Text

Rev 1 to Auxiliary Feedwater Sys Reliability Analysis Final Rept
	ML20129D548
Person / Time
Site:	Davis Besse
Issue date:	12/31/1981
From:	Holderness J; EDS NUCLEAR, INC.
To:	;
References
	02-1040-1095, 02-1040-1095-R01, 2-1040-1095, 2-1040-1095-R1, NUDOCS 8507300083
	Download: ML20129D548 (152)
	v • d • e

._

DAVIS-BESSE UNIT !!o. 1 AUXII.IARY TEIWATER SYSTCi RZI.IABILITY AllALYSIS --

ri -

.g FINAL REPOR*

l N l33l l . NUw.. g,]

Prepared by:

ICS Nuclear Inc.

for Toledo Edison Company Ncambe r, 1981 Papor tio. 02-1040-0195

.% vision 1 0507300003 011231 PDR ADOCK 0D000346 p PDR

EDS NUCLEAR INC.

NEW YCRK RECT 0NAL OFFICE REPORT APMCVAL COVER SiiEET Client: Toledo Edison Comoany Project: Auxiliary Feedwater System Reliability Analysis Job Number: 1040-003-671 Report

Title:

Davis-Besse Uni No. 1 - Auxiliary Feedwater System Reliability Analysis Final Report Report Number: 02.io.to.inos Rev. o The work described in this Report was performed in accordance with the EDS Nuclear Cuality Assurance Program. The signatures belew verify the accuracy of this Report and its compliance with applicable quality assurance requirements.

Prepared By: 'l

/ _/LM OF [ N b I CT M S Date: h /h /8/

ucM 1 A flh k 8/

Reviewed By: Date: k Approved By: 6t[ dM -

Date 9 /c /

I i Cencurrence By

,%w lh //[. Ye '

Date: 'i //"[ .31 Q Regional Quality Assurance .4anager f &

4 REVISICN RECORD r

Rev. Approval No, prepared Reviewed Approved Concurrence Date Revisien I

t MM p294/w JL.~ A u la t V p a @\[% .(([ k ' H SI M *

\

I -

i ,

Report No. 02-1040-1095 Revision 1 Page i TABLE OF CON"CIN Page 10 SL%'tARY l 2.0 SCOPE AND CBJICTIVIS 6 2.1 Background 6 2.2 Cbjectives 6 23 Scope of Work 7 3.0 SY3 TEM DESCRIPTION 9 31 ATWS Safety Punction 9 3.2 Pre *MI System Configuration '. 0 33 Post-TMI System Cor. figuration 11 3.4 Third Train System Configuration 13 3.5 Analysis-Based System Configuration 14 3.6 ATW3 Support Systems 15 4.0 MIT:500CLCGY 29 41 System Fault Tree Development 29 4.1.1 Tault Trees Ceveloped 29 4.1.2 Tault Tree Methodology J1 4.2 04ta Analysis 33 4.2.1 Industry Data Review 33 4.2 2 Review of Cavis-Desse Experience 34 4.2.3 Recommended Cata Base 39 1 4.3 System Unavailability Analysis 40 4.3.1 Ouantitative Analysis of 40 Fault Trees 4.3.2 'Jncertainty Analysis 42 4.3.3 Importance Ranking 42 4.4 2nitiating 27ent Analysis 43 4.4.1 Frotaency Istimates 43 4.4.2 Factors Influencing Event 45 Frequencies 4.5 Combined System / Event Analysis 46 l

l

Repor~. No. 02-1040-1095 Revision 1 Page 11 TABLE OF CONTENTS (contirtued)

Page 5.0 RESUL73 73 5.1 Relative Unavailability Ranking of AFW3 73 Configuratio ns 5.1.1 Differences Among Event 73 Categories 5.1.2 Differences Among ATWS 73 Co nfigurations 5.2 significant Centributors to Arus 74 1 Cnavailability

!.3 Kesults of Combined 3ystem/Ivent Analysis 76 5.4 Potential Cceman Cause Contributors to 77 AFW3 Cnavailability 6.0 CONCLCSIONS 3:

7.0 RETERINCt3 95

Report No. 02-1040-1095 Revision 1 Page til i LIST OF TABLES PAGE l

l-1 Results of AFWS Fault Tree Analysis 4 1-2 Overall Figure-of-Marit for APWS Reliability 5 3-1 Steam-reedwater Rupture Control System Actuation la 3-2 AFWS/ Electric Pcwor System Interfaces 19 4-1 System Saf ety Functions 47 4-2 Plant System Designator 49 4-3 Ccmponent Code 50 4-4 Failure Mode Code 5 4-5 AFW3 Reliability Study Tailure Oata 53 4-4 Failure Crperience for Davis-Besse Valves 63 4-7 Failure Experience of Oavis-3 esse Auxiliary 64 Teodwater Pumps 4-9 railure Experience of Davis-Pesse Diesel 65 1 Gene rato rs 4-9 Initiating Events Challenging AFWS 66 4-10 Initiating Event frequency Estimates 67 5-1 AFWS Unavailability 70 5-2A Significant contributors to ATW3 Unavailability 77 Pre-m! Configuration - Category 1 5-2n 317nificant Contributors to ArWS Unavailability: 80 Pre-St: Confi7uration - Categorf :

5-2C 317nificant Contributors to ArW3 Unavailabiltty di Pre-m! Configuration - Categorf 3 5-3A 317nificant Contributors to Aru3 Unavailability: 6:

Post-OtI Oonfiguration - Categor*r 1 5-33 317nificant contributors to ArW3 Unavailability: 13 Post = :t! Confi7uration = ste ,o ry 2 -

s. . ..

Repo rt No. 02-1040-1095 Revision 1 Pa le iv L:3T OF TA3Lrs (cant. ) PAGE 5-3C Significant contributors to AJMS Unavailability: 84 Post-OtI Configuration - Categorf 1 5-4A Significant contributors to AJMS Unavailability: 95 Third Train Configuration - Categorf 1 5-4D Significant Contributors to AJMS Unavailability: 86 Third Train Configuration - Categor/ 2 1

5-4C Significant contributors to AIMS Unavailability: a7 "hird Train Configuration - Catagory 3 5-5A Significant Contributors to AJVS Unavailability: 80 Analysis-Based Configuration - Categorf 1 5-!D Significant Contributors to Ards Unavailability: 99 Analysis-Based Configuration - Categorf 2 5-5C Significant Contributors to Ards Unavailability: 90 Analysis-Based Configuration - Categorf 3 5-4 Combined Systa:a/tvent knalysis Pasults 91 5-7 P41ative Significance of the Pre-?t! - Poet-UtI Changes 92 L

hport No. 02-1040-1095 Revision 1 Page v LIST OF FIGURES PAGE 3-1 Pre-T:C AWS Configuration 20 3-2 Pre-TMI Main Steam System Configuration 21 3-3 Test-TMI AWS Configuration 2 3-4 Post-TMI Main Steam System Configuration 23 l

l 3-5 Post-TMI Stanup Pump Configuration 24 3-6 :takeup System Configuration 25 1 3-7 hird Train Configuration 26 3-8 AWS Analysis Based Configuration 27 3-9 Startup P=p Analysis Based Configuration 28 4-1 Davis-Besse AM Fallability Program Flow Diagram 68 4-2 Symbols Used in Fault Trees 69 4-3 System / Event Analysis :tatrix 7:

Report No. 02-1040-1095 Revision 1 Dage vi l

l AT""ACIDEN"S Drawing No.

APJS (Pre-OtI) Fault Tree 1040-003-001

!!ain Steam System (Pre-OII) Fault Tree 1040-003-002 Electrical System (Pre-24I) Fault Tree 1040-003-003 APJS (Post-OtI) Fault Tree 1040-003-004 Main Steam System (Post-UtI) Fault Tree 1040-003-005 Electrical System (Post-Di!) Tsalt Tree 1010-003-006 Start-up Pu=p (Post-24I) Fault Tree 1040-003-007 Start-up Pump with Feed and Bleed 1040-003-008 APJS (Analysis-Based) Fault Tree 1040-003-009 Main Steam System (Analysis Based 1040-003-010 Fault Tree 1 Start-up Pu=p with Feed and Bleed 1040-003-011 (Analysis Based) l l

l

l l

l Report No. 02-1040-1095 Revision 1 Page 1 1.O

SUMMARY

This report presents results of a reliability analysis of the Davis-Besse Nuclear Power Station, Unit No.1 auxiliar/

feedwater system (AWS). This analysis, performed by EDS Nuclear, Inc. for Tbledo Edison Company (T Co), is in suppor.

of the TECo commitment to the U.S. Nuclear Regulatorf Co= mission (NRC) for a continual review of auxiliarf feedwater system reliability. This analysis also provides a comparative probabilistic risk assessmant basis for various design modifications to further upgrade the AWS for added reliability and performanca. The analysis has resulted in the quantification of AWS unava21 ability, identification of major contributors to system unavailability and recommendations for system modifications to =inimize unavailability.

The analysis is based on development of fault trees for the AWS and other plant systems supporting the AWS saf ety function, i.e. , the deliverf of cooling water to one or both steam generators whenever the main feedwater flow has been interrupted. The fault trees depict the logical relationship between the failure to deliver sufficient feedwater to the steam generators and the basic mechanical, electrical and hu=an factors which may cause an individual system compenent to fail. Failure data, derived f rom industry sources and reviews of Davis-Besse plant-specific cperating experience, are used to assign probabilities of failure to the basic ccmponent f ailure mechanisms. These basic event probabilities are then propagated through the fault tree, using Boolean algebra, in order to derive a probability for f ailure to achieve the AP4S safety function. For the purpose of this report, failure to achieve the AMS safety function is defined as "ANS unavailability", even though the f ailure may result frem failures in other plant systems which support the AWS safety function.

Initiating events wnich challenge the AWS can be conveniently categorized as follows:

. Categorf 1- Events in which the main feedwater flow or reactor coolant system forced circulation is interrupted, but offsite electrical power is available to the plant.

. Categorf 2- Events in which effsite electrical power to the plant is interrupted.

. Categorf 3 - Seismic events.

Raport No. 02-1040-1095 Revision 1 Page 2 The AWS unavailability is determined for each of these categories of initiating events. 3e annual frequencies with which these events occur are estimated from industr/

experience and from reviews of Davis-Besse operating history.

The frequency of the initiating event is then multiplied by the AWS unavailability. The result is the annual frequency with which the AWS is unavailable when called upon to perform 1*.s safety function. This is the overall figure-of-merit used to judge the relative reliability of various AWS configurations.

"he following four AW S configurations are considered in this analysis:

" Pre-StI" -

The AWS configuration that existed in March, 1979.

" Post-tt!" -

The AWS configuration that contains TMI-related plant changes, including those planned to be implemented in the 1982 refueling outage. Included in this configuration is a written procedure for fulfilling the AWS safety function using the main feedwater startup pump, reactor coolant system makeup pump and power operated relief valve as a backup to the AWS.

""'hird Train" - A potential configuration, which utilires the main feedwater startup pump, in an altered aligreent, as a backup third train of auxiliary feedwater.

" Analysis- A configuration which incorporates lt 3ased" reccc:mendations resulting frem this reliability analysis, but does not include the realigned startup pu=p.

The unavailabilities of each AWS configuration for the three categories of initiating events are summarired in Table 1-1.

Se overall figure-of-merit, the annual frequency with which each AWS configuration is unavailable when challenged, is presented in Table 1-2.

The following conclusions are reached as a result of the analysis:

The Pre-ttI configuration unavailability is dominated by potential human errors, primarily in valve misalignment.

L _ _ _ _ _ _ _ . _ _ - - _ _

Report No. 02-1040-1095 Revision 1 Page 3 The Post-TMI configuration incorporates many changes to plant procedures which diminish the likelihood of human errors. The unavailability is then dominated by

=echanical failures, primarily associated with motor operated valves.

The third train configuration reduces the AIWS unavailability by over an order of magnitude for Category 1 and Category 2 events through the addition of a third redundant train. .

