATER SYSTE!4 o Foun TRAIN SAFETY INJECTIOli SYSTEM

~

i o SAFETY DEPRESSuRIZATION SYSTEM

t i~

SYSTS1$$#

E6 l

RISK _REDFDlG_EEAERES_ EOR _D01RUANT SEQUEt[CE_TIEE Il4 l

SEQUEL [CE_TlEE - St4ALL LOCA REPRESEl[ TAT _IVE_D M HANT SE0lLEtLCE l

' (Sf4ALL LOCA) (FAILURE OF SI RECIRCULATIO!1)

(S!4ALL LOCA) (FAILURE OF SI IllJECTIOld EREQUENC1 OLD - 9t-6 NEW

4E-8

.EEATURES O IN-CONTAINi1ENT REFUELING WATER l STORAGE TANK O F0uR TRAIll SAFETY INJECTION SYSTEM ;

0 ELIMINATION OF RAS O SAFETY DEPRESSURIZATION SYSTEM l

( -

SYSTEM &#

nt 1

t i

87

\

i l

RISK REDRIRG FEollRES FOR_ D_0LIINAHI_ SEQUEL [CE lL5

_SE_0UENCE TYPE - ATMS REERESENIAIIYE_D0LiINANT SE0Vft[ CEES l (ATWS) (ADVERSE MTC) -

FREQUENCY ,

OLD 6 NEW - 2E-7 FEATURE _S

., o LARGER PRESSURIZER 0 LARGER STEAM GENERATOR o SAFETY DEPRESSURIMTION SYSTEM O DIVERSE PROTECTION SYSTEM SYSTEM &#

t s

Ouestion 72M i i

Please provide us a floppy disk containing the IRRAS input data for calculating all the seismic core damage sequences delineated by the j seismic event tree presentoo by figure 7,3 3. .

3 i

Response 720.47 -

The core damage frequencies attributable to seismic event sequences were calculated by using the Seismic Integration Program (SIP) code, flowever.

the cutsets for seismic event sequences from wnich the core damage frequencies were calculated are presented in lables 8.3.1 ? through 8.3.1-15. '

i I

1 I

t A

v b

+

1 l

e t

t

- _ . . . . . - - , . . . . - . - . . - . - - . . . , . ~ . . . , - , . . - , - . _ - . - , . . , , - _ - . . , , - . . . , . - . - - - - . - . . , - , . . . . . . . . . - . . ,

CV.cf119.11 220_tAE Please provide the lint of random failure probaDilities an Well as the fragility data of all the reicmically-induced banic events used in quantifying the neintric fault treen and event treen, llc D E O D E e 2_2 f t4.Q The random failure probabilition for banic events used throughout thin PRA are presented in tablen S-2 through 5-7 of the System 80+ PRA Report (DCTR-RS-02 Fev. O, January, 1991).

The fragilities for noismically iraduced basic eventa used in thin PRA are precented in table 7.3-2 of the PRA Report.

6 f\

f f<

\

P' f

1 1 _ .l

i Questior. 720.49 l Although seismic failures of the containment spray system (CSS /RHR) hoat exchangars for long-term decay boat removal are considernd in the seismic fault treo analysis, such failuros, however, do not includo a heat exchaugor pipo break that could i drain the contents of the IRWST and load directly to core damage. Justify omitting such an accident scenario when estimating the soismic coro damage frequency.

ECAnonBA 7.2.0.4_12 This accident sequence was .not included in the System 80+

soismic PRA. C-E is currently updating the System 80+ PRA and this seismic accident sequence will be ovaluated, but it i s, not expected that the results will be significantly affected.

Question 720 11 A closo examination of the soismic event analysis performed in the System 80+ PRA revealed that virtually no consideration was given to possible failuros of important plant structuros that may onsuo from a soismic event. Please discuss tho ;

possible impact of such omissions on the coro damage frequency estimates. In this connection, also provido the System 80+-

specific fragility paramotor calculations for the following structures and components: containment, reactor vessel, reactor internals, reactor-coolant piping, pressurizer, turbine building, main control room (including control room suspended colling, if any), condensor hotwell, omorgency '

feedwoter tanks, foodwater hoe rors, CRD guido tubos, CRD housings and fuel assemblios. ;

Rosponsg_220.,11 ,

An Advancod Reactor Sovero Accident Program (ARSAP) contractor performed the initial portions of the System 80+ Soismic PRA.

Part of those analyses. as describod-in section 7.1.8 of the System 80+ PRA Report, DCTR-RS--02, Rev. O, January,1991, and

.in~ Reference 6G of the PRA report, was a qualitative assessment of the design features of System 80+ as compared to the design features of plants for which a detailed seismic PRA had_ boon performed. . Based on this review, it was datormined that the seismic capacity of the plant structures comprising the nuclear annex - would be in excess of 3 9 Thorofore, failure-of thoso-structures was not addressed in the soismic PRA. Equipment in structures outside of the nuclear annex which might have lower seismic capacities were not credited -in the seismic analysis. System- 80+ specific fragility parameters woro not calculated for the seismic PRA. Thoso calculations will be based on as procured-information as part of the detaiJ ed plant' design phase. The component frag 111 tics

llcanonne 720.51 (Cont.)

Assumptions and croundrules. Those fragilities are considorod to be achievable. It is bolioved that any uncertainty in the assumed fragilities is overshadowed by the uncertainty in the seismic hazard curve.

Q\tention 720.52 The staff believes that firos and internal floods can bo significant contributors to estimated coro damage frequency.

Pleano provido a firo PRA and an internal flooding PRA.

RoSDonne 720.52 As discussed in sections 7.1.6 and 7.1.8 of the System 80+ PRA Hoport, DCTR-RS-02, Rov. O, January, 1991, internal firos and internal floods woro not considered to be significant risk :

contributors for System 80+ because of the high degroo of separation and compartmentalization used in the System 80+ ,

design. C-E is in the process of updating the System 80+ PRA. l C-E will perform a more dotalled ovaluation of the risk !

potential for internal firos and floods as part of this-PRA l updato. In addition, if the results of the qualitativo !

analyson indicate that quantitativo analyson are warranted, ,

the quantitativo analyson will also bo included -in the updated !

PRA.

t ouestion 720.53 On page B-67 of the CESSAR, Appondix D, it is. stated that '

"fa:,1ure of a PSV to rescat af ter the primary sido prosauro decreason will result in a small Loss-of-Coolant Accident F (LOCA) with offsito poWor unavailablo.- This-is considered to be a small LOCA initiator for cuantification of small LOCA :

frequencies." Essentially ident;. cal statements are also mado on pago D-144 for Tornado striko sequence analysis. Woro those sequences actually transferrod to the small-LOCA ovent troo? If so, what are their frequenclos?

Response 720.53 ;

Tho statomonto regarding PSV LOCAs following a LOOP or tornado ;

striko initiator are incorrect. Thoso " consequential" small .

i-LOCAs were treated within the appropriato event troca as chown )

in- figuros- D3.1.8-1 and D4.2.3 (prosented hero for convenienco). The statomonts on pages B-67 and D-144 woro ,

inadvertently loft in from a previous itoration of the-PRA.

C-E is in the process of updating the System 80+ PRA and those statomonts will be doloted. '

l c

P

.-,..y, s . - . . . . . . , ~ . , . ._,_,..,.,m,,_,. ,_.,m, ...--..~._,...,-._,,,._,,,.-.~,...-,.,,.,7_ , . , , , , _ , . _ . _ . - , , . . . . _ _ . , . , .

m W U U U m V U U m U U

- N m

w e e- ce m S "

01 3!$9 5

&d5 yg _= _

3-

~a t is C

.$ A

6. -

E5 N&S

ch :

.h f

.hY a _.

k a;