The analysis-based configuration provides over an order of =agnitude improvement in AIWS availability by addressing those specific mechanical factors and procedural limitations which dominate the Post-MI results.

Significant improvements in the Davis-Besse AIWS reliability have already been achieved since the original NRC requests (1) 1 (2) for a review and upgrade following the Three Mile Island, Unit 2 ( 3I-2) event in March,1979. Further improvements are planned. Cf the alternatives examined in this study, the third train and analysis-based configurations offer the greatest i=provement in AIWS reliability. The cost of the third train configuration is relatively high. The analysis-based alternative offers an even greater improvement in system reliability, and its associated costs are likely to be relatively low. The design and procedures modification of the analysis-based configuration are now planned as a means to enhance the Davis-Besse AIWS reliability and performance.

Report No. 02-1040-1095 Revision 1 TABLE l-1 Page 4 Results of AEWS Fault ?ree Analysis ArdS Unavailability (per demand) l Category 1 l Category 2 l Category 3 APdS Confieuration l Drents l Drents l Dre nts

! I l Pre- mI l 3.3 x 10-2 l 4 1 x 10-2 l 8.8 x 10-2 I I I Post-mI l 6.6 x 10-4 l 5 5 x 10~3 l 1 9 x 10-2 I I I nird Train l 4.5 x 10-5 l 1,4 x 10-4 l 1,9 x 10-2 l I 1 Analysis-Based l 3.3 x 10-5 l 9 3 x 10-5 l 1,1 x to-2 g l i !

i o ._

Report No. 02-1C40-1095 Revision 1 TABLE l-2 Page 5 Overall Ficure-of-Merit for AWS Reliability l Frequency of AWS Chavailability l When Challenged (yr'l)

AP4S Confieuration I("btal of all event categories) l Pre-SiI l 8.2 x 10~2 I

Post-SiI l 3.3 x 10'3 1

2 ird Train l 2.2 x 10~4 l

Analysis-Based l 1.4 x 10'4 1 l

u , .

e e

b. . . . . _ , . . . _ _ _ . _ _ . . . . _ . _ _ , _

Raport No. 02-1040-1095 Revision 1 Page 6 20 S$PE AND OBJECTIVES

2.1 Background

As part of its review of the Three Mile Island, Unit-2 event, 1

the NRC issued, on May 16, 1979, a Confirmatory Order (l) to the Toledo Edison Company as part holder of the operating license of the Davis-Besse Nuclear Power station. Unit No. 1.

This order required, in part, that the licensee review all aspects of the safety grade auxiliary feedwater system to further upgrade components for added reliability and pe rfo rmance . On July 6, 1979, the NRC issued a letterI2I lif ting the above Confirmatory Crder, allowing Davis-Desse Unit No. I to return to power.

The safety evaluation attached to that letter indicated that the NRC would at some future time require system diversity through the installation of an additional 100 percent capacity notor operated auxiliary feedwater pump, or an alternative acceptable to the staff.

In reviewing the NRC's intended purpose for such a modification, and relating it to the magnitude of the cost impact, TEco determined that a quantification of the relative risk reduction actually provided by such a modification is apprcpriate.

2.2 Cb*eetivee The overall objective of this analysis is to evaluate the reliability of the Davis-Itisse Unit ni.1 auxiliary feedwater system in delivering feedwhter to one, or both, steam generators whenever main feedwater is interrupted or whenever reactor coolant system forced circulation is interrupted.

Each of four potential ArW3 configurations, identified in Section 2.3, is evsluated.

3pecific objectives with respec. to the evaluation of each Ard3 confi7uration ares to determine the APf3 unavailability for various categories of plant initiating events, to identify the most significant contributors to APdS unavailability, when challenged.

1 l

I i

l

l Caport No. 02-1040-1095 Revision 1 Page 7 The objectives of the comparative evaluations of the four ArW3 conf 17urations ares

- to establish an overall figure-of-merit with which to judge the relative reliabilities of fered by the four ArW3 cc if17sra tions,

- to determine cost-effective system nodifications to upgrade the ArW3 reliability and performance.

Formal f ault tree techniques, as discussed in Section 4.0 are utilized in achieving these analysis objectives.

2.3 Score of Work This reliability analyeia examines four potential ATW3 configurations. These a re

" Pre-St!" Cont if ar ttion The ATW3 and other plant equipeent as confi7 ured prior to implementation of TMI-2 related plant modifications.

" Post-Ot:" Canfiftretton The AW S and other plant equipment as configured subsequent to implementation of certain 011-2 related, and other, plant modifications. It includes those modifications planned to be implemented through the 1902 34vis-Desse refueling outage. It also includes a written procedure to fulfill the ArW5 safety function using the main feedwater startup pump, the reactor coolant system makeup pump and the power operated reitet valve (ICRV) as a backup to the AW3.

- 51r4 Train

canfitirteten

he same AfW3 confiraration as for then Post-011 case, except that a third manually initiated notor 11 riven nain feedwater startup pump would be aligned to supply auxiliary feeitwater flow, without tne necessity for reactor coolant system makeup flow 4nd steam venting via the POPV.

"Analysta asset

canfiraretton The same AMia configuration as for the Post =U1! Jase, except that cor ain reconsond+1 system molitications, resulting f rom this at ady, are taeumed to Le implene"to1.

I Repo rt No. 02-1040-1095 Revision 1 Page 8 The first two configurations are analyzed to demonstrate relative improvement to AIWS reliability resulting from systen modifications alreacy planned or Laplemented by TECo. The third configuration represents a system designed to address explicitly the NRC concerns with respect to ATW3 reliability as outlined in references (1) and (2). The fourth configuration represents a system designed to address the most significant contributors to the Post-TMI system unavailability, as determined f rom a comprehensive evaluation of system reliability.

Report No. 02-1040-1095 Revision 1 Page 9 3.0 SYSTD1 DESCRIPTION 3.1 AMS Safety Function The AIVS is designed to provide coolant to the secondary side of the steam generators wherever the main feedwater flow has been interrupted or to establish natural circulation whenever the reactor coolant system forced circulation has been interrupted. This is necessary to maintain adequate core cooling and prevent fuel damage. In the Post-OtI configuration, there are two ways in which this safety I function can be met

1. The AP43 can deliver full capacity flow f rom at least one of the redundant Ards turbine-d;,1ven pumps to one steam generator. The water deliver / to the steam generator (s) must begin within ten (10) ainutes of the initial loss of main feedwater or loss of forced circulation. The water l1 deliver / must continue until the reactor coolant system cools down and is depressurized to the point where the decay heat removal system can be operated.

2. The main f eedwater startup pump can to manually started and aligned to deliver coolant to either steam generator. The present capacity of the startup pump is not sufficient for complete decay heat removal.

"'herefore, the manual initiation of feedwater via this path must be accompanied by the manual opening of the power operated relief valve (PCRV), initiation of primary coolant makeup flow (through at least one makeup pump) and isolation of the reactor coolant system letdown line. In this mode, partial reactor coolant heat removal is obtained by venting fluid from the primar/ system through the PORV. The makeup flow is necessar/ to prevent excessive reactor coolant inventory loss until the high pressure injection pumps can provide emergency core cooling. In this mode, the safety function is accomplished if all actions are initiated withia thirty (30) ainute.: of the initial loss of feedwater. The systems must function until the operating conditions for the locay heat removal system are reached.

ther7ency procedures for this second approach exist in the Post-Ot! configuration only for the situation in which of f site electrical power is available at the plant site. Emergency procedures for the second approach are presently planned for the added situation in which offsite electrical power is not 1 available at the plant site. This extension to the emergency procedures is credited in the analysis-based configuration.

The Joabination of the startup pwep, mabeup purp and Porn is referrei to as the " feed and bleed" metnod througnout the belance of this report.

l.

Paport No. 02-1040-1095 Revision 1 Page 10 The first of the above methods is the anticipated technique for f ulfilling the AFWS safety function. The second method is designed only as an emergency backup in the unlikely event that the first method is unsuccessful.

32 Pre-TMI System Configuration The Pre-TMI AIWS is illustrated in Figure 3-1. The system consists of two independant trains, each containing one steam-driven auxiliary feedwater pump, AC powered motor operated valves, crossover piping which allows the pump to supply water to either steam generator, refundant water supplies.

The primary sources of auxiliary feedwater are the condensate storage tanks (CST) with a minimum water supply of 250,000 gallons. Should this supply fail, plant procedures call for the manual transfer of the AFW3 pump suction to the fire protection system. The service water system provides an automatic safety grade backup to the other two supplies. The service water system is connected to the ATWS through motor operated valves, which are initially aligned shut. They receive an open signal on a low pressure condition at the AFWS pump inlet, as measured by redundant pressure switches.

The auxiliary feedwater pumps are both driven by steam from the main steam generators. Normally, steam generator 1 provides steam to AFWS pump turbine 1, and steam generator 2 provides steam to AFWS pump turbine 2. However, in the event of low pressure in one steam generator, the unaffected steam generator can provide steam to both turbines through crossover paths, as illustrated in Figure 3-2. Normally, the motor operated steam admission valves are aligned closed. They receive an open signal from redundant Steam and Feedwater Rupture Control System (SFRC3) channels on low steam generator level, loss of four reactor coolant pumps or high main feedwater differential pressure. The STRC3 actuation logic for the valves is explained in Table 3-1. Individual valves would subsequently close on a low pressure signal at the turbine inlet, a low pressure signal at the APM pump suction, or a low pressure signal from one steam generator. The turbine contains a trip throttle valve which closes on a turbine overspeed s17nal. A turbine governor valve is used to control turbine speed. It is controlled automatically or manually f rom the control room through a DC-powered motor.

The exhausts frcm both turbines come together and are vented tnrough a cocoon silencer.

Report No. 02-1040-1095 Revision 1 Page 11 The AIW pumps are self-cooled and have minimum flow protection through a normally open recirculation line. In addition, there is a normally closed test line connected to the pump discharge. Steam generator level is controlled at low steam generator pressures through the closing of the motor operated pump discharge valves. These valves are AC powered and initially closed. Additional motor operated valves downstream of the pump discharge direct the auxiliary feedwater flow to the steam generators. These valves are AC powered, are initially closed, and receive open/close signals from SFRCS.

Normally, A2W pump 1 would supply the water to steam generator 1 and AIW pump 2 would supply water to steam generator 2. In the event of a steam generator isolation, crossover paths are available so that both pumps would supply water to the remaining active steam generator. The motor operated valves at the steam generator auxiliary feedwater inlet noz=les are normally open, and would only close on a steam generator low pressure isolation signal.

Prior to the TMI-2 event, no procedures existed for using the main feedwater startup pump, in conjunction with the " feed and bleed" procedure, as an alternative method for fulfilling the AFWS safety function. As a result, no credit has been taken for this backup success path in evaluating the Pre-TiI AFWS configuration.

3.3 Post-TMI System Configuration The Post-TMI configuration represents the originally planned 1

configuration of the AFWS at the end of the 1982 refueling outage. It incorporates a number of design improvaments over the Pre-TMI configuration. Flow diagrams for the Post MI AP.iS and main steam configuratiens are shown in Figures 3-3 and 3-4. Major differences between the Pre-TMI and Post-21I configurations are:

1. The Post-OtI configuration has diverse electric power sources for motor operated valves. Certain valves on train 1 (AF-360, AF3870 and the main stream turbine admission valve MS-106) are powered off OC power supplies. The remainder are AC powered.

2. The turbine exhausts are redundant and seismically qualified. The plugging of the exhaust pipe / silencer is no longer a common cause f ailure for both AFWS trains.

Report No. 02-1040-1095 Revision 1 Page 12

3. Administrative procedures have been implemented to lock in position all manual valves and local control stations and hand wheels for motor operated valves in the auxiliary feedwater supply paths, the recirculation line, l

the test line and main steam supply paths. This reduces the probability for human error in misaligning remotely cperated and remotely indicated manual valves.

4. "'he turbine admission valves now have automatic dual level control, with the option for manual control.