>0 E$

t>

bb p C. hnendment f C Decunber 15, 1989 Figure

,gyJ7g / Loss of Offsite Power Event Tree B3.1.8-1

m m V U U m O O U m O U

- r4 m .s n e 6 m e 3 % U

~~~~

f5 l

5

_Q . - - - - - . - - . -_ _ -

t

^dE Ji r-x l

l Gee ,

1 H

QO-S O -

xb b 3 s'

.h :i ) .

B

.kG st j'C G

>C _ _ _ . -__

$,2 e c Amendment H August 31,1990 Figure

/ Tornado Strike Event Tree (34.2.3-1

l DitCAti.on 7 2 0. 5_4_

On pago 6-522 of the System 80+ PRA, n is stated that the electrical power for the CVCS (chemical and volume control systom) is supplied by tho 4.16 kV non-class 1E, 125 VDC non-class 1E, and 480 VAC non-class 10 power systems. More specifically, it is stated that the 4.16 kV non-class 1E power system ;,rovidos the motive power to operate the charging pumps, the 480V non-class 1E Load Contor provides the motivo power to operate the boric acid makeup pumps, and the 480 VAC HCC system providos the motivo power to operato the AC motor-operated valves.- l'ocussing attention on the ATWS sequences i dolineated on the scismic event treo (Figure 7.3-3, shoot 2),

which presumably assumos a concurrent lons-of-offsito power t (LOOP), it is not clear whether the omorgoney dicsol can supply power to the CVCS. If not, how can the event-tree top event, "Doliver Doron" be achieved under LOOP conditions?

Ecpj?onse 720.S_4 C-E agroos that the amorgency diosolo cannot provido power to

-the_4.16 kV permanent non-safety busos for a seismic event ;

becauso the permanent non-safoty busos are not ooismically quallflod and are assumed to be unavailable. Thus, the CVCS :

can not be used to "Doliver 13oron" for ATWSs with a concurrent loss-of-of f sito power (LOOP) . The solamic ATWS sequences cro modelod incorrectly. IIPSI pumps would bo required for boron

delivery to provido long term reactivity control. The primary system would~ have to be depressurized using the depressurization valvos-before the !! PSI pumps could be used for boron delivery. C-E is currently updating the System 80+

' PPA to reflect some system design changes. The seismic ATWS sequences will be corrected in the revised PRA.

QugatJLQD 720.56 Pleaso concisely describe how the coro- damago frequency ,

associated with seismically-induced station blackout sequences was calculated.

Regnonse 720.56 Soo the responses to Questions 720.57 and 720.62. _,

h a,,_,,,-,--..,_,,,.n._,.-.,.. .n..n.,...-,.n,,,_,..,.--. , _ , ~ n_.n.-,,-,,-,n-,~

Qatat ion 7 2(L.J22 ploaco elaborate on the computational procedures for combining the f ault troon shown in riguro 7.3-4 with the noismic hazard curve and the rolovant fragility data shown in Tablo 7.3-2 to yield the coro damage frequency.

litHEtnDe 720.57 The fault troo procented in figure 7.3-4 reprenonto the coro damage coquence attributable to a noismically-induced station blackout with subacquent battery depletion. This modol van nolved using the methods discuoced in the renponse to Quention 720.62.

QM20t10p 720.Sji What 10 the fragility for the solumically-induced basic event, SEISLOP (soismically induced failure of the switchyard) appearing in Figuro 7.3-47 What recovery factor was uned for the basic event !!XZRCVR (operator f allo to recover CCWS in scismic event)?

Etangng.c_7 2 0. 58 The soismically-induced basic ovent, SEISLOP (soismically induced failuro of the owitchyard) appearing in Figure 7.3-4 is the sano as the event, IE-SEISLOP (noismically induced loso of of f sito power) prosented in table 7.3-2. The fragility for this basic event was 0.39 It was essentially assumed in the seismic analycia that all coismic events would result in a loss of offaite power lasting more than 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months .

The recovely factor used for the basic event, ;iXZRCVR (operator f ails to recover CCWS in soismic event), was 1.0E-1 with an assumed error f actor of 1.00. This value is presented in tablo 5-7 of the System 80+ PRA Report (DCTR-RS-02 Rov. O, January, 1991). A copy of this table was provided as part of the recponno to qur:..;; ion 720.48.

i i

i Ouestion 720.59 '

Although seismically-induced largo-break LOCAs are modelod in the soismic event troos, there is no modeling of possible seismically-induced s.nall-break LOCAs, such as a break of a small RCS pipe or a break of the lotdown lino. Also, unlike the caso with internal events analysis, no consideration is given to consoquential steam generator tube rupturo (SGTR) ,

followir.g an ATWS in the soismic events analysis. Justify the omission of those ovonts from your soismic core damago ,

frequency evaluation. l 4

Ruponso 720.59 t

Tho contractor that prepared the System 80+ soismic ovent !

trocs dotormined that soismically-induced largo and small !

LOCAs would occur at about the same soismic peak ground accoloration level and thus only modolod largo LOCAs as being ,

the moro limiting of the two break sites. Consequential SGTRs woro not modolod because they are low probability soquences. ;

C-E is currently updating the System 80+ pRA. Solomic41y-induced small 1,0CAs and consequential SGTRs will bo ovaluated for the soismic ovent troos in this update.

QRution 720. 60 Was any distinction mado betwoon ATWS ovents and non-ATWS ovents in assigning a probability for the soiomic event-troo !

(Figure 7.3-3) top event, X (PSV Roscat)? . What is the probability assigned to the top event, X, in quantifying tho :

scismic accident sequences #11, #12, #31, and #32. Note that for ATWS ovents, the initial RCS pressure can be expected to .

be much highor, requiring more PSVs to open in order to t relievo RCS pressuru.

jtosn9ngo 720.60 There was no distinction mado betwoon ATWS and non-ATWS ovents in assigning a probability to the soismic event tree top event X (PSV Roscat). The probability assigned to top event X was -

2.8E-2 per domand. As - described in section 5.6.1 of the System 80+ PRA Report, DCTR-RS-02, Rev. O, January,1991, this probability is based on any one of four PSVs failing to .

rescat.

l I

r

+- - 6,-m,- . , - . , _ - e,- ,--w-rw i, e . . . - . . - - , - - - -r, ce-e, . - - - - - _ . m ---,,,3.,.e-,- - , -

Cutation 720.62 l ploano provido a conciso but syntomatic description of how the i soismic fault troos are used in conjunction with the scismic hazard curve, component, and structure fragilities and other random failure probability data to obtain, through discretization and convolution, the frequency of individual accident sequence based on the Doolean expresston derived from '

the soismic event troon.

Eqanonce 720.62 Tho seismic fault troo models prosented in Appendix 7A of the System 00+ PRA Report, DCTR-RS-02, Rev. O, January,1991, woro based upon the internal event fault troo modols and includo basic events reprosenting both random failures of compononto and poinmically-induced failure of components. Thono fault troos were initially solved to gonorato system cutsats using ,

-IRRAS 2 Bota Draf t and SETS. For this preliminary solution of

the soismic fault troos, the failuro rates for the basic

.olomonta representing noismically-induced componont failures were set to 1.0 to ensure that the.cutsets containing those basic events woro generated. Noxt, the IRRAS 2, Data Draft, event sequence solution module was used to solvo the " Boolean Equation" for each colomic coro damago sequence using the solomic fault troo cutouts generated in the previous stop. ,

This produced a set of coro damago cutsota for each sciamic core damago sequenco. Thoso cutsota included both random i f ailuros and colomically-induced failuros. Thoso sequence cutsots were manually edited to ensure that the .NOT. logic had boon proporly applied and to combino cutsota containing ,

seismic failures of like components based on the assumption that "ono fail, all fall" for equivalent componento.

Hoxt, tho Solomic Integration Program (SIP) computer code was used for the final solution of the cutuota for each eolomic ;

core damage sequenco. SIP is a modified version of tht SAMPLI:

code. The cutoots for a given soiamic core damage sequence are converted to a Boolean equation in a FORTRAN subroutino which is then linked with the main body of the SIP codo. The #

data file used by SIP contains thron types of. data; 1) a discretized vorcion of the seismic hazard curvo, 2) the median fragility and combined uncertainty for each solomically- i induced failure, and 3) the mean failuro probability for each random failure. SIP randomly selects a scismic accoloration ,

from tho input poismic hazard curve. Next, SIP calculates a '

probability of f ailure for each basic event representing a l soismically-inducod failuro, given the selected noismic l

The equation set for the solomic coro damago accoloration.

soquenco-is then solved to gonorato a conditional core damage frequency estimato for the selected soismic ovent. Finally, an unconditional coro damago frequency estimato is calculated as the product of the probability of having a scismic ovent with

l l

hangnpe 720.62--,dat.)

the selected accoloration and the conditional core damago frcquency estimate. This sampling process is repeated several thousand timos for a given soismic core damago sequence. The coro damage frequnncy ostimatos are used to calculate a mean core damage frequency and error factor for the specific seismic core damago sequence.