5. An emergency procedure has been implemented to manually start and align the main feedwater startup pump to provide feedwater to the steam generators in the event that both trains of the AFWS f ail. This procedure includes the feed and bleed procedure for relieving fluid through the PORV while maintaining makeup flow to the reactor coolant system. This procedure effectively provides a diverse and redundant third train of AIWS.

The feedwater startup train consists of a single AC powered pump, which is supplied frca three water sources, and which discharges to either steam generator. The water sources are, first, two deaerator storage tanks and, secondly, the CST.

The fire protection system is available as a backup water supply should these two sources fail.

2b initiate the startup train the operator performs the following operations:

block the SF3CS signal and open either, but not both, of the main feedwater stop valves FW-601 or FW-612 (operation performed from the control room), 1 block the SFRCS signal and open either, but not both, of the main feedwater startup control valves SP-7A or 3P-7B (operation performed frem the control room),

manually open the startup pump discharge valve TW-106 (operation performed locally),

manually start the startup pump (operation performed from the control room).

I l

l l

l' Report No. 02-1040-1095

! Revision 1 i

.Page 13 A flow diagram for the startup ptusp 18 shown in Figure 3-5.

In addition to the startup pump the operator must initiata the feed and bleed operation. This consists of manually opening a PORY and its block valve, and operating the reactor coolant system makeup pumps. The PORY and block valves are controlled from the control room. Normally, the PORV is aligned closed and the block valve aligned open.. The makeup system is illustrated in Figure 3-6. . The system consists of two trains of pups discharging through a common pipe to the reactor coolant system. The makeup water tank provides a water supply of 4,480 gallons, after which the water supply is automatically switched to the borated water storage tank. .A motor operated valve provider the switchover function on a low level signal from the makeup tank. In normal operation, one train of makeup is assmed to be operating at =11 times. In the feed and bleed mode of operation, however, the normally open reactor coolant system letdown line must be isolated to prevent additional loss of primary coolant inventory. The isolation of the letdown line involves manually closing a motor operated valve from the control room, a routine procedure with all reactor trip conditions. . The feed and bleed procedure in the Post-TMI configuration applies only to situations in which offsite electrical power is available at l1 the plant. .It is not credited for initiating events in which offsite power is assmed to be lost.

3.4- Third Train System configuration The third train configuration examined in this study consists of an N* T adaat, manually initiated train of auxiliary feedwater in parallel with the two present AFWS trains.

Manual initiation is required so as to prevent excessive feedwater flow in the anticipated event that the two safety-grade steam-driven auxiliary feedwater trains function as designed. The third train would be started only if both of the steam driven trains failed.

The third train flow diagram is shown in Figure 3-7. *he train consists of a single AC powered motor driven pump, supplied from three water sources, discharging into either of the steam generator auxiliarf feedwater, inlet nossles. The puny is considered to be the main feedwater startup pump, upgraded in flow capacity such that the feed and bleed operation is unnecessary. The time requirement-for initiating.

. auxiliary feedwater via the third train is 10 minutes from the initiating event. The water supplies would be the same as for

the present startup pump.: - However, the discharge piping would

.be' rerouted to bypass the'feedwater heaters and discharge directly'into the AFits steam generator inlet nozzles. Either-of two AC-powered .1-

Report No. 02-1040-1095 Revision 1 Page 14 motor operated valves, normally isolating the train from the steam generators, would be manually opened from the control rocza when the pump is started.

In all other respects the third train configuration is identical to the Post-TMI configuration.

3.5 Analysis-Based configuration This configuration is based on results of the reliability analysis of the Post-TMI configuration. The Post-DtI configuration is found to be most susceptible to failures of motor operated valves (MOV) to open/close on demand and to the inability to implement the feed .md bleed procedure following loss of offsite power events. The analysis-based configuration represents the presently planned AWS configuration at the end of the 1982 refueling outage. It incorporates several design modifications as well as improvements to the feed and bleed procedure. These g additional system nodifications include the following:

1. The speed switch control for the pump discharge Movs AF-360 and AF-388 is eliminated and the valves are normally aligned and locked open.

2. The MOVs AF-3870 and AF-3872 are normally aligned and locked open.

3. All four turbine steam admission valves, including the valves in the crossover paths, open on an SFRCS signal.

In this case, both turbines are supplied with steam from both steam generators through parallel paths. In the event of a staam generator isolation due to low steam

. generator pressure the isolated steam generator discharge valves close and the steam supply system to the turbine is identical to the Post-TMI configuration.

4. Flow indication is temporarily installed in both AW pump mi. h recirculation lines during surveillance testing.

This permits flow testing of pumps to be performed without opening valves AF-21 and AF-23 (for pump No. 1) or AF-22 and AF-23 (for pump No. 2). Thus, an auxiliary feedwater pump remains available if the AWS is challenged during a surveillance test.

s

5. . Se startup pump discharge valve W-106 is locked open i and a check valve placed between the pump and W-106.

5. The startup pump bypass valve W-102 is locked closed.

I

Report No. 02-1040-1093 Revision 1 Page 15

7. Redundant steam generator pressure and level control rocm indicators are provided for each steam generator. The startup pump feedwater flow indication is upgre.ded.

8 The makeup system valve MU-33 and startup feedwater control valves SP-7A and SP-73 are controlled from the station nitrogen system or local nitrogen bottles and are therefore available following a loss of offsite power.

9. The original feed and bleed procedure is modified to better reflect the steps necessary to implement the feed I and bleed operation. Improved descriptions of various parameter responses enhances the ability to recover from incorrect operator actions. "'he revised procedure format will be similar to the Abnormal Transient Operating Guide 11aes wergency procedures.

10. 2e feed and bleed procedure is extended to the situation in which offsite electrical power is unavailable at the plant site.

3.6 ANS Supoort Systems For the purpose of this study the makeup system, POKV and main feedwater startup pump train are considered part of the ANS, since they directly support the ANS safety function. Other plant systems indirectly support the ANS as well. These include:

electric power system STECS

- service water system fire protection system station nitrogen system. g Cf these support systems, the reliability analysis results are only impacted significantly by the electric power system. The electric power system is, therefore, considered explicitly in the reliability analysis. The impacts of other systems are conservatively estimated (as discussed in section 4.3.1) and found to be generally insignificant.

The importance of the electric power system is based upon this system providing the electric power for valves, motors and pumps throughout the ANS. These power supplies can be categorized as:

powered from essential AC-buses powered from non-essential AC-buses powered from DC panels l

P Report No. 02-1040-1093 Revision 1 Page 16

! *:he DC panels are normally powered by battery chargers powered from essential AC buses. In the event of a bus failure, however, the DC panels are backed up by battery power supplies. As a result, the DC panels have relatively high reliability. The essential AC buses are powered by the turbine generator (through the auxiliary startup transformer),

i offsite power sources or a diesel generator.

For events challenging the AfwS it is assumed that a turbine generator trip has occurred and that this power source is unavailable. For events in which offsite power is assumed to

be lost, the essential buses must, therefore, be powered from 1 a diesel generator. One diesel generator powers the "C1" bus I while the other powers the "Dl" bus.

i The startup pump is powered from bus D2 which can be powered from either diesel generator. In the Post-MI configuration, selected non-essential buses can be fed from the diesel j generators through operator action (control room operation) .

For other non-essential buses the power supply is limited to offsite. power following a turbine generator trip.

1 Table 3-2 lists the interfaces between the electrical power system and the AIWS and shows the ultimate power supplies to

. individual AFWS components.

Major differences between the Pre-TMI configurations are:

1. In the Pre-MI configuration a ground fault on any of the t

Essential Motor Control Centers would cause a loss of one of the two redundant electrical systems. Thic has been

modified in the Post-MI configuration with the i installation of ground-fault detectors to trip the individual breakers on all loads attached to an essential bus. (Note that this is not a "lMI-related plant modification, but was undertaken by TECo to upgrade the reliability of plant electrical systems.)

2. Davis-Besse has an automatic switching system that changes the plant's electric source from onsite power (main generator) to offsite power in the event of a turbine trip. Ther1 is a 30 second time-delay between the turbine trip and the generator trip. When the 4

generator trips there is automatic transfer of the plant's electrical source from the auxiliary transformer

, to the startup transformer. In the Pre-3I configuration 1

there was a possibility that the generator 345 K7 breakers could be manually opened before the 30 see time delay and thereby fault the entire switching system by not allowing it to switch to offsite power. Procedures 1

t

-.mm. , yen- 4 m ---,g--- - . , . - ---..w m --, e ng., g,,.,-v,enu e ..w-.7w,, -,-e.wme, , y

Report No. 02-1040-1095 Revision 1 Page 17 have been added to insure that there are no actions done until the 30 see time delay and automatic switching is completed. Also, if the 345 r/ breakers are opened for any reason other than a degraded offsite power source, a fast dead transfer to the offsite source will occur in the Post- MI configuration.

3. In the Pre-TMI configuration there was no way to know if there was a ground fault on one of the D.C. MCC Essential Buses. The Post-mI configuration includes a load fault detection-system to correct that situation. (This change is not being planned for completion in the 1982 refueling i outage, but will be completed later.)

I i

_ . . . . - . _ . - - . . _ - _ - - - - _ _ _ _ . - _. -- .- . - - - . m R: port No. 02-1040-1095 R vision 1

! Page 18 i

1 i

1

Ei PL b t t t t t 5 5 5 5 5 5

- a. - _ .a. _u .a _a .e. p u. ,

a I

2' :'

x.

a. -

l

, y t- t t g g

{'

2,;

e-E

= 4

.s. a.

-L .a. .s .a .r t' 5 5 %_a

5 5 a

_a f.:. .m .e .- 3. .a. .a . -r -E.e

.F = -i _5 _> .a-> - _

> _f _ .a._ >

.E : ,

p* t + .ym I l, . l* t ts D t b V E t aE e E. N Ne Es i

3 a = -a ! l e + e a,

c'

. i

.2 x -

_ _-. __..=--_u.u._, e .- a

--=

-____u -

r. r r t 9 '

4 #

! ; e s

t t t ta t.g !S l t -E N a P E

4

+

4 4;

?

3

_ -- _ _ a .:.

g I I p p

.e e I

g

, a e r r r e r-

., _ _ _ _ .a '

a _ . _ a _ A . .:

, f .s ,

w- '

i i

l'

.4 hs t t at be ba a

$ 22W a+ t t ta t t ra a

,'l I. _c +

2__gm _E _E .E. .E.

e i .

_ _A.E. _5 _ 2 _2 _ 2 _a_- 3: -

Na; 2 1 i

1 3 a i I -

! p

~

i t t t t t t -

? i Mu r t t t t t; '

i tt 5 u% ! * =

4

d. 2l

...+ - ,. - - - - _.=_.m. - -

5

= = e 5 5 5 5 5 t aa I e - j

_e _s _ _u _ -a- _a _ 3 _.at.

5'

~

7 ! M

_ -l .~ I. -

1: t >a t st at ts a ME t. @ i. -

4 I #

P.si e"l v 12 em e + m -

c .,

--a

_ a. _ _= - .m. _.3 e , . g-_ -_

u: - .- - 2

=. s

_a_

r x _=. -_;r r

_. 7* - -[

8 Ej j 3 - 6 g at - =t '

l ss I

l t >3 > b st ta t a 'z I tpu M = M - = 3 2;

P t:

5!

e

' a E

-i G gC= _l !'

set 3E* e E 5e : tE 9 E 9 -

= .:-

,, . _ , d. g- - - .3_ _2 _ _E

__Gy _ _

_ ___ -_____ p-

, ag e .e

- ,~

j e .

g g

El e t t t t t t~ tMW t t t t t 1

i 8!

iI*,, _ -PS.- _.5 5 _ _g _ _.5 =

-I 5

g

.. 5tD

.e 5. _

g 5 *.s. _.g 5_m.

.3 I*

> u. 7 a ,l

} ,l i

- 1 -

, e '

3* '.f *-

-u *

> > > > t #

j .

9 1

z a.

,. e e t t i . t'.

M3 , ! ! -!= ! 5 ;

4

) 5m *G E* f e8F 1i

.z. e

_= - -. __a_.a. - _.m__5__a5 -! . .

-. 0 5' _ _ - _

a al' ~

ed. = ,

.e. -.' na*

> ~ ,. I .

t e .s.

e

.a ' ** 8' k t t at ta t -

j :: i7

{M

. t t g :W* M*

g ..:: _ .g _---..a_.3.,.

a E *4 ft 2 21

] g a 3 s : j ~" ~.-

3 ,

s=5- _.m

.:. s: s: =l a':

i ..s. -

- .-. s .1 ; -

i

w , . --

i .

a -

[- 3 J' O ! -II

-t .I t ts t t ta ts t M* E

- : : : a .5 se=1

-. => '

5 s , . g-. . . * @ e

.s. -

. .=~

E- e

,e -.

eg :

.a

_=. _ . . .

m

.=.~._

m i

- 1

- _ ._ _ _ . s- =. *i _ _ .m I

r

. -._ r. r

=. .

d + e.:c e-J-%

e.

3 3g

- :- ! t t. t t t t ' 'I - t t t t t t -

!.' a -

3 8

5 5 5 5 5 5 ! i, s u

a i 5 3 5 5 5 .

e. ':,

_= .--_ _ _3 _ _m _ _n _ _= _ _a _ _u _ - _ _ . ._ _a _ _. . _ _n _. .a. _ _u _ _

u4. .. .. i , - ,. . .. ,..

i-t t e t e t

4 ! t e e e e- e

.-irI-g '

- _ _=_ 5_=__=_ 5 5 5

_=__.5_5_= _.r. = _, .

.t m, . i_ _a5 __5_e _5_=5_ _=I_ _a _ ! ,

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

i, l ,,,

. .i

=32.- e p. i pl aas i

-a-a = =

! , -=3: 3 a- - -n g,, ...

n. :l

8 c;

l i i z - wns E .. -a2a5. - - ,

1 3 = ~5 = , a u :, n "1",,! !

L , a. p h.

F z- - - z . 4 ~ 5 - A 4, t E - . 5, i a 3

a

-w-v a - -

z i

.= ' s .

i

..n *

=J- -an !' 5=? a-

.r a c.;.a i 533..

g -

a a n. .E z , . s - a a

p

= .:

p 5

s. 4 e -. .

q-9=4*4=.az. 2

- .. 2 a J a .> e- a r.a

.r i h

5 s 3 ;23 I 6

.a p e-2 se ~cl.a en-*

co a: l y te I8

s s.s.i w M. a~in a s. ae sr. : .

=~ v : 5, .n l a g ., y . n - . n. .FF

, ~~ -

ar

- til 5 e :* e., :! n e r. ,-

.ae

a

= 3:. .

u e

5.-f. .4r .*. e *- = *..t :

E.2,. .i.q.gp: z. ..- e

=..=. - s.]l' 4 s

a.s

.-.- . .. . !' .--a I. l*.e., .

.. .s.n . r -5

- .. bw .h. T.4: . f.*.

. =* . . . - .

'u ;

l t

i I

I I

l.

-7.- -i . t --

- ,.-e, -.-.,m -g q,-.w .- ,--fin -. +--.--.m v-- my npg - - - i3

. - . . - . ~ . ._ .- - - ._ .. __ - -- . .__- . . _ _ _ _ _

Mport No. 02-1040-1095 Mvision 1

! TABLE 3-2 Page 19 l ANS/ Electric Power System Interfaces i

f l Electric Power l Power Supply l Ultimate l ANS Component i Supply 1 Type l Source

! I i l AF-360 (Pre-MI) lMCC-ele l Essential AC l Diesel Generator #1 j AF-360 (Post-MI) lDC Panel DlP l Essential DC l Battery IP

{ AF-3870 (Pre-MI) lMCC-EllD l Essential AC l Diesel Generator el ,

AF-3C70 (Post-MI) lDC Panel DlP l Essential DC l Battery IP AF-3869 lMCC-EllE l Essential AC l Diesel Generator #1 AF-388 lMCC-F12A l Essential AC l Diesel Generator #2 AF-3872 lMCC-F125 l Essential AC l Diesel Generator #2 4

AF-3871 lMCC-F12A l Essential AC l Diesel Generator #2 SW-1382 lMCC-E12A l Essential AC l Diesel Generator #1 SW-1383 lMCC-FilC l Essential AC l Diesel Generator $2 MS-106 (Pre- 2I) lMCC-EllC l Essential AC l Diesel Generator #1 MS-106 (Post-MI) lDC Panel DIN l Essential DC I Batterf 1N MS-106A lMCC-E2B l Essential AC l Diesel Generator #1

. MS-107 lMCC-Fila l Essential AC l Diesel Generator #2 I MS-107A lMCC-FilB l Essential AC l Diesel Generator #2 j ICS-38A lDC Panel D2P l Essential DC l Battery 2P

ICS-388 lDC Panel DlP l Essential DC lBatterf IP

AV-1 lMCC-F13 lNon-Essential ACl Diesel Generator #1 & 2

$ AV-3 lMCC-F13 lNon-Essential AClDiesel Generator el & 2 Startup Pump l Bus D2 lNon-Essential ACl Diesel Generator #1 & 2

Makeup Pump #1 l Bus Cl l Essential AC l Diesel Generator #1 Makeup Pump #2 l Bus D1 l Essential AC l Diesel Generator 92 N-601 lMCC-FilD l Essential AC l Diesel Generator 42 l N-612 lMCC-E11C l Essential AC l Diesel Generator #1

) MU 3971 lMCC-EllD . l Essential AC l Diesel Generator #1 N-786 lMCC-EllD l Essential AC l Diesel Generator el j N-790 lMCC-F12A l Essential AC l Diesel Generator #2 i N-460 lMCC-F32A lNon-Essential AClOffsite Power i

MU-2B lMCC-E115 l Essential AC l Diesel Generator #1

! PORY lDC Panel DBP l Essential DC l Battery 2P g Block Valve lMCC-E6 l Essential AC l Diesel Generator #1 1

t FIGURG s-1

l $6-WI AFWO OWFIGURAfloN W

@**h a

CD167 C0:65 CD8G4 Flow g FJtt PtoT.bl61tM

'1 ~>k n q's l

'"8 5 o ,

Iw,cwice r, ' AWR 4

6$1tiu o g,, J 81 t

S w An' sG-1 (Q

O.

"l1}

-rn)h-$s,u, h

nz W & (E24 -l) cn i an "

ssC^W sess araos

.a b.

W AF-81 $ >$

C YI' [$1 AF14 O LIMIT OF yo erts ~

ces Ar m AF 5 Stil5MlC 'O 05'I (D

, PIfldG h >$ y AF21 4 '

,, n

,, 'E * @

!! @ erzog u ; ,, uS $rb,s 'l uY f gf $2N n,a o

g&g q g I*

?01-101 4 sT tog r%e ug5a ggg,yagg5 j

Sn 6 Sd4 sy, W

$ b SG 1 SG-2 g L.O MS 87G L.0 MS-675 I

IC5 IIA Y icsfl6 3 El k k Q u A A 2 - -

5/L-o

'75 Liuff of

$ ! Sel5MIC _ , ,_

2 PifitlG a

JL 17 MS 10G Jk b J k MS107 3

MS727 MS 72G[ M5U 45735 i h 3 Ps.) f

' k J ISAA

)JN M5130 A /

s PS 2 l

}i La Es /

R j MS 130 {,] gQ F MS729

! d MMgAl su gMUAL 6TCP

,!; VALVE W V1 VALVf MSV4 3

IC5-3SB 4 sca-S8A i m 4

i b j ~! AFPT-l fgggg,g.2 AFPT-2 N, L - - $6'T)Al }AAld Of6hl 6Y3f6M (JNf160fAfidtj db 7;e.u2

'k & ?BI ILE>)CER R_ F N wTo atmos.

$ w

. 4 toGATlFIES 1i8066 cstarco)Es4TS TilAT WEr16 p f409ffl60 FROM *flE Pg6 -TUT ConfleulATIO*S

< h q Q csn To sucWASe oveRAGG 61'6TEsA RELnAerusTV-

. si Colef aii 4 A a".

4(045 (854+

f1AM{ row b

g a Fstt Prot.efettM R

dj ->d Fns C @ csNuc6 G

~

>k I5 theta E U t IsIt 2 2 D

Q .s n 6T257+ me rontg) (or tbreR)

@ rinza + M (D colei(tstl[W E gp Ani @ cG-l

8 I 31 # #88 7s . ., m , .. as s

A l1 k &

Q

~

C (ssI1 syjg O SU_ ntst igc fC-,4 T5 ) NkTl 3m PIPn1G t

-) m

(""***&

O '

$ @ 4 ,7 g s.o QM -

l E M A N- Af M A

e e o

- ' L D. gggq(g

, fft$5M 'f 108 4 tt-20lo copfM LGA blusttERO Ins $ N-!ST fun Mtal 61RAstifK i FIGURE S-3 j F , t'061-%4i AFW6 CONFIGURATION Y

f n -k IOGMlFIG6 1M06G COMt'or\tisnG TttAT*

WERif MopsfldD FROLI Tile life-THI (j CotillGuRyiotl To leicfleAda oveRALL 6/6TEFA ku LIAblLiff. '

f. SG 1 SG-2 E ' I LO MS 87G LO (S.675+

ICS IIA ICSfl6 8

01

. u j

k

'n

~

L;gif of

$ 661dlA16

{Q Vit'lIIO (1

in (h $ MS 10G ---@ o.c. MS-107 a

MS*127 O MS 12Gf ag.73 45.]35

, t L J k J B PS2 S3G4 IAS130A g 4 tAS-130 L 0- 4{,] d5'T21 4 {d5 !

b MAdtAL % aktuAL GTOP

,1; VALVE 8dSVl VALVf IISV 2 lCS 38 B Li l.p scs-S8A

'; FIGURE 6-4

AFPT-2

'IDST-Ril mal}4 6TgAbi c/3f9A a!n cc wnsuRATIcx1 - - b I

g

$ 6fLEllCliK

-, g SILBEER M

l e sh rt

%_ - i@

s X- ) \ .

- b F 2 t ,t i

G am gawr )M e

'a S rt o )fi M )y' q Tte s -

nw '

e1 m,4vrQ -

=4 -

(3

<.cV >6 4

a

{w r

o 4 sf04 s

u y &, g 4

r T a VA o s

s w8 r l F4 s w-"' e I 4 v

s

, 3 4 hs 3

m a

'h "f aHs _.

g

,1 nr hu

" a oo .m S

-*~ wt H4> t e

uF be v t

s w

qt rP r aU d

W es r

r r X e

> m 1

r s'

r$

eR,5 *8 g ~3! r e

o s

) %w MA

>., i fg T ;

s y G s W

A v 6 A

M g

. r , M n

r g T 6

- }l G -

- Ts' Sud 4s Me8I LSP I

sP g.Y

- - <E $ ; , ;; , t

- 1 .al$

! )

!q~

! qga?cE@k .E s

~ ' y2 E. e '

s .

=

(f ink O o ! b t.o. 860110 A

> \' g L_. WAftfR 00

k. --b Q On 1'1 OhD o DH94 b Y f* L htl f'i O f g]

f

& G6t9Mi&

a PIf!# $V" 5 3 S 4I O

.h Y- a

?a A N

3 C

~WW W10 kW31 MdL T WliE' O O ~ *2 1G u Q 7s @ 43 LT 5

ji b . L o ,,

d g eb"N N#'S' Hf'51 6 ht43 Md -lG1 gTog o F.c. uo 26 in t.\

LETooWil FIGUR6 3 6 .

Laus MAK6 UP #6fBM co4FiGURAfl0N g gogg3t,ty ctoggo N No htAL,1,f Of W

5

N 4 FIGURE 3-7 o

i O THiTD 1TAIN CONFIGURAflON h \; otALora L 6TCRAGE

" 0 9

y ll 1

3/3-LnAIT6 of lh l1 66l6MlG a

2 PIPlH6 l S @

FWI FWJ4 66-1

{ '} _ k a II74* b L.O. yW.g $g.[ '

j

@ w2 h

J &.