SAMPLE was then used to calculated an overall mean seismic coro damage frequency and error factor based on the mean core damage frequencias and error factors for all seismic core damage sequences.

Qaqntion 720.64 Please provido the detailed information regarding the soismic capacity or the fire protection system including pumps, valves, and rolovant-equipment. Please indicato the median capacity as well as the paramotors representing randomness and uncertainty.

Rgsnonse 720.64 C-E did not credit the fire protection system in the seismic analysis. Thorofore, the solomic capacition for the firo protection system equipment wcro not ovaluated.

- . . . _ _ - . _ . . . _ - . _ - _ . ~ . _ . _ _..- _ _ _ , . . _ . _ . _ _ _ . . . _ . . . _ . _ . _ _ . - _ . _ _ _ _a

- Ouestion 72025

- Should the seismic fault tree top event, PA0GSMDX (motor-driven EfWP-102 unavailable. Figure 7A-1) .contain, under GATE 2127, an additional event ~

r,VNOV138-1437_ -Similarly, should the top event, PA0G9MDX contain, under GATF 2140, an additional' event CVNOV238-2437

- Response'720.65 ,

C-E'is currently updating the System 80+ PRA. These fault trees will be -i reviewed and-revised appropriately as a part of this update.

- Ouestion 720.66- +

It appears that the following basic events are missing in some of the fault trees developed for high pressure injection system (Figure 7A-2) and " feed and_ bleed" (figure 7A-3).

- Basic Event- . Fault tree Top Event-where h 3asic Event is Missina

. CVNOV134-135 PHSG08DX, PHHB07DX-CVNOV136-137. PHSG090X, .'HBB08DX '

CVNOV234-235 PHSG10DX, PHHB09DX CVNOV136-137 . PHSGllDX, PHBBIODX Flease coment on this.

Response'720.66

' C-E is currently updating the System 80+ PRA. These fault trees will be

- reviewed and revised appropriately as a part-of:this update.

stion 720.67 In ~the seismic fault trees developed- for-containment spray system-(Figure

-7A-5), should the fault tree top event,- PG0813BX, contain-an additional event,1GVNOCS542-(flow diverted via the mispositioned valve CS-542)?-

-Response 7P0.67 C-E is currently updating the System 80+ PRA. These fault trees will-be reviewed and revised appropriately as a part of this update.

, A.-. ....--g, a e :-*w. --w-c ..,,,,.9,re v,.ee,m , , _ . -- .. pen.# w. ,ww.....w,, ,mw..-,-..--+,.._-_ _w~ _ __ _.m, . , -m= s. _ _ . - ; -w ,e u a,-

Question 720.68 Comparison of the fault trees developed for long-term residual heat removal shown respectively in Figure 6.4.1-2 and Figure 7A-7 revealed that there are noticeable differences in specifying the input events to OR-gates PJOB07RX, PJOB17RX and PJOB22RX. Please explain these inconsistencies.

Resoonse 720.68 The inconsistencies i:ssentially consist of omission of basic events ,

JVNOS0757 in the fault tree PJ0B07RX and JVNOS0756 in the fault tree l PJ0817RX, and incorrectly calling the i.ame set of valves differently in the l Figure 7A-7. C-E is currently updating the System 80+ PRA. These fault-trees will be reviewed and revised appropriately as a part of this update. '

Ouestion 72Q M Is there a missing event, PMSA0lMX (failure to deliver sufficient startup c

FW to either SG) under the the fault tree top event, POLX02BX (failure to establish RHR and failure to maintain secondary heat removal) developed for long-term decay heat removal (Figure 7A-6)7 Response 720.69 i The Startup Feedwater System is a non-seismic system and, the; Tore, can not )

-be credited to provide cooling water to the SGs during a seismic event.

Hence, the event, PMSA01MX (failure to deliver sufficient startup FW to either SG), is not included under the the fault tree top event, POLX02BX-(failure to establish RHR and failure to maintain secondary heat removal) developed for long-term decay heat removal during a seismic event (Figure 7A-6).

Duestion 720.70 It appears that certain basic events are omitted from some of the fault trees developed for long-tem ceondary side heat removal (Figure 7A-8).

They are summarized as follows:

Basic Event Fault tree where Omission is made AVMX-EF2B2 PA0G4MBX CVNOV138-143 PA0G5MBX (GATE 2510)

AVSX-EF2A2 PA0G6MBX AVMX-EF2B2 PA0G8MBX CVNOV238-243 PA0G9MBX (GATE 2532)

Please commant on this.

Response 720.70 The long-term decay heat removal is initiated later in the transient following the plant cooldown to residual heat removal entry conditions using secondary side heat removal. The Residual Heat Removal (RHR) System is then used for long-term heat removal. If the RHR System not available, secondary side heat removal must be maintained for long-term decay heat removal using the Emergency Feedwater (EFW) System. Therefore, successful operation of the EFW System is assumed when the long-term decay heat removal is implemented. The basic events CVNOV138-143 and CVNOV238-243 (Component Cooling Water valves for the EFW pumps not available due to maintenance error) need not be included in the fault trees since the emergency feedwater was successfuliy delivered during initial plant cooldown. Furthermore, since it was felt that the basic events AVHX-EF2B2 and AVSX-EF2A2 (common cause failures of SG isolation valves) are demand-type (fail to open) events and since the EFW System operated successfully prior to initiation of long-term decay heat removal, these events need not be included in the fault trees.

Question 720.71 Should the fault tree top event PC3N02BX (failure to deliver flow from CCW loop 1A) developed in Figure 6.3.3-2 contain an additional event PEEN 01A6 (loss of 125 VOC control power for train A)? Similarly, should the fault tree top everst PC4N02BX (failure to deliver flow from CCW loop 2A) developed in Figure 6.3.3-3 contain an additional event PEEN 0186 (loss of 125 VDC control power for train B)?

Resoonse 720.71

The Electrical Distribution System (EDS) support systems for the Component Cooling Water (CCW) pumps in a particular division (or train) are similar to those for the front line systems in the same division. These EDS l support systems are addressed under the frontline systems. Therefore, the duplication cf these EDS support systems in the CCW model is not warranted.

Hence, the additional events mentioned above need not be included in the corresponding fault trees.

l

0uestion 720.72 In Figure 7A-8, the basic event, AVSX-EF2A2, is defined to be "CCF of SG isolation valve set 2A(2), EF-100,EF-102." The same set of isolation ,

valves, however, are defined to be EF-100/EF-101 in Figure 6.3.7-2. A l similar discrepancy can also be found for the basic event AVSX-EF2B2 appearing in the same figures. Please clarify these inconsistencies.

Response 720.72 The definitions of SG isolation valve sets given in Figure 6.3.7-2 reflect the latest design of the Emergency Feedwater System. In Figure 6.3.7-2, the basic event AVSX-EF2A2 represents the CCF of SG isolation valve set EF-100/EF-101. However, in the same figure, the CCF of SG isolation valve ;

set EF-102/EF-103 is represented by the basic event AVMX-EF2B2. On the <

other hand, in Figure 7A-8, the basic event AVSX-EF2A2 represents the CCF I of SG isolation valve set EF-100,EF-102 and the basic event AVSX-EF282 represents the CCF of SG isolation valve set EF-101,EF-103. The basic events were called dif ferently in Figure 7A-8 when the fault trees used in l

the seismic analysis were modified from the fault trees for the internal events. However, this inadvertent error does not impact the results since l the appropriate numerical values of the probability and fragility been used )

in the respective analysis. C-E is currently updating the System 80+ PRA.

These fault trees will be reviewed and revised appropriately as a part of this update.

Qufstion 720.73 Comparison of the 125 VDC bus fault trees developed respectively for internal events (Figure 6.3.1-7) and seismic events (Figure i A-16) revealed that many of the fault tree top events (inquiring about the availability of battery power) constructed for the latter contain an extra basic event, EBTA8HR (battery depleted-no recovery in 8 hours9.259259e-5 days 0.00222 hours 1.322751e-5 weeks 3.044e-6 months ). (a) Should this basic event also be included in the corresponding top events for the former?

(b) What probability was assigned to this basic event in the seismic quantification?

Response 720.73 A prolonged loss of offsite power is anticipated during a seismic occurrence, thus placing demands on the onsite power sources, such as the 125 VDC buses. In addition, a seismic occurrence may render a 125 VOC bus unavailable and the 125 VDC battery is more likely to be depleted since the offsite power cannot be restored for a prolonged time period. For this reason, the basic event EBTA8HR was included in the fault tree top events i constructed for the seismic events. In the case of the internal events, l Station Blackout is the only event where the 125 VDC bus may be depleted

, if the alternate standby AC power cannot be established or the offsite l power cannot be restored within a certain time. A fault tree was developed

! to account for this scenario when determining the frequency of occurrence for the Station Blackout event (see Figure 4.8-2 on page 4-63 in the report l

Response 720.73 (cont.d)

DCTR-RS-02, Rev. O, Volume 1) . Therefore, the basic event EBTA8HR should not be included in the corresponding top events constructed for the internal events.

Question 720.74 Are the fault trees, PEEN 02A5 and PEEN 02B5 (shown in Figure 7A-16) specifically developed for seismic events? Why there are no corresponding fault trees for internal events?

Response 720.74 The fault trees, PEEN 02A5 and PEEN 02B5, were initially developed for the internal events, but subsequently deleted because they were not used in the analysis. Although these fault trees were not used in the seismic analysis, they were inadvertently included in the report.

Question 720.75 It'is not clear why the basic event, FSERAPS (no actuation signal from alternate protection system) is included in the actuation signal fault

. trees developed for seismic events, but is not included in those developed for ir+.ornal events. What is the fragility assigned to this basic event in thi seismic quantification?

Response 720.75 In the analysis for the internal events, it was assumed that, except for an Anticipated Transient Without Scram (ATWS) event, either the reactor trip is not required (Large and Medium LOCAs) or the reactor would

-automatically be tripped by the Reactor Protection System (RPS) on a safety parameter. The Alternate Protection System (APS) is provided to address an event in which the RPS falls to generate the trip signal. ATWS is such an evert. The analysis for the internal. events defined an ATWS as an occurrence of a transient requiring a reactor trip for reactivity control coupled with the failure of a reactor trip to occur. The failure of a trip could be due to either mechanical failure of the Control Element Assemblies (CEAs) or the failure of both the RPS and APS to generate a trip signal.

The basic event FSERAPS does show up in many of the ATWS sequence cutsets.

In addition, the APS generates an Alternate Feedwater Actuation Signal.

Figure 7A-17 shows how the basic event FSERAPS is modeled in the fault tree model for the actuation signal. The seismically-induced failure of the actuation system is differentiated from the non-seismic failure at the top.

No fragility value needs to be assigned to the basic event FSERAPS.

However, the (mean) probability for this basic event is 2.6E-02 with an error factor of 3.0 (Table.5-2 in Volume 1 of DCTR-RS-02, Rev. 0).

b.

Duestion 720.76 What ue the probabilities and/or fragilities assigned to the following basic events: PC3N0lMX (Figure 7.2-3), PC4N0lMX (Figure 7.2-3), PCSEls

-(Figure 7A-11), ECHATTER (Figure 7A-13), and ECHGSEls (Figure 7A-16)?

Response 720.76

- The basic events PC3N0lMX and PC4N0lMX shown in figure 7.2-3 (Tornado-induced Station Blackout with Battery Depletion) were derived from the fault trees PC3N0lMX and PC4N0lMX for the internal events by adding an additional basic event CSWINTAKE (service water intake blockage due to

-tornado generated debris). This basic event,-CSWINTAKE, was assigned a probability of 0.01 per demand. The fault trees PC3N01MX and PC4N0lMX for the-internal events were statistically combined with the basic event CSWINTAKE when analyzing the tornado strike sequences. - Therefore, no direct probabilities were calculated for the basic events PC3N0lMX and PC4N0lMX shown in Figure 7.2-3. Similarly, the basic event PCSEls (seismic-failure;of [ Essential Service. Water System) module) was_ derived based on the fault tree developed in Figure 7A-10 (on page 7A-108 in Volume 3 of DCTR-RS-02, Rev. 0). The basic event ECHATTER (failure to recover seismically induced bus chatter) was assigned a probability of 0.05 per demand (Table 5-7 in Volume 1 of DCTR-RS-02, rev. 0).- The fragility;for the basic event ECHGSEIS (seismic failure of battery charger) was inadvertently left out of the Table 7.3-2 of the above mentioned report and has a value of 1.69 as specified in Table 7.3-1.

l.

, Question-720.77 Please supplement the detailed fault trees (illustrating the fault tree logic) for the following zero-level fault tree top events which are missing in the System 80+ PRA:

LCCSAPWR PGSB01DX LPWRCCSX PGSB01EX l LPWRCCSY PHBB01BX i LPWRPCCS PHBB02EX PAIB0lMX PJ0B01BX PAIBOIRX PLCH01BX PG0B01DX POLB01BX PG0B01EX PPAX1MBX PG0B01SX PVBB01EX PGSB01BX PVDB01BX PGSB01CX RCVR1 I Response 720.77 1

-The fault trees LCCSAPWRe LPWRCCSX, LPWRCCSY and-LPWRPCCS were used;only .

for design evaluation of the component control system but were not used in l

'the System 80+ PRA analysis. Similarly, the fault trees PLCH01BX and. '

PPAXIMBX were developed to model, respectively, the failure of safety injection. tanks injection for medium LOCA and the RCS path unavailable for RCS pressure control logic but were not used 'in the PRA analysis as they >

were not needed. Therefore, these fault trees are not included here. The fault trees PAIB0lMX, <PAIB0lRX are part of the zero-level fault' tree top event 4 POLB01BX,:which is presented in. Figure 6.4.1-3 starting on page

._6-668 in volume 2 of the report DCTR-RS-02, rev. O. The fault tree

. PAIB01MX (page 6-671) refers to two fault trees PAIB2MBX (page 6-382) and PAIB5MBX (page_6-385), which are part of the zero-level fault trec ;op event PAIBlMBX presented in Figure 6.3.7-5 starting on page 6-381 of the

-report.: The fault tree PAIBOIRX (page 6-672) eventually refers to two

ault trees PA0G3MBX (page-6-360) and PA0GSMBX (page 6-362), which are -

part of the zero-level fault tree ~ top event PA0GIMBX presented in Figure

- 6.3.7-2 starting on page 6-358 of the report. The fault trees PG0B010X, PGOB01EX, PG0BOISX, PGSB01BX, PGSB01CX, PGSB01DX and PGSB01EX (and their subsequent level-trees, if any) are included in their entireties-as a part of this response. It is to be noted that the subsequent level (s) in any

, .of the top -level fault trees need not be in a sequential numeric order since the subsequent levels many times refer to other fault trees. The The fault tree PHHB01BX, is included as a part of the response to Question 720.11. The fault trees PHBB02EX, PVBB01EX and-PVDB01BX are presented in Figure 6.3.6-13 starting on page 6-323, Figure 6.3.10 4 on page 6-460 and -

Figure 6.4.4-2 starting on page 6-874, respectively, in the report.

Although only subsequent levels of the zero-level fault tree PJ0B01BX are referred to by other fault trees,= it is incl ided here-in its entirety. The' fault tree RCVR1, which was used only to get the basic events representing the recoveryLactions in the IRRAS database, is;also attached, l

(- . __. _ _ . _ _ __ _ _ -

d

3GO30~ JX FALURE OF CONTAINMENT a

SPRAY SYSTEM i

f' )

PGOE 01DX-1 LOSS OF LOSS OF

'CTMT SPRAY CTMT SPRAY LOOP 1 LOOP.2 PGOB02DX PGOB08DX

\

4-

---,------------,--,---..-------------.----.-----,---e ,

I r

8tg l

D$

e 7~w O

ne 9

a: --

_ 8e 0

~

yl m

b i!ic Gl !ii Gl a -

x -

~

ige -

9 jds -

f g

s.g -

~

h

!a e

<l n_l.

.. ~ , er #-.w.... i + ,,,.. .e ,r...,-m-. .. -. -- -,e. ..-,.--n.*-- .~-......-.-_c.,

l i

1 o55]

h !

6 g)l j- 1 25 (' g $

lif n;

s a

.r Gr be I "e

IC bb l

.s

- - I l fe i b n, s_

g gydp,,,

,.- G. ~L I

+$ G'ir y'

~

1 s .

L p[a 1

f, Ok~~ .

r .

ae a .

$d' k,.

~

I i.

u

}

l-

, . . . . - - . ~~ 4

4- ,

i. -

j.

l

4 b

i i

! .PGOB06DX l L

- r ;

i- FALURE CF 1. - l, !

RHR LOOP .i I

j. PUMP OR VALW3l

, .i 1 .t -

.. X-1 4

1 I -1 i .

kW4UAL VALVE RHR PUMP CHECK VALVE j UN4UAL VALW SI-123 RHRP101- SI-124 St-125 -

l N.O. DUE TO ME. . INOPERABLE FALS'TO OPEN N.O. DUE TO ME._ ?

i

l. 1 PJOBOSDX ;

i i ' JVNOS123 JVCA9124 J.WO9 t25 [

, t I- -[

l !

, t 4 I t

't js

~

~ :

r

'i i

a b

{

, 6

~

.. - - ._=-_____-_- _ - -_.--__ __.__--__ _---

,--,,-._-----,-._----p, _ , _

ja I

~k,3 f

h2

~p n g x

gfUI- $CJ -

ljg a,

W r9 De' -

$- lhh h ~IffOk e

jj (l 5 ll Of fp k

~

fo

, , , , - - - - - -N, n -,, n -- , e

i PGOBOEDX 1

' LCSS OF

. CTWT SPRAY LOOP 2 s

I i i i 1

LOSS & C5 I LJS5 OF CS COuuCN l CS hT EXCHGR ' LOSS OF CS ,

- CALC FLOW TO FLCW TO .- 1 FLOW TO ,

CSHX2 isttFES l INOPERAiLE PUMPS CS MX 2 -] CS HEADER 2

! \

racceUCx . Ca. 413 PcctmBx MXB10BX CA. s 4 U j

' 'l 7 I i 1 CCF Or 2 MAf M . VALVE FAL OF CS CCF Or 2 CCF OF 2 53-242 Ato RHR C5 HEAT SET A SET A CTWT N.O. DUE TO MI. Ptjup OR vtt'45 EXO4ArCERS uoTop vstvEs esct v.svEs

/ ~ I q CWLC HMrA2 CawA2 CVtC5342 1

t

'. . - _ , . _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ . _ . _ _ _ _ _ _ . . _ _ _m. _m

i

'l l

1 h

h/_jhau gs$

s 3"

~

\ E h k

b" 19

- ~ ..

f %

.e )

' 1 a ,D l

j x

.H-(

Q l-agg %

g* l To-gf A.I

( -.

8 ecd wa

- h 5 s

n$

hy #A t

$s Gr %a s%

k 1m u 3 -

g>g --

1-t Lt

., Gb u,,,- .

Lt l --

l-

._.r. - - , , , - . . , _ _ _ . _ _ . _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ . -

PGOB12DX l

FALUPE OF RHR LOOP 2 PUMP OR VAL \ES PGOdt2DX f i i i i :

CHECK VALVE t.%NUAL VALVE mat 4UAL VALVE RHR PUMP SI-225 l l

RHRP201 SI-224 '

I SI-223 FALS TO OPEN PLO. DUE TO ME.

tJ.O. DUE TO ME. INOPERABLE I

N h i e l

I PJOB6DX J.'CAS224 JnCO225 J/NOS223

1 l

l i

)

l,301

@Oi- $ c! I)

I l

5 8i!GI-

~

L 8- $i al .

e J,31 /l L _

E3 nn ih 3 s ur 1 s Fn 95 Il OI e Q %

3 pj bl ji l

J

y'*'+-' .F. 5%'r 5 p ,, y. 5 a - -

l I. ,

j.