IC f' V 6 TAU C #d" AfWlRA84 A Fu-32 S O @ @rhD'0-w m custom AFA ffAIN 6 7 6TolfME TM

~

5 N

l 2 (FROM CST l g

] A.o. b3 O

O o

n @

t. ,

~ ~

M4 V7 (g 4.f)

F#2 FW 42$

44S Af& 1 1

({

4 (AO. VALVE)

,r#lI W

6 -

4- lOGATa i'IGS 'litost!QaAIMnEUTS plat W68G

). M00lYl&O FRotA ~{18ti F1th '(L41 Conflun'ATI)*S R

x LA h g3g g to. t o. 131 2 To lucREAst! OreRALL 6/6T6:4 ReLt4tsitsTV-N NN E g(1 C0147 4C0:45 CD 8G4 + Ro#

s a a Fstt PtoT 6161fM b

y -Id Ffts C @ ttRhCS y

C -

hI I I he>A

$$dS SC ,a 2 p2 Q sT 257+ me rowtg) (or fbegd m

colai Gl? [n!

r1 W g M41 h h ~ rinz )e h cG-1

[

.ib hhikum, en as i. a* N Io LO-t j

4 A -l1 e 5c C

O gg_

AFISV

,,@ [$1 Ap.y l 11 Aff4 3 ] II i-l ne,,w

@ 4,,,

s.a L ><

y -

o -

N 1

u a l; -D,@y* N uso mu La D4 me-E%)4- ms a + (*8*2 fil 24-D E

333 ,

"Q tf O

's @

LD .%sNL^2

'N '

' ' l WAX:R

~

eQM'4 W n $$?'b sinenA ST tal 4 st-to:, enc 66 USH eranega 61'!)1 flin M26d &fCA>lNK i FIGURE .S-8 h

W ANALYSIS DASED AFWS CONFIGURhTiokl

O

'; I

[ceAERAToR 4169 AGE TANK *)l l

h >4 yWE C

~$ r

~

N } W"$yg? h f 6.6 41 )

V t i l

?

FW 1

~

Q )

9 g ,

I $ 15.

B_ _ rw at a g pg j ig S/I-30 B rw450 twerue 'NYs) ggol-g g

LiHITs of seismic j2 d Xfd32 r r_)

-3s/1 4g1 m PIPiWG R d m

() mm I O o - O_, ,w . , 07,

,- o GiBFwise

@ o ri.A k.se 51I V - !% -

n Grurue FW11 -

Twh

- ruur 3 446 -r g* .- -

--Q B

bS- N4 0 [

ingNe j%es F fnTo2 TAh8)'

di 2

%FJ)X-@

~~

6

... rwr L.

("'

! (veseasron F ys u f e S - 9 h

(*" *### ANALY6IS BASED 6 TART-Uf' FUMP I

( j h 6OhlFIGURATION W

i

Report No. 02-1040-1099 Revision 1 i Page 29

4.0 Methodology The reliability analysis is based on the development of f ault trees for the ANS and its support systems. Probabilities for basic events appearing in the fault trees are assigned based on reviews of industry reliability data sources and Davis-Besse plant-specific experience. A Boolean manipulation computer code is then used to determine the ANS unavailability. The unava11 ability of each ANS configuration l 1s determined in this manner.

7 The ANS unavailability is dependent on the plant initiating 1

event which causes the ANS to be challenged. The relative

' differences in ANS unavailability for the four configurations are also dependent on the specific initiating event. In order to develop an overall figure-of-merit for each ANS configuration, the annual frequencies for various initiating events are estimated, again based upon industry and 4

plant-specific operating experience. The frequency of the

, initiating event is multiplied by the ANS . unavailability for that initiating event. The results are summed for all initiating events. The result is the overall annual frequency with which the ANS is unavailable when challenged. This is the overall figure-of-serit for each ANS configuration.

The analysis can, therefore, be described in the following phases:

I -

system fault tree development data analysis system unavailability analysis

-initiating event analysis ceabined system / event analysis The interrelationship among these analysis phases is illustrated in Figure 4-1.

4.1 System Fault Tree Development 4.1.1 Fault Trees Developed 4

Fault trees are constructed for four AN S configurations: l1 Pre-SI Configuration Post- SI Configuration

- Third-Train Configuration

- Analysis-Based Configuration l1

. - . , .. n_ , . . . - . ,.

Report No. 02-1040-1095 Revision 1 4

Page 30 1

The fault trees contain as the " top event", the failure of the AIWS to perform its safety function. The specific safety functions are described in Section 3 1. In addition to the fault trees for sach AIWS, subtrees are developed for other plant systems which support the AIWS safety function. These

, subtrees are limited to only those parts of the support systems which directly affect the AIWS safety function.

Support system subtrees which are developed are:

electric power system main steam system main feedwater system and startup pump makeup and purification system i

power operated relief valve The attachments include fault trees and subtrees arranged as follows:

A2WS (Pre-MI) Fault Tree Main Steam System (Pre-MI) Fault Tree Electrical System (Pre-MI) Fault Tree AIWS (Post- MI) Fault Tree Main Steam System (Post-MI) Fault Tree Electrical System (Post-MI) Fault Tree Start-up Pump (Post-MI) Fault Tree Start-up Pump with ' Feed and Bleed ,

AIWS ( Analysis-Based) Fault Tree Main Steam System (Analysis-Based) Fault Tree 1 Startup Pump with Feed and Bleed (Analysis-Based) 4 Subtrees for other systems supporting the AIWS safety function are not developed. Such systems include:

fire protection system service water system SFRCS station nitrogen system l1 The reliability of these systems is found to have a lesser impact on the achievement of the AEWS safety function.

Conservative estimates of the unavailability of these systems are used in the quantitative analysis of the A1WS fault trees

+

as discussed in Section 4.3.1.

4 i

i t

Report No. 02-1040-1099 Revision 1 Page 31 i

4.1.2 Fault Tree Methodology For each system fault tree or asubtree, the safety function of

' the system is first defined, and failure criteria applied. An absolute determination for failure criteria is made for systems where a reduction in capacity leads to failure or unavailability of that system. A list of systems, safety functions and failure criteria assumed in this reliability study is presented in Table 4-1. Note that a 10 minute AIWS actuation criteria has been arbitrarily assumed, for i conservatism, in this analysis.

Detailed fault trees, or subtrees, are then constructed using the methodology and symbology of WASH-1400(5) and IEEE-352(6). The construction of the detailed fault trees, or subtrees, is done in a rigorous, systematic manner, j considering every component and event which could contribute to the failure of the system. Quantitative judgements about the likelihood of failure of a component are not made during J the detailed fault tree const=uction phase. The following mechanisms for failure are included in the fault treest i Pre-Existing Faults

- outages for test and maintenance

- demand faults for initially inactive components

- failure mechanisms which are dependent on the duration of

, the standby period, and which will cause failure on demand for initially inactive components-pre-existing human errors, e.g.: maintenance faults Faults occurring During Mission

- failure of an active component to change its state, e.g. a demand fault

- failure of a component to continue operating 4 - human errors of commission human errors of omission !

only pre-existing faults which would not be detected in normal l P lant operation are included, e.g. ; pre-existing faults in 4 active systems (other than test and; maintenance outages) are not considered.

i

+

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

, Report No. 02-1040-1095 ,

Revision 1 i Page 32 l 1

)

In developing detailed system fault trees, single passive

failures and double active failures are considered within a single process flow path (e.g.: within a single train of a ;

multiple train system). A single failure is the failure of

, one element within a process flow path which causes the failure of the required flow path function. A double failure is the combined failure (either random failure or dependent j failure) of two elements within a process flow path which

! causes the failure of the flow path function. A passive ,

failure is breach of a fluid pressure boundary or blockage of a process flow path in a fluid systems or short circuit, loss of electrical charge or loss of ability to conduct electricity l due to physical defects in electrical systems. An active i failure is a malfunction, excluding passive failures, of a component which relies on movement to complete its intended function upon demand. Examples of active failures include the failure of a powered valve to move to its correct positions or

< the failure of a pump, fan, or diesel generator to starts or 1 failure of a circuit breaker, relay, or solenoid to change

position"when energized. Ihtaan errors (acts of commission or caission) are considered active failures.

In constructing fault trees the following rules are applies ;

The fault tree is developed to the level where acceptable !

failure data exist. ,

Components and basic events are coded using an eight character m nelature as shown below: ,

1

A PM 001A F i 1 l l l System Code .l l l l Failure Mode !

4 (See Table - l l l l Code (See i 4-2) { l l--Table 4-4) t i I i Component Code l l Descriptive (See Table 4-3)----l l---Nomenclature i t

) -

Ports of the fault tree which are only applicable to j specific plant conditions (e.g. t loss of offsite power

events) are combined through a gate with the " house" logic ,

l symbol. ;

i The symbology shown in Figure 4-2 is used.

4 4 In addition, the following assumptions are made in developing fault trees:

t 4

r I

Rop;rt No. 02-1040-1093 Revision 1 Page 33 The plant is assumed to be in a normal operating condition at 100% power at the time of the initiating event.

- Pre-existing faults in active plant systems (e.g.t one train of the makeup and purification system) are not considered, since such faults, if present, would have been readily detected and corrected.

- Cperator action to recover from a faulted condition is only credited when the operator has sufficient instrumentation to detect the fault and a written procedure directing his recovery action. The probability for failure to take the recovery action is discussed in Section 4.2.

Component alignments, as shown on plant P& ids and electrical drawings, are assumed for the initial plant ccafiguration. However, the possibility of misalignment is considered when such misalignment would contribute to system failure, and when such misalignment might not be detected in normal plant operation.

Spurious human acts of cocsission, such as taking an incorrect action when there is no indication that action is required, or acts of sabotage are not considered.

j 4.2 Data Analysis l

Data on the probabilities for fe.alures or unav..ilabilities of basic events are necessary far quantification t.f the fault trees. Such data consist of system and compen.nt failure data and human error data. Indust:7 data sources ar.d Davis-Besse plant-specific operating experience have been reviewed to develop a recommended data base for this analysis.

4.2.1 Industry Data Review Sources for failure rate data are listed in Section 7.0.

These sources contain sununaries of recent nuclear power industry experience for electrical and mechanical components generic to the industry. In some cases, the data is supplemented by experience with similar components in other industry applications. A major data source for nuclear power industry com WASH-1400(5)ponent .

reliability This study is the contains Reactor Safety reliability Study, data on most components found in nuclear power plant safety systems. It is based on compilations of many reliability data sources avaialble at the time of the study, in 1975. Several more recent NRC-sponsored data summaries (7,8,9) document the reliability experience of common nuclear power plant components, specifically valves, pumps and diesel generators.

Report No. 02-1040-1093 Revision 1 Page 34 These summaries are based upon Licenaee Event Reports (LERs) through which safety system malfunctions are reported to the NRC. IEEE-500(10) represents a thorough compilation of electrical component reliability data. Reference (13) contains summaries of nuclear power plant equipment malfunctions as reported through the Nuclear Plant Reliability Data System (NPRDS).

Failure data from these sources have been reviewed and tabulated in Table 4-5. The table lists the recommended value for failures (expressed either as a failure rate - units of inverse time, or as a failure probability - dimensionless) .

Note that the recommended value is not necessarily the mean value of the data sources revieweds in fact, the recessended values _are generally the highest of the reported values. Also listed, wnen available, is the uncertainty factor representing a measure of the spread in the reported data. The uncertainty factor is defined as the square-root of the ratio of the maximum reported value divided by the minimum reported value.

The uncertainty factors are rounded to the nearest half decade. Where only one data source is given an uncertainty factor of 10 is assumed unless the data source reported the data spread. The data in the table are presented by component type, e.g.: motor operated valves, and by failure mechanism, e.g.: failure to open on demand.

Human reliability data consist of human errors of commission and omission. Errors of omission include omitting steps from written plant procedures during routine operations (e.g.:

maintenance), during emergency operations (emergency procedures) and during attempts to recover from a faulted condition. Errors of commission similarly include those committed during routine operation and those committed during the course of the accident in attempting to mitigate the accident. The primary sources for estimatin the pecbabilitiesofhumanerrorsareWASH-1400(g)andarecent NRC-sponsored study (11) of human reliability in nuclear power plant operations.

Table 4-5 also lists failure probabilities, and uncertainty factors, associated with various types of human errors. The values listed are from the above two sources.

4.2.2 Review of Davis-Besse Experience Where possible, the generic data sources have been supplemented by analysis of failures experienced at Davis-Besse, Unit No. 1. This analysis is based upon a review .

of LERs for Davis-Besse from the time it began commercial operation until February, 1981. Due to the rather limited

Report No. 02-1040-1093 Revision 1 Page 35 data base, this analysis concentrated on ccaponents and failure mechanisms which occur relatively more frequently in nuclear power plants and which could have a more significant impact on the A2WS reliability analysis results. The components included in this plant-specific analysis ares i valves auxiliary feedwater pumps

- diesel generators

- test and maintenance outages

- human factors analysis for the feed and bleed operation, valves Failure rates for Davis-Besse motor operated valves, air operated valves and check valves have been determined.

Failure mechanisms are ! allure to open on demand and leakage (far check valves only).

Failure rates are computed by dividing the total number of f ailures reported in the LIRs by the total number of valve demands (for f ailure to open/ciose on demand) or the total number of operating valve-hours (for leakage).

LER's are limited in terms of the plant systems in which failures are reported. In this analysis, the following six plant safety systems are considered:

auxiliary feedwater system main steam system containment spray system high pressure inject!on system low pressure injection system

- chemical volume control system Five of these systems were considered by the NRC in the development of their generic data base (Reference 7). The sixth system, main steam, was considered in this analysis since it is directly pertinent to this program.

Table 4-4 susmaarizes results of the analysis. The total population of valves in the system is listed along with the total valve demands, total operating hours and total valve failures. In computing the number of demands placed on valves, it is assumed that valves are only operated during testing and that Davis-Sesse the minimum Technical testing (schedule Specificatione contained

12) is used. The in the resultant failure rate is thereby conservatively estimated.

While this estimate is conservative, the procedure used is consistent with that used in the data analysis of all U.S.

operating reactors considered in Reference (7).

Report No. 02-1040-1095 Revision 1 Page 36 The failure of Davis-Besse motor operated valves to open/close on demand, computed in this manner, is a factor of three greater than the same failure probability computed for all operating U.S. reactors, as reported in Reference (7). The Davis-Besse plant-specific value is used in the quantitative analysis of the Ams fault tree. This failure mechanism turns out to be a major contributor to ANS unavailability, as discussed in Section 5.3. Most of the motor operated valve failures are attributable to torque switches and limit switches being out of adjust:nent.

Auxiliarf Feedwater Pumps This analysis includes failures of the auxiliary feedwater ptmps and/or turbine to start and to continue operation. The probability of f ailure to start is ccmputed by dividing the total number of reported f allures by the total number of attempts to start the pumps. It is assumed that each pump is only started for monthly testing, and that there is one demand of each pump per test.

Failures to continue operation are generally attributable to faults occuring during its standby period. The failure probability is calculated from the total reported failures, divided by the total standby hours for the pumps. Results are shown in Table 4-7.

During the period covered by the data reviews, there are a total of three demand failures of the AETS pump / turbine. All three of these failures occured during the first year of cosmarcial operation of the plant. Faulty speed control relays were the primar/ cause of the failures and the relays were replaced with relays capable of operating under design conditions. No subsequent failures of this type have occured since 1977. Since this type of failure appears to be associated with the plant " burn in" period, it is felt that generic industzy f ailure probabilities are more appropriate to be used for analyzing the plant in its present phase of operation. There has recently been a four*h demand failure of the AN S pump / turbine. It's cause is unrelated to the earlier reported failures and, while it is not included in this data review, it would not alter the conclusion that the Davis-Besse AMS pump / turbine demand faults are consistant with reported industry average failure data. The Davis-Besse experience in failure of turbine driven pumps to continue operation is in agreement with the generic data reported in Reference (8), so again, the generic data are used in the reliability analysis.

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

r I l Repsrt No. 02-1040-1095 l Revision 1 i

! Page 37

)

l Diesel Generator

Diesel generator failures reported in the Davis-Besse LERs can [

be categorised as failure to start, failure to stabilise and !

failure to continue operating. 7% failure to start includes l 1 actual failures of the diesel gen %rator to start on demand. ;

In computing a failure probability, only demands during ,

. monthly testing of the diesel generators are considered. This j results in a conservatively high estimate of the demand failure rate since other diesel generator demands (e.g.s

,' demands imposed by other system tests) have not been included i

I in the calculation, although any failure occurring during such l demands are included. Failure to stabilise includes faults ,

which prevent the diesel generator from operating for more I than a very short time after starting. These failures are generally due co pre-existing faults occucring during the ;

standby period. The failure rate for this mechanism is !

computed from the total standby hours for the diesel 7

generators. Failure to continue operating includes faults i i

! occurring as an actual result of running the diesel j generators. The failure rate for this mechanism is computed i i from the total operating hours logged for the diesel J

generators. Table 4-8 summarises the Davis-Besse diesel l generator reliability experience.

) In the analysis of the electric power system fault tree a l probability of failure to start and stabilise is computed. ;

This probability is the sum of the probability for failure *o

! start on demand and the probability for failure to stabiliza, l which is calculated by multiplying the failure to stabilise j failure rate by one-half the mean test interval.

1 4

In general, the Davis-Besse diesel generators appear to have experienced slightly higher failure rates than the reported !

industry averages contained in Reference (9). Many of the

{ !

diesel generator failures have occurred as a result of faults j j in the turbochargers. TEco plans to improve the diesel generator reliability by installing new high capacity

(

r j turbochargers and modifying the lube oil system for the [

} turbochargers. While these changes are planned for the 1982 I

! outage and should significantly improve the diesel generator !

i relishility, their quantitative impact on the reliability is not known. Therefore, the higher failure probabilities computed from past Davis-Besse esperience are used in this l

reliability analysis.

l !

I t

4 l

i

J s

Repsrt No. 02-1040-1095

-Revision 1 Page 38 3

i

~

Test and Maintenance Outages

! Test and maintenance outages for the AFWS and diesel i generators are computed in the same manner as reported in J WASH-1400(5). This calculation is dependent on the "

j plant-specific frequency of testing and the =ewintam time r allowed by the Technical Specifications (12), during which a

component can be out for maintenance while the plant is in .

l operation. '

! The unavailability for a component being in maintenance is i

hm !

) G>t = 720 ;

j where f,is the frequency per month at which maintenance ,

]

is performed and ta is the average outage time per maintenance act (expressed in hours).

l l The unavailability for testing is:

ft tt

) Qt

= 720 l :'

I where f t. is the testing frequency per month andgt is

! the average time per test (expressed in hours). For the AFWB and diesel generators, the Davis-Besse Technical-Specifications (12) limit maintenance outages to 72 hours8.333333e-4 days 0.02 hours 1.190476e-4 weeks 2.7396e-5 months j before the plant must be shutdown and specify monthly test j intervals. An hourly test duration is assumed. The i frequency of amintenance acts is taken as .22 acts / month,'

I and the mean duration of the maintenance is taken as 19

hours for the AFWS and 21 hours2.430556e-4 days 0.00583 hours 3.472222e-5 weeks 7.9905e-6 months for the diesel generators.

The maintenance values are developed in WASH-1400, Appendix III, section 5( 5), '

i i lluman Factors Analysis ,

l

! The htuman factors probabilities used in the fault tree l analysis for the feed and bleed operation are computed l

using WUREG/m-1278. In each instance, the operator action contained in the procedure is analysed and compared to specific events in NUREG/Q-1270. It was necessary in many

}

cases to make assumptions concerning the operator system, 1 i since perfectly analogous exemples do not exist. These l assumptions are documented in this calculation. A typical l list of assumptions is as follows:

1. Aamotely operated and locally operated valves are ,

)

treated under the generic category of " Manual valves" i

n . r i

s i-l

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

A Report No. 02-1040-1095 Revision 1 Page 39 No recovery from error is assumed unless a specific 2.

control room annunicator is available.

l1 i *

3. Operator errors are assumed to be consistent with i populational stereotypes.

For the purpose of fault tree analysis, one number, representing the probability of failure for that operator action, is indicated. It must be noted, however, that this single entry is a composite of unique errors that, when.

combined, form the operator error probability shown in the

fault tree diagram. For example, the fault tree entry

" Operator Fails to Open Valve" consists of the following j componeats:

1. Operator omits step in written procedure ,

or j 2. Operator selects wrong valve from grouped system or 1

3. Cperator operates valve incorrectly multiplied by

4. Operator stress factor (moderate) 1 4.2.3 Recommended Data Base i

The results of the plant-specific data evaluation and the review of generic data are presented in Table 4-5. Also t

listed are recommended values for use in this reliability study.

The recommended values are generally based on the following prioritization.

Whenever possible, plant-specific data are used.

The highest value of recent generic data sources, 1 (References 7, 8, 9, 13) is used.

j -

The human reliability data of Reference (11) are used, j since this represents an expansion of the earlier work reported in Reference (5).

Also shown in the table are uncertainty factors on the data.

These are determined by taking the maximum variance of the tabulated data sources. Some data points are discarded if they vary from the mean value by more than a factor of 100 (In 1 all cases, such values are smaller than the mean value so that, in no case, are reported high failure rates discarded).

I Such values are not considered in determining maximum

, variances. The uncertainty factor is then rounded to the l nearest half decade. If only one data source is available, an

! ' uncertainty factor of 10 is assigned.

I,

, , . , , - ---.~ w._ , - - , ,%.,- -

- , - . .--"r- ' a-"-

l Report No. 02-1040-1095 Revision 1 Page 40 4.3 System Unavailability Analysis The system unavailability analysis includes quantitative analysis of the fault trees, the uncertainty analysis for APWS unavailability, and the importance ranking of fault tree basic events in contributing to AFWS unavailability.

4.3.1 Cuantitative Analysis of Fault Trees Each AFWS configuration fault tree is analyzed for each category of initiating event. This analysis resn ts in the qualitative determination of minimal cut sets ft: . the fault tree and the quantitative determination for the point estimate for the probability of the top event. All support system fault trees, except that for electric power, are evaluated as part of the overall AIWS fault tree.

The electric power system fault tree is evaluated separately.

Probabilities for failure of the electric power interfaces with the A2WS fault tree (see Table 3-1) are computed separately and values inserted into the AIWS fault tree. In cases where the dominant failure mode for separate electric power supplies is actually a common failure, these interfaces are treated as the same basic event in the AFWS fault tree.

For example, with loss of offsite power, the dominant failure mode for failure of MCC-E12A and MCC-E125 is the failure of diesel generator il to start and continue running. This is treated as a single event wherever MCC-E12A and MCC-E125 interf ace with the A2WS fault tree.

The WJetCUT computer code (14) is used in the fault tree analysis. This is a Boolean manipulation computer code which deterraines the probability of occurrence of the top event (and any specified intermediate events) in the fault tree. It also identifies the minimal cut sets of the fault tree.