! 1

? 3GO 30" EX f'

FALURE OF.

CONTAINMENT:

SPRAY : SYSTEM

~(' h PGOdO1EX'

-LOSS 0F. I LOSS OF CTMT 3. SPRAY , CTMT SPRAY-LOOP l.1 . LOOP 2 -

l PGOB02BX PGOB08EX

.--g( y yF 1 .g,a q gi1s---- t!W, e7 y 1 p -p e4.:: eTia vr gi. ---y -m-y--gg

,J. .n,.,&y.a. - 4 a JL 4)4d4-+,u.,w41 . h,,J _.4 .%

- ,a 5 m 3,4

N3

pp- n g -

es a i

\

f b d,_

f6f 3 g5 5 a

i n h'

- . bk e- num )

_ bhe 1 le I ,

a t

N !

9 6{o5 {

>< -, yag

. uj > t Don u

.O DE g- gd g a gee s

3

a. y'm, 7

>CS

_ g

$h 8 og 80 -

k

~

g 3

s M

~

hk.

k v m k e v , , - . . - ,miwr,m-+- -,-er- --- - wm ww 3,w-s .-w - , c.-e,--v v ee-,=. - --- -

4 --

L -A- - 99's2 A4e5--sJ-s- k4m Jhuus 4- .s 6.

~9 -9 161 i Sr

~!g5

w --

[

I S

~$

B" 3 r I

b! Of-

~

8 EJ g

W (a;i!>m

- a '-

h$5 uf Y3 -

Ji o

~7 ~7 d- $g g;<]_ >L D

' ,lJ 3 6 w w)!

b

-l@

pi Gi-se n!* 0

{FAs !

IE _

3 h' [

~.

F i

3 1-r 1

, PGOB12EX

( ..

. FALUE ' OF -

RlR LOOP 2 DUMP OR VALVES i PGOBr2EX

. I i 1 l

-MANUAL VALVE 'RHR: PUMP CHECK VALVE MANUAL VALVE:

S1-223 . RHRP201 S1-224 St-225 N.O. DUE TO ME FALS TO OPEN FLO. DUE TO ME l- NOPERABLE

!- I emx .

J

yesm msn. uosas e

i-

. , . - . , , + .. - ____ _ ___ - _ _ _ , _ = _ _ _-_-__._ _ _ .- _ _ - - -

w w -

s n

f F N X ,

O R2 P S

B S SPO O .

B .,

S TO O

G X O ML LC T D S -

' TW rN O X OE O

S L

E NMF 1 O

p B RI UAY TA L NR d

O G

O A OP rCS P

G F A Y .

X J O R1P S

2 y S SP 0

O B .

S O w

s OTO ML G P

LT / e C

4 : i 2 l! ;; ;i;! . : 4

g -v- -.,- -.- , . ,.-- , . ..-- m_ -.- . . - - - - . -- - , . , - - - - - . - - . - - - - - - , - , - - - - , - - - - - . - - - - - , . - - - - - - - - - -

r I

0 8s ._ g S89 90@M9 4

~

sM

~

6"!

lf g5b>

N loc-

~

hh a -

ch h, a

~

q!s! O -

882 egs f De-ii: Gi e '

,1 y

~ 9 lAs l o, '9 .

if F l L

f 1 .. -

4 h

r l

PGOB08SX i,

I

. LOSS OF ,

CTMT SPRAY

i. ' LOOP 2 1 'l l l LOSS & CS - LOSS & CS ~ LOSS OF CS COMMON FL7# TO FLOW TO - FLOW TO CAUSE PtM'S CS HX 2 CS HEADER 2 FALURES PCOB133X CA 4U PGC8tIR CA 41!! .;

1 1 I I ._

WAffdAL VALVE TAL OF CS CCF OF 2 CCF CF 2 S-242 AND RHR SET A . SET A CTM7 '

N.0, OUE TO ME., PUuP CR VALVES MOTOR VALVES' r5CL VALVES y' f

- i .

M t

i i

t l

1 ! !l 1!l; 4 ~

?

"I l

k

-Y

" FA F2 X B

OP' 8 I

S SPO 0 B.

S TO S G

X OML LT C

P I

B

" -TN FNO OE L X 0 ;MF E Wl R

B 1

O B1 UAY LNRTA r' d

S G

AOP

$ F CS P

G Y _

? F-A' R1 X.

~

OP SP 8

2 0

S S TOO l

B _

C _

OML G P -

LTC' s

4 4 : ;l ri l1i l 4i[ i i:!f 4 i,i }j

. ; ; . a

l I

n50

~ ~

g pj Gi -

a b:.Y -.

}

p?S $

9 00-

@{

\

X E-en E,!i Gl

!ad l O g~ * -

O_ S s yh a

dye V ME l 1e' c

hlke

~ b td s

7

)

I U

4! -

l )

bas

~

Y g

.. 1 a%s b

9 O"g .

_ e0 5-g5Y B s c :

m -i

~

. co

~

bgn g

~

Q b5 '/ iB

!sh $ \l m

o n-

-1 e

h a

due =

d$

3-e"

(.

e 5

b]m gi n

, + -aca-c 4%,- , + . -,-,-- r.~ - - -. - - - - ,, .,...,%--. -o -A e,. ~2. , . . - . - , r- .we.. , , - + ..,.-,--,,--,,-.--c.,,.,,, - . . , , - -

00-

_ be=

hh '

4 sg x l ,

N s D- 85*8 o' b e j'w !g-l g Gal jg G[ -y lye r

~

$05 f 6 .

'9 a

~

g5 h i

4 1

h b

h xi Es"

. o r 688 O- 2e ro x -

E k [ .

V) da$

o ..

O-.

,le b-ssa M

i e

G-.

f f:

L I

l l'

WWW M D 'd - - _ .

_ ng a v" f5 Of

.g ,

sg' _ ;

00 1 n '

._ bR"- .

llh x

O i CN

~f:>

0 *

~

<> _ v a "e- 9 jht) we 4R

~

ps

~~

,Eg se4.

ti o ,

s

.a

,,-a-,,, w_ , , - - - - - - -,------,,---,-,-,.w--------- - - - - - - - - - - - - - - - - , - , - - - - - - - - - - - - - - - , - - - - - - - - - - - - - - - - - - - - - - - - - -- - - -

I 4 --

'l Dd 9

\. i q+, w

\_

n-Es

_ bGgY

@d

.o ,

j shN m

Gl b5 i

l :

n"R D +

, ,e >

$U? '

d5 , i og is g ,

1 . -

gj g

F

+

5 }

=-v"*-m +-~

vv~.ee+,+-+,c-w n..., m.,=,-w,.ewee c we- e--w n . ear =++ ,,e-,

m ye s,,, ._, e w t-r ay-vn ew ,m e-.reww tvg wf

' \. .

)

i 1

1

)

r E

+

k a

bN -

h"t" N d to .

O

rd CD- lk-m a up t L

. a_ . b ." k;.

~

@,GsS s 1- ,

L Y

?

i k-srv== .r- ---y ~ n , +- e 4 w g 'we--rsweyw+e wy v- 3---,et--

4'.+-- M-- -s*,s4-M.W.4 ,wh 3 t. W 4 h4 En- ,y

]

"l9 8.r -m Jj u na Id.ai .a 1 u=W P >

+

2

-l

.l 1

kY

. b g5Y 'N P j

.n ,

{'

m s ,, g agg :

b gg l

gg w

I w l U"R b

@hu a!~e n .

s le' 1 Go

~

g.h

+

i 1

l.

l-3 5

,.l ,- , ,-C, , e,,- , - , , , - - , , , . , - , - - - . , , - , - , . - , - , - , - . . ,.,-v.-----,.--,~,,...,--,4.,,,:,a,-, ~ ,,..,vn,- - - - - , . - , , ,, .,,,,,, ~ .-- -,,, ,

I l

$N' h '

ee*--- t"'Ntw"'

- Wm.9- **--eSe vs. +-+Nwtev-='-4r ----.er_amwe-re-sm,++>

- we

i

)

i I

I i

as s 3 I

'I M

~ '

o

y cn
, g' g5 Q h i
m vd -
b n,g.
d a
a. S v --
w5 _
U>v "g 9 g.*- B
$hd g '
$h a Ae?e .
er b"j k kN"0
,R Q
u &
@~
a i
a 3
i l
i 1
bh l ,
~
y[~
l^
,l 1 fglgg.3}
1 u j, pal :
2p D W
D h 4 8_
g 45 ci _
>ae -
l4 9 gas
@. g ,
E
[I -
l de 89
Of- [
g &
dj t -
P
~
ae ~
g fds (
1g l, r 1
l
,,.-.-p , ,. , - - , , , , , , .. -,,,,.- n,,,-,,,e ,,,,me,. ,, , , . ----,,w.,,,
., c. ,a+ , - , , . , , ,,.w.... g- -n ..e., .0+n-, .,
> 2.
.J / v
'l. Y
> "1,o -
c
(
Tii3 o '.
p 6 4 d.;
.mma m ,- ~4eam "Y
A Lb 4 h
_ >00 _ a
(, vL H 6
%,,>m g a
O < tu A
(3 CD t >
4- e_ g,s-o .-
6Js,w cr.
b "j O
p duge- -
O_ d ta v /x 0:se; ss 8 h.ct (h Mil 8 s;.i iZ u
LAJ tj1 ,,'
>qno 93
~
W T tij ~ " -
f3
.\,
e mo 3 f; a
=
___--m- -- -
_-.p_e_& _ .__ i he .-,
+-w===- a+=. .sh +4-A--. S--umm iMC,4+de4-s8- m--1-,e .M da-Mae.E, - AAomhau -aa* MA-.aaa.Ali_4-A.OL#Ah .u . 4_Al.4m_m._d442A-M,2 4,M4--t?=ah mm.u--+A-A I
1 I
I t
le - g '
$Q .
W (D -
I-e
!!l Gl !Ie l.gi3 . .
w5 6 N
~
EAk Ci E85 65 lW ;
0 i -
en w
. o d
a 33 a
I l
P 1
l l
_ - . - . - - - - - . - - . . , . - - - - . - ~ . . _ _ . - . - _ - . _ _ _ . . . . . - . _ . - - . . ~ ,
i I
k i
f h
eS e
_ .5 s :
i a i t
i g e X >
CD E .
O ag_
g$,
o on .
es g ,
a
%./
e s Na .
i f
b <
gh~ug e4 $"'
es :
-D E s lle -
g eg< g
,;. 88 -
, 5 >
i.
l i-i.
i s
t n .[.-e. .
wer-e n .-ww -r--c, e,.*--e r, e -r---we-r-er-m-ww,emn-an-.m, w---y,.ew.r-w,wmv= , ore,-~ arew ,w -w m m we-.we+-+.--e.,-,'v<ww,--,c-++.,i. <-,p-m----w-+-.yw .- 4-= gw e i
i p _l Gr gr o
6 8
- I -. _ O u
s n , e f5 !
2 -
f 8 y[ Gj-8 g-6 lj@-
f l T?
te . . _
u W
u,c a S@ h L lt s > -
l ,
u I
q- . - - _ . ..- - - - _ _ _
J I. '
i i i i !
a ,
bR. 9 l
gf'
\0 l i
i
! i
! -~ >(,
L CD +
CO b- p- --
- O ~
up " ,m .
. - r-6p 61; g_T -
h O.
v ;
i i :
i !
J- i b
+
4 i I
- 4 a gg ..
1. ;
$g$ /t '!
}
.l p.
-t t
t l
f 3
.i
'l r ,
- 2 .
I f
l i
I 4
4
_ 40 _
a '
I a$h, A
b :
B 58 e G"u! <
e -
- Q
- l)
D i Q ~
g.
r%
>( ,
p6 e O 8 -
a V g38 3- U 8
o Re
-- )j. eg$U 8
E a
- Q,- s'd a5 v
v t-g-
gg s s; .i
?
e I
.-w--.a. e .,E<-i- .-1,--._rerse s'-r,%_ e - -, . ,s . w w b , - w c-r .,-..w....v. w ,,swer v.w-w,w,,-%,, . - w w -r r i-y- r, -e y- ey'-et-e, i-y-w-.n-, --,.ce--- .y+,v%=-wy..
- A^
i t
l l
$j o b aa -
b,,,
si!
8 fb 3-_
b-g6 s
m f x 6 CD y g y '
$U y up Gr@ er>U Ir
~ -
, n, s'9 S:
CL , s6 v
148 )
5
>& - g '
3,3e l
t
?
-f at.--~~, , ,- .,.- -rv,.... ,w.v.,,-,,,nw., v y-_-- --,w...--+e,v,.,4,, , - . rt-r,rm-. -,---,,,r...-,----er,--,w..-v,., . . . . - - . . . . , , _ ,
...,w<..,.-w . . , - ,=
r &_ w &Is ARES &-<EeW J-w6,L 4 Ae 4 44- w--oAOM' 5bmb 0=M&*H*-&-hM5"*M"5'*- 'O"-:*=SM**MA'*cw *'&#**MAA " b "' A A" A b"-a**""---b^WN-"*W&45 '
i t
i h
k B,2 x ,
a ,
<to
, i.
5 6
A X '
CD W e
W g .
co - g5 >Ri ia O -
9 !
n 9 '<
Q., ,
v
.)
v
- >0 9 --
Bh
-m 3Yw -i i k
(
n'r g ersw eeg--d" gkMf'"-M g*g w*s- 18Pd-T'F'TT="'*
4 4
1 S I CEE B I VSP Ul 2
2 val E OCA C F C
J P
2 CX LH X F V E E N
$R T T
V L M
&R O O C A2O A T
J
) SMT P V2 2O Y
X [4l R'
SO OR i
R6T 2
S O
B L O
LF O5S T
J P
FO 2 O L 2
1 OC X
B M AF SN P 2 B
l S# O OTO O LKL I
P J J P P NSN OE E E EN
( lSi TVP 5 V P L X AL LAO 2 A1O E OV O 5 V2 1
bT i 2O t 2
G RT RkT E O
rOTL A G OS T S J P
uOA OL T
CMF M A F
i E N V
L PE .
W A0O .
O I V2 2O m e .
Rm2 X
B RbT OS G
J C T S_ P TR C I ,
X B 1 OH O J
u AF ST P _
S' O _
L
m.L..4 ..wa e - w e * *
4ea- uw. _ mm.s.m waw 4 =4m 4 a ew es4:5- A wus.w,-a a am mma- a
s. . . . ~ . + , 4*.*.mma.w--+.a.~ae-.=x.dmm---..A. s.-=_ u- w s-e +--h o l
l q
b"& hd5 '
n>a l- 'd '%
t 5
$$$ q OW u-8 9 -
($
ba ff u
{u L h b e
"S s
gg Eg ;1 4 Qi e>
s 6e s
.al I c.
IN
. m. >P!o n M" X tJ Q3 6_ A2 q . [$ {--
eg-
',Q dgm sne 1 b -
'T
? Id ct O
" go v __ e v
- >ne V9 1 i
es a '
Is e 6 d'@g A" -
1 e
- a}
w V) .9kb Ne e
8ez 9
de
. , , ., - .,- c. 7 .,-,.s , , -. w g-,. .#-, y, #+ ,,,,w,w.,- ~ v y--w.ma g .-e-, , -ew%, ,* w-- -e e h s i _ g. ., ,v
! ( l , ,
, ;i }! -!tIiI> L ! ,f[!;'
E E. -
V M L _
5 -
A5O 3 V3T 2 i 2 l S LE O At J US:D n N G
- A . -
. M O.
N -
7 E N
=
E VL P 4 A4 O 3 V3 T i 2O S KT A C
Ct v E SS G
H L C A F
. )
X S E
B F2VL 4 O PV A (
8'
. 1 EO M _.
B I
ROR h. u ULG L /, O _
O J - .
P -
ASP J FCM .
U
.P -Pl E _
( P:1L X M E0 .
A E
! U2R PPE S SP O
. SCO C N I
/ P J
E .
E t .
V L
V .
3 A3O 3 V3T C .
2' .
c I
w1 }
L E At USD N
A U
( VS w G _
U O.
N 1
^
n nl -j ii !i ii!i.i I ti, Il!vl 1' l4 i ll l[ j i!! ,
iti !ll1 l ; !t
. m4 -_aa ab. #waA, ,e.p,AA An- ...A.4Aw.Aha ad 6 A mmv+AAd u--Am4 AdA#h p'hal khF4 MAe AM+----*-da$w+--_4+h-.e+EA-.A% hu 1Ad MM 9elLM4h48 MA 4444ewd4S A AWW r 44-.
'I .
t f
. , r I
r i
L s
I r
i hN -
i r
t gh 3 c ,
..s p m
[isa (
m Y N CD 3 E $
$ hw5 Y 5
}fg**Q n? E m
v" b>gq (5n a
\
o. [3- 5 -
b -
I. ^
CL ' g v @ -
~
' ~
@l$
s [ RGE 85-l a
?
6
)
e 5
b, y 1 8= s I Elso .H o to -
w i
t g.
'?
I e
, 1,.,-- . ,- .,, ..v,,. %.v.-..,, .__,- .,..,,,u ...,y ,,,m..... .,ns,,,.-,yor
.,,-,yy..#,,,.,-.ry. m#.. we ,, m--,,
w _ _ _ _ _ _ _ -_ _ _ _ _ ,
g-- *-m.
'~
I;"E _
b-hw Meu q . ti -%
4 (M ' .o. -,
s}
C w m
r 5, - - .
m u, -g y> m n
to 4 .po T.
'~~ k ;M / gW "g rP-7 g* h
' ~j hc
~ '
{$ c$ xT Dj f fg ADe 4kgJ sse o 'e l
gte 's]~h-esp, se 3 JU
, l
.h )
~N !I $ g3
$ p,iA en,
/g 36-a..
.,_ 4sN ,p; 8% ,- b g,3
-, 8>2 ss n g ..A
. u __ - - -- - . - . - - - - - . - - - . - - . - - - - - - - - . - . - - - - . - - - - . - - - - - . - . -. - - , - - , - . - - - - - . - , - - ---,----- - - - , . . -
1 i
i n_
P Y
vnb Ng r. . ] !
i 9
i
~__
i 9" W i
~
fb(Of sts e
f{c(
ee f ;
l l
t y
s",. ..
>' . nc . - .
7- bh ?
-l 5 ._ ;
gg"' /. l j \f 4 f s re
<n 89 3[lt} -6 n N J !
I D' [gt e_ @~cG, it 6, [
]!;sGl- .
l p
e ,l b
l>9
~ ~
.e n '*
[f e F ,o a i 4
I9 ,
e-f, g ,_ '
9 13 g5 :
!f r wg; -
9 .
y ,v 1
__ i i*R g g fA$ ( ,
ay 1
,.he-- .,,-,-v,, + , . - , , , , , ,~~-e.-. -,w.,,, .w. ,,, v.~-..,_...,-., ,.'.w,.,,,,,,,,,,,,,-.E,m,,,w,.,+,=-r,,,,m.,,w--,,,--,,we.c.,-.,.,,,,w.r.,,.,,vw-E , +.e e..,,a .w--,,eng,,
- . _ _ _m..-
-_ -my _
ll I
)
l I J
1 i
1 1
1 4
I t
. i I
4 l
1 4 BT om g
bS"y' g- !
e ,
M p QB
.A e
l'
! A
! X 1 I CD I I 00 b t
y l wg;N 18 i l' '
i b bey a
e( b 7
!. ' O, !
I v t 1
l- f 4
_ ty _
'g-Na ff O D
T3
-3,
-I ,
m F
WWW* IN MVMN W WE'r *9r mWhh g tug'u W NT s FwMO gr4mg '
U
k I
I i
i 1
I I
t 4
a i
o y e
-m jsg Gl. ,
E i
E a '
Og l g M e l
QJwg e 8g:n .
W i
m
$gag
"?2 $ -
@ -s V
W !
i ke . - - c gy +
4 T
1 I
3 I-
.{
l- ,, - .c _ .__:_._:.,._...
.._a..._... .._...;_,.__,.__.-_-
m_--sii
(
gop ,.
g ,j t - . . .
gu v O
g.
gk 3If __
f
- e x -
CD w
48 h _; [3 /"71 bE !
m sr ( 14a eag o +s $
p 4 0 n h 1 >5 fl.
U -
bl a d-O
[j
_ ' Fio .
D h$$ .k sea -
I
--,,---------n. - , , . _ _ - - - , , . - _ . - ,,.-,_ - - - , _ _ _ _ _ __
i f
l f
h 5
I 6
5 e
i i
t
~
Da '
u 9
Rgl4 i a gi 5 .
1
^ _
X .
03 c) -
v- 5 f
ca g e '
'ye :
o n
4 t
Q !i "o v
M 4,.3 3.e d( -i t
?
i .-
l
\ ,
v e
[. 4
.=* . - , . . . ~ , . - - - - - - , , , ~ .--,--m.,.... ~.._-,...~.....,,_.,m.,-.m..mm-,_.,,._,..-y,r 4_r, m.....__,mm,_.., ,,,,,,,,,.,_..,7m.,_,_,~.
I i
l j
.4 g
$he ;
!~
4 r;p-
- b ~
I ,
N
~ ' ' '
m (N ~
l
.o 5y sg,- ._
g bis i f*
sgu -
j b@A l
GEli sa _p ,
bja
^I I
__________________u J
I e 01 li!
=
!! OI
~(
B!OI-lCl E l-'jo; e 4.0a h,
" '~
3 V
f ,
s .
g T 4$ hf 4!!
m . O 2 F
4j!Ol ->@Of v
qe, i
ij! OI m
NO
Ogestion 720.78 suct, aful operation of high pressure safety injection pumps in loops B and
. D requires-control power f rom 125 Vdc bus B. In the event of a failure of
-125 Vdc bus B, only HPSI pumps in loops A and C are available for mitigating accidents, in the fault tree (Figure 6.3.6 13) developed for failure to deliver safety injection with 125 Vdc (bus B unavailablo),
therefore, should GATE 1836 be defined to be
failure to deliver sufficient Si flow to 2 of 2 loops" instead of 4 of 4 loops? Similarly, should GALE 1838 be defined to be *CCF of 2 of 2 loops" rather than "CCF of 4 of 4 loops"?
Response 720.78 As described in subsection 6,3.1.1.3.11 and depicted in Figure 6.3.1-2, the 125 VDC class lE power system consists of six independent and physically separate load groups. Each load group includes a battery, a battery charger and DC distribution center. The battery-chargers of load group channels A, C and division I are powered from division 1 of the 480 VAC power system. - Similarly, the battery chargers of load group channels B, D and division 11 are powered from division !! of the 480 VAC power system. Furthermore, the ESF equipment which is loaded on the 4.16 KV or-480 V bus is provided with redundant tri) coils. Control poter for the trip coil circuitries is assumed to be 03tained from the-125 VDC buses A and I for division I equipment and from 125 VDC buses B and !! for division 11 equipment. The HPSI pumps in loops B and D are powered from '
division 11 power system, thus the control power for these pumps can be from 125 VDC bus B or 11. Therefore, in the event of a failure of 125 VDC bus B, the HPSI pumps in safety injection loops B and D do not necessarily become unavailable. For the HPSI pumps in loops B and D to com>1etely fail, failure of both the 125 VDC buses B and !! must occur. T1e fault tree -(Figure 6.3.6-13) developed for failure to deliver safety injection flow with 125 VDC bus B unavailable correctly models such logic.
b L
(L
~ - - - - . - -... ~ ~
.0uestion 720.79 Based on the fault trees shown -in Figure 6.3,7-3 (8 of 9 and 9 of 9), loss of 125 VOC control power. (bus B) will lead to loss of flow from emergency
' feedwater (EFW) sub-train B2 because of the failure of motor-driven EFW pump-104. In the fault tree (Figure 6.3.7 7) constructed for failure to deliver sufficient EfW tc either steam generator with loss of 175 VDC vital bus B. thera'.?m, .hould GATE 2110 contain just one event "lors of flow from EFW train Bl"? Mes ti.e failure probability of the event, PA0G08EX, taken to be 1.0 in the faalt tree quantification? Why are the availability tne emergency feedwater tanks not modeled in the fault tree?
, Resnonse 72Ll.2 As described in subsection 6.3.1.1.3.11 and depicted in Figure 6.3.1-2, the 125 VDC class lE power system consi-ts of six independent.and physically separate load groups. Each load group includes a battery, a battery charger and DC distribution center. The battery chargers of load u -
- group channels A, C and division-1 are powered from division I of the 480 VAC power system. Similarly, the battery chargers of load group channels B, D and division 11 are powered from division 11 of the 480 VAC power system. Furthermore, the ESF equipent which is loaded on the 4_.16 KV or 480 V bus is provided with redundant trip coils. Contra power for the i trip coil circuitries is assumed to be obtained from the 125 VDC buses A j and 1 for division I equipment and from 115 VDC buses B and 11 for division <
41-equipment. The motor-driven EFW pump 104 in sub-train B2 is pow red l from division 11 power system, thus the control power for these pumps car, be from 125 VDC bus B or II. Therefore, in the event of a failure cf 125 ,
VDC tus B, this pump does not necessarily become unavailable. For the EFW l pump-104 to completely fail, failure of both the -125 VDC buses B and 11 !
must occur. The fault tree (Figure 6.3.7-7) developed for failure to deliver sufficient EFW to either steam generator with 125 VDC bus B unavailable correctly models such logic. The failure probability of the event PA0G08FX was accordingly derived, and not.taken to be 1.0, in the Sult tree quantification. The (un) availabilities of the emergency Jdwater tanks was _modeled in the developed events- ALOSFPTINDO, loss of
..ction flow to turbine-driven EFW pump, and ALOSFPMINDO, loss of suction
< bw to motor-driven EFW pump.
Qgstion 720.80 Loss of one component cooling water division, loss of 125 Vdc vital bus,
- or loss of a 4.16 KV vital bus all have some impact on the availability of the containment spray system. No clear treatment of the relevant impacts, however, can be found in the fault trees (Figure 6.3.13-2) developed for failure of the containment spray system. Please explain why they are not i modeled, l Rescense 720.80 Treatment of relevant impacts of loss of one component cooling water division, loss of 125 Vdc vital bus, or loss of a 4.16 KV vital bus on the availability of the containment spray system was considered when developing fault tree for failure of the containment system. However, some of the fault trees detailing the-logic were inadvertently left out of the report and the rest of the fault trees were included in other system or special functions. The impacts of loss of offsite power, loss of one division (8 or 41) of component cooling water and loss of a 125 VDC vital bus are modeled in the development of the fault trees for the special function
" Failure to Successfully Cool the IRWST given a Loss of Offsite Power" (Figure 6.4.2-6), " Failure to Successfully Cool the IRWST given Loss of a Component Cooling Water division" (Figure 6.4.2-7) and " Failure to Successfully Cool the IRWST given Loss of a 125 VDC Vital Bus" (Figure 6.4.2-8) in volume 2 of DCTR-RS-02, Rev. O, respectively. Only the fault tree for " Containment Spray (CS) pump CSF-101 inoperable (given loss of a component cooling water division), PGIB15CX, is included in Figure 6.4.2-7 (on page 6-846). Other relevant fault trees attached herewith are:
PJ0B06DX : CS pump CSP-101 inoperable given a loss of Offsite Power PJ0B16DX : CS pump CSP-201 inoperable given a Loss of Offrue "ower PJ0B16EX CS pump CSP-201 inoperable given Loss of a 125 VDC Vital Bus The consequences of losing a 4.16 KV vital bus are identical to those for i
loss of a component cooling water division and, therefore, the case for losing a 4.16 KV vital bus was not explicitly modeled. Instead the fault trees developed for loss of a component cooling water division were used.
Similarly, the corresponding faelt trees for the Residual Heat Removal (RHR) or Shutdown Cooling System (SDC) pumps needed in the special function
" Failure to Successfully Cool the IRWST" were inadvertently left out of the report. These are also attched.