Since the A2WS fault trees developed in this study are very detailed, many thousands of minimal cut sets exist. In order to limit computer running time and to avoid exceeding the capacity of the code, the code has an input minimum probability cutoff. Any cut set whose probability is less than the cutoff value is discarded from the calculations. So as not to eliminate any potentially significant cut sets, this minimum probability cutoff is generally selected to be three orders of magnitude (1000 times) less than the probability of the top event of the tree. In a few cases, excessive computer time requirements dictate a minimum probability cutoff of not less than 500 times smaller that the top event probability.

l l

Report Wo. 02-1040-1095 Revision 1 Page 41 Mean values for basic event failure probabilities are input to the WANCUT code. These are assigned from the recommended data column of ':hble 4-5. Where failure rates are given in this

{

table (units of inverse time) the failure rate is multiplied by either one-half the mean test interval or the mission time, as appropriate. For all initiating events, a 24 hour2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months AFWS mission time is assumed. This is based on a conservative ;

estimate of the time required for auxiliary feedwater prior to setting the decay heat removal system in operation. Also, the APWS reliability analysis is based on the assumption of non-repairable component failures (except for human recovery actions not actually requiring repair of the fault). Twenty four hours after a plant initiating event, repair of components in plant safety systems would almost certainly be initiated.

For operation of the diesel generators, a ten hour mission time is assumed. This is based on the review of offsite electric power restoration experience reported in Reference (5). Once offsite power is restored, the diesel generators would no longer be required as an electric power source.

For certain categories of events, various components may have l an unavailability of unity. For events involving loss of offsite power, all components not powered off the diesel generators have an unavailability of unity. Also, for seismic events, all non-seismically qualified equipment is assumed to be unavailable. One exception to this general assumption is the Pre-TMI configuration in which the common turbine exhaust silencer is not seismically qualified. This is a potential

common failure mode for both APW8 trains. However, the failure mode is not the rupture of the silencer and exhaust pipe, but its becoming plugged to an extent that steam cannot be exhausted. A probability of .01 (uncertainty factor of l ten) is arbitrarily assigned to this event.

Fbr failures of support systems for which explicit fault trees are not developed, the following order-of-magnitude unavailabilities are assigned:

System Unavailability fluid systems (service water .01 systest, fire protection system) a specific single channel of SFRCS .001 station nitrogen supply to any 9.'6 x 10-5 ,

single valve (involves passive failures only) 1

._ _ _ . . . __ . _ . . . . .. . ,_ _ . - . - 4 _

a l Report No. 02-1040-1095 Revision 1 4

Page 42 These values are based on judgments . formed from various L

reliability studies on fluid systems and safety grade electrical control systems. In all cases, an uncertainty factor of ten is applied to these unavailabilities. It should I also'be noted that the fire protection system has an unavailability of unity for loss of offsite power events and seismic events.

4.3.2 Uncertainty Analysis The standard deviation of the fault tree top event unavailability, due to data uncertainties, is determined through a mn==nts calculation. First moments (mean values) and second acaents for the fault tree basic event probability distribution ~.nction are input to the WAMCUT code. The code computes the resultant top event first moment (point value) 1- and second moment. The standard deviation of the top event, is then computed from the relationship 02 =

M2 M1 2

where M 2 is the top event second moment and Mi is the top event first moment. -The assumed form for the probability

distribution functions of basic events is a log-uniform distribution. The second moment calculation, in conjunction -

with the large number of events contributing to the top event, tends to make the top event standard deviation insensitive to the assumed probability distribution function. The 11mits of the distribution are the mean value multiplied and divided by

' the uncertainty factors listed in Table 4-5. Since the reen===nA=d values are greater than the true mean values (which could be compiled from the various data sources), this procedure tends to bias the standard deviation towards higher j unavailabilities.. The true + c value reported here is, therefore, too large while the'- a value may actually be

', - somewhat larger. However, the intent here is not to develop .

absolute confidence limits, but rather to evaluate relative

changes in A2WS unavailability and to develop a qualitative

! measure of the uncertainty in the results. since the same bias.is used in all cases, the results can be compared in relative tezas.

4.3.3. Importance Ranking

l. The importance ranking is used to judge the relative significance of basic events.in contributing to the unavailability of the ANS. Events with a high importance measure are more significant in contributing to AFWS failure.

Such'a ranking is most useful in evaluating various means to improve AFWS availability.

\;

-s- , , - - . . , , , - , h~,,. v , ,,e, .-

,,,.y n,,. ..r..-w%,. ,.e-e-.--,-, .v. .,, , a~%v ,

1 ,

Report No. 02-1040-1095 ,

Revision 1 Page 43 i

There are several measures of importance used in reliability studies. The measure employed here is the Fussell - Vesely

! measure (4). This measure, applied to a single basic event, is defined as the total probability for the occurrence of a cut set cor.taining the event, divided by the probability for

the occurre.nce of the top event. Basically, this is a measure

-of the sensitivity of the top event ( AFWS unavailability) to small changes in the unavailability of a basic event. The Fussell-Vesely measure can be applied to generic categories of events, e.g., the failure of motor operated valves to open on demand, which is a feature utilized in this study.

1 The importance rankings in this study are computed through hand calculations, using the miM=al cut set identification and probabilities generated by WAMCUT.

4.4 Initiating Event Analysis Estimates for frequencies of initiating events requiring actuation of the A2WS are developed. Initiating events considered in the analysis are listed in Table 4-9. The events are grouped into three event categories which differ in

! their assumed availabilities for certain plant components, equipment and systems. These event categories ares l

! Category la Events in which main feedwater flow is lost, but offsite electrical power and non-seismically qualified equipment are availabla.

. Category 2: Events in which offsita electrical power is assumed to be lost, but non-seismically.

qualified equipment is available.

4 Category 3: Events in which offsite electrical power and all non-seismically qualified equipment are assumed to be unavailable.

i Table 4-9 shows the categorization of initiating events 4 challenging the A21tS.

4.4.1 Frequency Estimates l_

! Initiating event frequency' estimates are developed'from reviews of generic industry sources, and are supplemented from

,' reviews of Davis-Besse operating experience. The primary l generic data sources are References (5) and (15). Davis-Besse

~

experience is summarized in LERs and unit trip reports.

Results of these reviews are tabulated in Table 4-10.

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

Report No. 02-1040-1095 Revision 1 i

Page 44 Mas of Main Feedwater Reference (15) cites 3sw reactors as experiencing this type of transient slightly less frequently than other vendors' NRs.

There is evidence of a "burnin" period associated with this type of transient, with a 50% increase in the frequency during the first two years of plant operation. The generic PWR frequency listed in Table 4-10 is the frequency after this two year period of operation. The Davis-Besse experience is in agreement with the generic PWR frequency, so the generic value is used in the reliability analysis.

Steam Generator Cverfill Davis-Besse has experienced three steam generator overfill events, but none of these actuated the ANS. The generic value is therefore assumed.

Small Break in RCS Reported events in this category includo control rod leakage, primary system (primarily pump seal) leakage, pressurizer leakage and opening of the pressurizer safety or relief valve. Davis-Besse has experienced one initiating event of this type, which occurred at less than 104 power during the first month of operation. This event is not considered to be representative of post "burnin" operation, so the generic PWR frequency is assumed.

Mas of Forced RCS Circulation

> - This event includes the loss of all reactor coolant pump forced circulation as the initiating event. It does not include loss of offsite power as the initiating event, which would also result in loss of forced RCS circulation. The generic data indicates that this type of initiating event is relatively infrequent. Davis-5 esse has not experienced a complete loss of RCS circulation as an intiating event.

Davis-Sesse did, however, experience a partial loss of forced RCS circulation (two loop flow) which resulted in low steam generator level and ANS actuation. The " initiating event" in this instance is considered to be the partial loss of RCS circulation.

Since the frequency for partial loss of forced RCS flow is

( expected to be an order of magnitude greater than for total j loss of flow, and since a partial loss of flow may result eventually in AFWS actuation, the larger Davis-Besse based l

frequency is used in this analysis.

l

i i

j Report No. 02-1040-1099 j- Revision 1 4

t j Page 45 Ioss of Offsite Power

, There have been three loss of offsite power events at Davis-Besse, however, one occurred during the initial power assentation and a second occurred as a result of power i transfer logic which has since been modified. h Davis-Besse experience since this modification was implemented agrees well with the generic experience of all PWR's reported in Reference (15). h generic value is, therefore, used. ,

j tme i

h frequency estimate for tornadoes is taken from the 1 Davis-Besse FSAR. No incidents have been recorded. This l value is in general agreement with the similar frequency j eatNes of Enference (5), which is presented as a conservative upper bound for the entire Eastern U.S.

I It is assumed in this analysis that a tornado would cause a j loss of offsite power. Its impact on the ANS and other i support systems is assumed to be negligible, since the buildings housing support systems are designed to withstand the effects of a tornado.

j Earthquake 3

h only source for this event is the estimate contained in Reference-(5). The value cited is, the frequency of earthquakes in the Eastern U.S. which result in ground accelerations greater than 0 1g. "

i 4.4 2 Factors Influencing Event Frequencies j There have been many Davis-Besse design modifications, either implemented since the UtI-2 event or planned to be implemented l by the 1982 refueling outage, which may affect not only the I ANS reliability, but also the frequency with which' the ANS may be challenged. Modifications which reduce the frequency

, of initiating events challenging the AN S may actually I contribute more to the overall plant reliability and the

<timinishing risk then do modifications intended to upgrade the

+

AN S availability.

)'

Unfortunately, there are generally insufficient or no plant i operating data available with which to quanitfy the reduction.

i in challenges to the ANS resulting from these modifications.

! This is true not only for Davis-Besse experience, but also for overall industry experience, since many plants have implemented significant changes in design and operation as a result of the 3tI-2 event.

I I

I i

4 1-

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

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

i-

Report No. 02-1040-1095 Revision 1 Page 46 l The initiating event frequencies used in this analysis are, t therefore, generally based on " Pre-MI" plant operating experience. The frequencies are, however, applied uniformly in developing overall figures-of-serit for all of the AIWS configurations. ( As discussed in Section 4.5, the overall figure-of-merit is defined as the annual frequency with which AFWS is unavailable when challenged).

The resultant figures-of-serit for the " Post-MI", " Third Train" and " Analysis-Based" configurations are, therefore, conservatively hight the figure-of-merit for the " Pre "JtI" configuration contains less conservatism. Since the same

<- conservaitvely high frequencies are applied to each of these 4

configurations, however, the relative differences in the results are indicative of the relative benefits attainable from each configuration. It should be noted, that significant reductions in the annual frequency with which AFWS is unavailable when challenged, could be attained through modifications to plant design and operations which are not directly linked to the APWS.

4.5 Combined System / Event Analysis Results of the AIWS reliability analysis are combined with initiating event frequencies in order to derive an overall figure-of-serit for e&ch AFWS configuration. The figure-of-merit is derived as follows:

A matrix is constructed listing each APWS configuration and each initiating event, as shown in Figure 4-3.

The frequency for each initiating event is multiplied by

, the AFWS unavailability for that event.

The products are summed for each. AFWS configuration.

The resultant figure-of-serit is the annual frequency with i which the AFWS is unavailable when challenged.

i i

W I

1 l

l l

. -~ - . ~ . . . - - - - . . ..

. . . . - - = --_

_ _ . _ . _ - _ .__--.- - ~.~- .- - _ _ _ - _ _

L R port No. 02-1040-1095 Revision 1 l

TABLE 4-1 System Safety Functions System l Safety Function l Failure Criteria for Study I i i i i A N3 (Pre-15tI) l Primary system decay heat l1. less than full capacity flow from 4

l removal l at least one pump train, or

! l l2. flow to stsaa generator (s) delayed

, I l more than 10 minutes from initiating l l event, or

'l l3. all AMS flow is interrupted during

I l required mission time.

i i ANS (Post-13tI l Primary system decay heat l1. less than full capacity flow fron

and Analysis-Iremoval l at least one p g train, or Based) l l2. flow to steam generator (s) from ANS l l delayed more than 10 minutes from j l l initiating event, or l3. all ANS flow is interrupted during

~

l l l required mission time.

l lAND j l l1. full flow from startup pump delayed l l more than 30 minutes, or l

j l l2. full flow from one v.akeup pump to l l primary system delayed more than 30 I

l l minutes, or l l l3. letdown line not isolated at reactor j l l trip, or

! l l4. less than full discharge from one

! l l PORV within 30 minutes, or l 15. feed'and hieed procedure is inter-

.l l rupted prior to HPI initiation, or l,

l l6. startup pump flow is interrupted l l ' during required mission time.