PJ0B05DX : RHR pump RHRP-101 inoperable given a Loss of Offsite Power PJOB15DX : RHR pump RHRP-201 inoperable given a loss of Offsite Power PJOB15EX : RHR pump RHRP-201 inoperable given Loss of a 125 VDC Vital Bus The fault tree developed for unavailability of RHR pump-101 given loss of a component cooling water division is presented as PGIB14CX in Figure 6.4.2-7 (on page 6-845).
Response 720.80 (cont.d)
J Also, it is to be noted that, as described in subsection 6.3.1.1.3.11 and depicted in Figure 6.3.1-2, the 125 VDC class IE power system consists of six independent and physically separate load groups. Each load group includes a battery, a battery charger and DC distribution center. The I battery chargers of load group channels A, C and division I are powered from division I of the 480 VAC power system. Similarly, the battery chargers of load group channels B, 0 and division 11 are powered from division 11 of the 4B0 VAC power system. Furthermore, the ESF equipment which is loaded on the 4.16 KV or 480 V bus is provided with redundant '
trip coils. Control po ver for the trip coil circuitries is assumed to be obtained from the 125 VDC buses A and I for division I equipment and from 125 VDC buses B and 11 for division 11 equipment. The CS and RHR pumps in train B are powered from division 11 power system, thus the control .
power for these pumps can be from 125 VDC bus B or 11. Therefore, in the l event of a failure of 125 VPC bus B, these pumps do not necessarily become i unavailable. For the CS or RHR pump to completely fail, failure of both the 125 VDC buses B and 11 must occur. The fault trees, PJ0B16EX and 4 PJOBISEX, developed for unavailability of CS pump and RHR pump, given loss l of a 125 VDC bus B, correctly model such logic. l l
l
as_LA, p._4_ _ - - L.. . A d .Al sA m 4w ded k # E J.Amr3 meme h 80 m>==d--A =6.-.4-a-3-4+%.m. '+dra dff_a+s'.4&Ap.24,- 40 4 %-T4M e na.hAdM 4 ,JJ's+-d aM J A44.W 254 5- . M.m te .4 -J_h4.Jrd 1 .$
g~2 g
}
Y L gJ B gps a ._ _
B 89 b d
l :
BW
$N" 8 rBR A X -
P IMg -
O- 3: .o 6
. ta @ $< x 8a .8
$sz
[o- hl5]N c7 855 B -
gD-
,p g ,= a
$g - -
byR 9 89 6[ -
d 0 1 @4 b 37g .g+ a m@
3 i
f 8 L sa 6 ggg e 38"
-4t : z<
$6 -
S O
5 28 9
_ g3 0 __
B _
t im 6 u /pe 68b yM" mf+w moe l; 80 g >
l l' . .
.. . ~ - . ... ..., . .-- .. -.-. - =. - . , , - -, ,.-~. _ .., -- . .. . . . . - .
. l y! f 1!

! ,! t I , 1 , j:
o th ! i" l7 r k, i
k fi i 4 :r hf
~.
C F
OSS CP 2
A V x u
! I 72U C P P
G
. -N
J 1 _
P1R 0 2
L0 P U2O S .-
i PPT C K
S SCS u r
C L c .
A _
F ._
X 1 0S 2 SR 3 .
3E -W D POL 'V t L2M M/
3 3
2 WC CC2 .
6 S/O C
X t2O C X 1 WP D 6
1 N3T A2- 2 3
MN O
E D
B l
P MLT E i Mf 2 i
&ROF!S N
/' W Vt v 4 O UAIS PVF A P
C .O C
W S V SWD C
P .
- J SP& C O.. C OO LL P CU N F CR)
DE VWB
&T RM
+
3 5ON 2
1PI 6
8 OET A A 1 0
i SPS LRT -
SPY OUS A TO R D
E .
T R P LS C SNO S
OF OC(
L T R R ENB A 1 0 WO -
P1TS 2 O S,U \ 2 N0 o P PB E U2 s G i
PPO ST C
C I
&OV M -
N B
SC A CK E C SL P' S S F A
G SO1 F OT4 L
[
,, ' ' i;! i , i l ! t # .
J a k %.a+ -- x a 86a..,n-,A +SSW-wa,t 6~A.s, Km *OL29r E nA J n --e-fin., ny MS-t 65A
b-A=4b-b AM--+b 4'ba-hA'"OA M W"'A'**A -"AYO A 4 0 A- **A^M*' "'4^m" ~~A-= +
W
, I.4' .
4 4
5 l
gg-
! y-
- b
u. 4 h b 0 5 ;
2:%" -
h 8s b a
l i
p( - 5 :^ . Qui LtJ i rRs j h,sm 59 bz m e ,g g
g g- gg GT ;ysg - -
4 g- b8[z g-su a sm
- o_ 8, ge bgj g g. W2-g g-
~
BENS
i)
Bg n
z se- .
9. $ 5 / 3; th
$o.8" -
byits R d [. .,
83 [ gog '
d s"* -
n
+ .,
& * * .. - , , - -- , .-r e , - = , . . , a , r% re , e-w.r' - gy e- 1 7., r
m -
f l
~
l.
m F 2 e OmDS A x
i RM i u
F U C p C2P' a
N P U 1 0
M1R 0
1 P
i U1O PP T 1 f
R m RHS K V
RHRL P A J
~
F 1
L1M 3I 1 A./ U I0O N3T -
G A1 - U ME l I VL /
W .O 07 CS X PE RL CVLO. . C D HB RAP VN 5 LO x 0
0g PAO MV- L 5 p
^ -
B UMH P UIT f' T R WC W
- O RW1 P
P A M1T 1
01 CC 1 C X e J H0 R1 U10S P
MN S P i PPORT P
H R I FOO OR S 1
0 N
RH G FI 3 S M C HRS W P SWD P
- R L J OO A LL F F OR D E )A VW FT 5 ON ORM D 2 1PI A 6
A I
SPS OE T X' 5 2 E F'LR 1
C SP T OUS LS Y f /Tf A G
OO S
RR T
SNO OF
/ D E
P OC(
L
.L
- R
- AT N ENA
'. V AAI WC 1
0 O
_ L'NM 1
P P .S U
2 A
UU
_ P R PB 1 C
1O i H R
I FM N R0T OOV CK B
E H1 W P RPCRL P
J S 6 SO1 HD OT4
_ R L
l > ',
l bkmh -
m gne s Mtd 3
tm9 a
~
55> > 0
.) g bvi 4 6
o>d s!=
X Lu a O Ed 5 -
88 n LO "kb q $5" 0 "s, 5 ro - ggg o, a=1 4 !3 mee
! g- gg g
sgg s e ,
maa e ..
Q_ m o
beg n") a5 /s
$$5 N I$g Bo'b hb
_io
,- 4 Z CD SS
'E ^$ $22 s
g2 m s g- bsg m
j -
s
& 8,e~* .
i 8
,y.
4 ...
4 u
PJOB15EX I
EHRPWR ER. ,
(125VCO . B N/A) c
(
P ,tSEX I
1 1 l l f RHR PUMP -LOSS OF ' CCW MANUAL ~ RHR PUVP ~CCF OF RHRP201- ' SUPPORT VLVS. V-230/231 RHRP201- 2 RHR.
FALS TO START SYSTEM - N.O.' DUE TO ME. FALS TO RUN PUMPS
l QV i GATE 2532 JPMJGRHRP201 CV?#W230-231 PWR4PP201 SMXA2 I I LOSS OF COWER LOSS OF 125VDC LOSS OF CCW TO COMP. ON - CONTROL POWER FLOW FROM CCW 4.16KV BUS B (FOR TRAIN B) DM310 2 .
PEBNOIS1 PEEtCB7 b
PC4to1EX i
l l
.= . _ . - . - _ _ - _ _ . .
_. - - . - - - . .- - - - - . - -.. -. .._.- .- -_- -.= .-
Question 720.81 Sections 1.2, 1.3, and 4.3 - It is stated that " bounding site characteristics were used for the evaluation of external events such as seismic and tornado strike events....".
However, it appears that the single seismic hazard curve (Figure B4.3.1-2) used in the CESSAR 80+ PRA has a much lower return period than those in the EPRI Requirements Document hazards curves at the higher acceleration levels. In addition, at sites where EPRI and Lawrence Livermore National Laboratory (LLNL) have both made hazard curve estimations, the LLNL curves have tended to be an order of magnitude greater in frequency for a given acceleration level. What is CE's basis then for stating that the single hazard curve used in the CESSAR 80+ PRA is bounding? Does CE claim that the annual seismic severe accident core damage frequency of 1.2E-6 presented in Table Bl.3-1 is best estimate or conservative?
Because of the wide range of uncertainty within the earth science community regarding earthquake potential and ground i motion estimation in the central and eastern United States, the seismic hazard curves developed by both EPRI and ( LLNL) ,
both with four and five ground motion experts, should be used for the seismic hazard estimation.
Response 720.81 The seismic analysis information, including the seismic hazard curve, used in the System 80+ seismic PRA was taken from the early version of the EPRI ALWR PRA Key Assumptions and Groundrules Document and was selected considering the PRA goals (e.g., core damage frequency). It is C-E's understanding that analysis using the LLNL hazard curve may not be necessary, pending NRC's final position on demonstrating plant safety beyond the safe shutdown earthquake (reference: Meeting with NRC on November 26, 1991). Should analysis using the LLNL hazard curve be required, however, it is likely that a new core damage goal would be selected.
Question 720.82 Section 4.3 - How is the buckling failure mode of steel containment incorporated in the containment fragility descriptions for the System 80+ PRA?
Respor,se 720.8?
The buckling failure mode of steel containment was not incorporated in the containment fragility description for the System 80+ PRA. C-E is currently updating the System 80+ PRA.
The potential impact of this . failure mode will be evaluated as part of this update.
l
Ouestion 720.83 Section 2.9 - What is the basis for the primary feed and bleed success criterion? In general, are the success criteria in the System 80+ PRA "best estimate" or design basis?
Response 720.83 The design basis for feed and bleed was that the reactor vessel level- would remain at least 2 feet above the top of the core if feed and bleed operation was initiated within 30 minutes. Using C-E's design basis codes, the bleed valves were sized so that this could be accomplished using one bleed valve and two HPSI pumps. Additional MAAP analyses demonstrated that feed and biced would be successful with one bleed valve and one HPSI pump. The upper portion of the core would uncover briefly, but the fuel temperature remained below 2200 Dog F. This was the basis for the feed and bleed success criterion in the System 80+ PRA. In general, the success criteria used in the System 80+ PRA are design basis criteria.
l However, in cases where the design basis success criteria were felt to be overly conservative, sdditional "best-estimate" thermal hydraulic analyses were performed using MAAP or CENTS to determine if less conservative success criteria would be viable. These "best-estimate" success criteria were used as appropriate.
Ouestion 720.84 Section 3.1.8 - For loss-of-of fsite power and station blackout sequences, the Standby AC Power System appears to make a significant contribution to accidcnt prevention, yet it does not appear in Table.B3.2-2, Component Importance for System 8 0+ PRA. Why not? Since the system is not safety grade, what l is its assumed availability and reliability? What is this o
system's mathematical measure of importance?
Response 720.84 Use of the Standby AC Power System to provide AC power was i treated as a recovery action. The basic event used to l represent failure of this system is "RCVRSBAC". As presented l in table 5-7 of the System 80+ PRA report, DCTR-RS-02, Rev. O, January,1991, the unavailability used for this event is 5.0E-2 -with an error f actor of 3.0. As presented in table 8.4-1 of l the System 80+ PRA report and table B3.2-2, the Fussel-Vesely
( importance measure for this element is 1.84E-2.
l l
l l
l l
Ouestion 720.88 Section 4.2.3.4 - What is the "best estimate" recovery factor of offsite power after a major tornado?
Repnonse 720.88 l
In the System 80+ tornado strike analysis, it was assumed that -
a tornado strike on site would result in a loss of offsite power lasting longer than the base 24 hour2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months mission time. j Therefore, recovery of offsite power was t.ot credited in the System 80+ tornado strike analysis.
Question 720.89 i
Section 4.3.1 - The generic fragility values assigned to i' components in the System 80+ PRA dif fer from those recommended
!- in Appendix A (Table A.3.4) of tha EPRI Requirements Document.
Clarify these differences. Note that the EPRI document has median capacity factors for different types of sites.
E9&ornse 720 d1 The generic component fragilities used in the System 80+ PRA were based on an early version of Appendix A of the EPRI Requirements Document. C-E is currently updating the System 8 0 + PRA . The generic component fragilities will be updated as appropriate for this analysis. See the response to Question 720.81.
Ouestion 720.92 Section 4.3.2 - Table B4.3.2-1 gives core damage frequency contributions for different sequences. Describe how these frequencies were derived. Provide the top fifty seismic cutsets. Does "ERF" mean " error Factor"? If so, how was it obtained.
Response 720.92 l The response to Question 720.62 describes how the seismic core damage frequencies were calculated. The cutseta for the dominant seismic sequences are presented in Tables 8.3.2-2 through - 8. 3. 2-15 in the - System 80+ PRA Report, DCTR-RS-02, Rev. O, January 1991. "ERF" does mean " error factor".
l i
Q11pp11pn 720. 93 Section 4.3.2 - provide or reference the random failure rates and human error rates used in the seismic PRA analysis.
Detail the differences in these rates between internal events and seismic events. Describe the human errors considered in seismic events.
Response 720.93 The random f ailure rates and human error rates used in the seismic PRA analyses are presented in tables S-2 through 5-7 of the System 80+ PRA report, DCTR-RS-02, Rev. O, January, 1991. The same random failure rates were used for both the internal events and seismic analyses. The same human error rates, to the extent that they overlapped, were also used for both the internal events and seismic analyses. The human actions unique to the seismic analyses tended to be special recovery actions such as reactivating equipment after relay chatter failures or isolating component cooling water to equipment whose jacket cooling heat exchagers had failed.
Question 720.94 Section 4.3.2.4 - Provide the basis that one to two hours will be available for operators to reclose switchgear breakers after seismically-induced relay chatter.
Response 720.94 The seismic event. is the initiating event. Therefore, the relay chatter failures occur at _ time zero and the plant systems fail to respond to the transient because of the relay-chatter failures. Relay chatter failures occur at relatively low seismic acceleration levels. Thus, in-the sequences of concern, there is little if any additional seismically-induced damage.- The transient can be terminated be resetting the relays and reactivating the safety systems, primarily the emergency feedwater system. This is equivalent to a standard transient in which -secondary side ' heat removal must be established within approximately 2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months of the initiating event in order to prevent core damage.
I

r-- --+-g .mp y-y , af

Ouestion 720.96 Describe how the System 80+ PRA and its insights were used in identifying equipment to be tested / evaluated in the ITAAC program. Describe how test specifications were influenced by the PRA.
Response 720.96 The PRA was used as an integral part of the design process to gain insight into vulnerabilities and to evaluate design features proposed by the EPRI Utility Requirements Document. Items to be selected for ITAAC will be based on design evaluations which may include PRA insights but , in ge*)eral, the PRA will not be used to define the ITAAC program. The PRA will however, provide significant input to the Reliability Assurance Program.
Question 720.97 Provide a comparison of the estimated core damage frequency, conditional containment failure probability, and offsite consequences for the System 80+ design to the Commission Safety Goals, if no credit is taken for operator actions other than control room-based alignment of alternative core cooling methods.
l l Response 720.97 I
L C-E is currently updating the System 80+ PRA. The requested comparison will be provided in the updated PRA report.
l I
1 Y 7
Question 720.98 The bottom line core damage frequency estimate for the System 80+ design appears to be very low. Given these low estimates, address how the System 80+ PRA ovaluated initiating events that have a lower frequency than those normally-postulated, but may have more serious consequences. Examples include the questions, "What does a lE-5/yr steam generator tube rupture initiator look like and how is it handled in the PRA?," "What do various 1E-5 common cause failures look like and how were they looked for and evaluated in the PRA?," and "What is the effect of multiple failures / equipment outages during modes other than full power?."
Response 720.98 The System 80+ PRA does not address initiating events with very low frequencies because there little information on what these initiating events might be. This approach is consistent with EPRI guidance and generally accepted methodology.
Question 720.99 Describe how the PRA has factored in the possibility of the need to deal with appreciable fuel damage in conjunction with RHR operation and waste processing?
Response 720,99 The C-E System 80+ PRA does not address the possibility of the need to deal with appreciable fuel damage in conjunction with RHR operation and waste processing. In the System 80+ PRA, sequences involving appreciable fuel damage are assumed to be core melt sequences and evaluated as severe accidents.
Ouestion 720.100 Describe how the PRA modeled control systems and - control system fail' Ire modes.
E_esnonse 720.100 The System 80+ PRA does not model control system f ailures. It was assumed, that with the improved component control system, control system f ailures would have minimal impact. C-E is i currently updating the System 80+ PRA. The potential impact
( of control system failure will be reassessed during this update.
Question 720.101 Provide a discussion of how the System 80+ PRA was used to >
identify equipment / structures / components to be covered by the Reliability Assurance (RAP).
Resnonse 720.101 A Reliability Assurance Program (RAP) plan is being developed in response to the " Request for Additional Information, Combustion Engineering System 80+, Performance and Quality Evaluation Branch, Generic Safety Issue II.C.4; Reliability Engineering". The System 80+ PRA is being used as the primary resource for development of this RAP plan.
Structures, systems, and components and reliability criteria in the RAP will be consistent with those in the PRA.
Question 721.1 The credibility of an HRA analysis is highly dependent on the mix of expertise in the analysis team. In this regard, please provide information on the makeup of the team that performed and reviewed the HRA portion of the System 80+ PRA.
E_esnonse 721d i_ For the System 80+ PRA, C-E performed a preliminary HRA l analysis consistent with the EPRI HCR model and the methods
! described in the Handbook of Human Reliability Analysis. The analysis team consisted of the systems cr. ysts with assistance from engineers holding an SRO. Most of the HRA was based on generic System 80 information and generally accepted operating procedures for C-E designed plants (e.g., Emergency Operating Guidelines in CEN-152). C-E recognizes the NRC's concern with the limitations of this analysis.- C-E is currently updating the System 80+ PRA. This update will include improved and detailed HRA.
Questions 721.2 - 721.17
[These questions address the manner in which human reliability was included in the System 80+ PRA)
Resnonse 721.2 - 721.17 i Combustion - Engineering agrees with NRC's comments and l will resolve them in the revised PRA. It is expected that. C-E/NRC meetings in the interim will ensure that resolutions in the revised PRA are adequate. This l-updated HRA will use the most up-to-date methods to the extent possible and will draw on more recent human reliability data such as that provided via NUCLARR.
. - _ . . . . , . . - ... . . ~ . , , , . . . - -
Resoonse 721.2 - 721.12 (cont.)
i A new subsection will be added to each event analysis section to describe the human actions and performance shaping factors applicable to each initiating event. In addition, Chapter 5 will be expanded to describo the quantification of each operator action credited in the PRA.
i l
l l
l
.

ML20094E943

Text

Navigation menu

Search