4 I

i 1

1 l

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

R: port No. 02-1040-1095 R;vicion 1 9"

TABLE 4-1 (Cont.)

System Safety Punctions System l Safety Wnction I Failure Criteria for Study i I A WS (Third

rain)l Primary system decay heat l1. less than full capacity flow from Iremoval l one AWS pump train, and less than l l full capacity flow from startup l l Pump, or l 12. flow to steam generator (s) delayed l l more than 10 minutes from initiating l l event, or l l3. all AWS flow including startup pump l l flow is interrupted during required l l mission time.

l I I l Electric Power l Provide AC or CC power to l1. inability to supply rated load to Syctem lAW S components l AWS components.

I I I I Main Steam System l Provide steam to AWS pump 11. inability to provide sufficient l turbines l steam to maintain full AWS pump l l flow.

I I

R: port No. 02-1040-1095 Revision 1 Page 49 i

! 4 TABLE 4-2 Plant Systemt Designator A Auxiliary Feedwater System C Condensate Systemt i E Electrical System 1

F Fire Protection System

' M Main Steam Sysl.em i

j S Service Uater System I P Feedwater System i

I 1

1 b o 4

f i

1 l

l

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

R port No. 02-1040-1095 Revicien 1 TABLE 4-3 Page 50 Component Code Mechanical Components Diesel DL Valve, Check CV Filter or Strainer FL Vaive, Hydraulic Operated HV Flow Element FE Valve, Manual XV Gas Bottle GB Valve, Motor Operated MV Nozzle NZ Valve, Pnetsaatic Operated AV 1 Orifice OR Valve, Relief RV Pipe PP Valve, Safety SV Pump PM Valve, Solenoid Operated KV Tank TK Valve, Stop Check DV Tubing TG Vent VT Turbine TB 4

l 4

b t

t G

e

i' R:: port No. 02-1040-1093 Revision.1

, i-I TABLE 4-3 (Cont.) Page 51 !

1 1

Component Code i -

4

}

j Electrical Components

,!- Battery ' Br aslay RE i j Battery Charger BC Relay or Switch Contact CN l Bus SS Switch, Pressure PS !

Cable CA Switch, Temperature TS ;

Circuit Breaker CB Switch, Torque QS t Control Switch CS Mansformer, Power TR r

, DC Power Supply DC Transmitter, Flow TF ;

5 Flow Switch FS Transmitter, Level TL 1 l

)' -Fuse FU Transmitter, Pressure TP ;

Generator GE Transmitter, Temperature M I Ground Switch G8 Wire WR '

Invertet (solid state) IV i:

Level Switch ES [

l Limit Switch 13

) Motor MO I Motor Starter MS 1

u I

I i

j k i

i l

4 4

1 7

j f .

i 4

1 1

I 1

i i

l l

l

. , . . . - , , - . . - - - , , . , _ - - - . - - . . - , - . - - , - - - - . ,-m , , , . -v. - . _ , _ _ . _ .-- _ , _ , . - _ - - , - - - . . - - - - , -

1 i

R:: port No. 02-1040-1093 Revision 1 Page 52 Failure Mode code 6

closed C Does Not Close K Does 100t Open D 4

1 Does Not Start A 1

Exceeds Limit M Mhp L ,

i.

j Loss of Function F 1 4

Maintenance Fault Y l

CPen o Open Circuit 3 a .

4 Operational Fault X i i

Plugged p Rupture R i

Short Circuit g ;

Short to Ground 3 1

4 4

i 1

i i i .

R: port No. 02 1040-1095 Revision 1

. a ,

_________-__-__._.____l=_.-_.--__-_-_l---__-

-_-- ___-- - __ _=_ _ - - _ - _ - _ - - -

4I Page 53

< . . . . ~.

i - . ~ ~ . , g g l

2 n. . .1 1 =-_. ._ .

.a .

--- --__=-____=_ ____-_-_---__-____.___-__ l

.I. .

i

_-_ _-____-____I_____'

a l

. r. .

~ r~

- 2 l L1 :

~.

6

- _ _ _.I_ _ - - _.I_ _ _ - - _I_ _ - _ _ _.I_ _ _ - _ I_ _ - - _ _I_ - _ _ _ _ .

i .S 3 g I a. . . - ?.

c

1. .

4 4 4 l , ,

_ - . _I_---__I__-_-_-_---_I____-I a

. __-_I__--__

s 1

.* 4 n -

-s -

.c.

1=_--__-

___.----____---_ __.__--__--_--- 1 : .

a... .

2 -

1 .

2 j :

j j 1 a .

J s : . aa s , . -

. . . . . . . - ] s :. . :. . .

s s sa

.s a a a a a a a g r. ; .

is la 4

i: . .

l . .r . .t . .".

i

_ ra laa . 2 - _, 1 _; t. -

) aa, .a.

a1. .

. 1.-

?. :

g e4i .

. 3*

h

.s 3 y e

1. -

. 9 e

p a4 *'.=e.E e .

e 74 J. i

.T..

ag 4 e i we

.s* _4_s.

e _ _ _a .3 .

e. . _ .a. .s.

.s. _

.--. ... $ a _.

I I

e l

. I i

J

3 Rep;rt NJ. 03-1040 1095

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

Revision 1 Page 54

~. ~.. ~ ~. e. , .

. e. .

,I.

f. .

e f.

. . e n e. .

fm. _ _~ _ _.;m __ _a _ .a. _

l

. ______ _ e_ __ _ _ _ -._ _ _ _ _ _

_- .m_ _ -- - _ . _ -_

. a e e e n e

_e

.i. .f

. ~. 1

.e .

w . =_ o._4 e.. - e. ~ e =.

o S. ..

3. .

s o. .. , .e_. . e

... =_ .- ._ ._ .

_ -- _- m _ __ _ _ -_

I l l l l l I

. - e .

. e. e. ..

e, . e. . e. f

. + +

e

.e. , , .e. . .e .e. _

d, . e 41

_ e M 8. P .

. =b . . . .

. e ;}

3 P8

.i. _ e .e__.- e w - - '.

e m 4

e _ .e 9 >

g" . . .

5 -3 ':

- I*. 3 3*.

P . j I. ~. _.m a.

_9e.

4 a .

4 g m.

4 e.

. e . .e.

9 .e .

!3

. - 7 .

2 ,

22 2A a 3 . .

. .m m._.m_ . .- m e .m r.. a l

b _.w .m _ _.m .m - m_ e.

.= g e -,

_m e.

.* e

.S. d =

. .y e. *

. m .e.

e.

e. .y .

.e. .a._

e g .y

. .e.* 4 e.d

.e. A e

e

=

. _3 . _$ . t. .

_ _. _. _. _ _ _. _ _. _. _ _ _v __ . ______a___ _.___._______

_I _s. . ._ ._. ._

__ _.. ~_ -

I l__:_., _ _ _ _ _ _.I _ _ _ _ __I . _ _ _ _ _.I_ _ _ _ _. _I_ _ _ _ _ _i_ _ _ _. _ .a .

.. t. .

. _e - . ,. _e

. a . . . . . .

t - e_

_ _ _-t._ _ .t.___a__.__. ___

_ . _ ___________.______.______ 1 ..

2 1 1 1 t

1 4..1

2 -

2 i n.

,I j a I j l J j

I 2.I.

I O' .

JJ

_. , a -

u : i a z z z e a z a i

}t .

a:

s

., :u !

- - 1 t_

4 4 .

.t,

.a .

M I

R:p3r: N3. 02 1040 1095

R wis t o rt 1

__ : . . _ __ .l

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

. j l

_=-_= ___

.j. .

., Page 55

*

1

f. ?. ?.

?. ?.

?. f. . _. . .

i . . . . . . . . . . .

3 i . . . . . . . . . . .

i A i

l i

. (

_- _____.--__= c___ =

'___=_______

.a .

l 1 .

. e ..

4

.. . .- f.. -

f. -

.e. .

3, . . . . . . . .

l n l l _ _ _ _ _

1 . .. -

1 -

. f

.i 2

2 *- *

o .; g g . g . . . ; .

f . . -. .t. . .k .

e !* : n e, -- i _ _ _ i _ _ _. n .

__ _I __-__ _ __________.

l U

2 .S :

la

i 1l =

=

z_ _ .:. __

t .

l l 1

A

4 .

4 t

- _ _ _ - _. -- _.J _ _ _ _

l

: l l l l i -

=

- {

8 f..

.t f..

........ 1, g . g g ;

g4 . . . . . .

.a. .f O

.. .r . . .

7.

! _f. _

=

4 31:,

4. .

5 1 1.1 1 . . .

1 1 . 222t 2 s

r a

f d

8 a

e d

a 3 3.. . .

, da

.I

. .Iv . . J .I. . ,

8 L i 8 2 5 8 i 2 l L i s 8

4 l 1. . g3 1

g ,,

. .i g . a. ;

l -.-

.- 1 .

.I .J .g.. I i, A.#

..e 8 -

- - A, .g fg j - a i

c 3

-2".3 a 1 #

, 2 4 . :. i I .

ii .

l

. .Pt y CIj-e . .

- _ _ - s- _c . - -:_ ) _. _ _ - - _.-___Ji -__--__--__x l i f

-__ .... J --

l l

~ - - - ,.

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

4 I

Report Ns. 02 1040 1095 i _ _ _ _

Revision 1 1

. n a n

. 84 80 G S M M 81 4 89 3 S

, _ __ _ _:____ _: _ .m Page 56 I

i 9

@ q .

G

. e. . d. d. e. @ . . .

e I

= .S .S. .Sa .e. e.

. m e. .e. .O.

J W W B 5 5 E E . W 5 4 . M . . e

=. =. . a.

e ,* 6 M m

.m__ am. .m .m e .e .un mmo an. '

I a l l

_ _ _ .ms um .m. emu u..mp amp .mp.mm.u. .so .ms.m .mm _ eu. m_

i me mms.m _.m..n - .- .m mes _ m.m _.m. .m _ .m .mn _ yme

.m. _

a a a e M et '

@ M l

.m um.am. _ _ .mm .me l l

.u. _ .mu un amn.u. .m. .mm um.e .mo a _ _ .ss .mp a , am 4 e.s e t e. 9 a$ .O co .O. .O. M .O.

II.

3 a=

e

.e.

M

. We -

a

.o

) 9*0 .mm__ _ _ m. .- _ p mmm .me _ m. .- .m. .um mme au. mme as. .ms l

l l l I

.e .=

=l 9 8. O W

e.t e A. * * -* d g g m O .e.

j - e

.e. o .eo e@ .e.

a.- 9 g SD 5 S . E . E An .P 9 e j

a

= 8e w

M. .h.

M . 6S

,r _. _ _ _._ _ _ _ _ _ _ __ __

,

n

, + . a a __ - _ _ _ _

1 -

at i

1, g .o r

s at-i

]. l3 I l l l

1_____

_S an

-.m.___ _ _

.l -

1 l l 1 I P. . . 3 .

..- ., f. 3 . .

a . - . ..

l . . .. .

, ;# , ; . ; . g . .g . . .g. g

. r- .

.2 , - . .- . . . J

~1

. . M.

. . . . M.... .

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

y . ,

d 1

- e 1

2..I. ..

1

. n

!e e I

a' f

d a E f 2 .j d f I .I *! 22 d e

. . . . . . . . . . . . . .1l . &

1 2 2 2 2 2 2 2 2 2 22 2 2 2

d2

jt1. I

, .t

.m _ __- _ _ - -.

. 4 11 f s

e 4 .t.> _

v . . .! .!.

3 o

r e.

s 3

g .

. 33 . 2 .s . n .

j* 4 4 4 e. 4 6 3 3 9 G s.

- 3 * -

_____e_____A(____________2__-___J______A____ . . ,-. 2 - .--l

. ===. J t_ _

R: port No. 02-1040-1095

_ _ _. _ _ _ _ R.t.v.ision 1

I l _l I_ _

I .I Pagg 57

.1

.a f.

l U 9 1 i 4

l 3 .,

= .- .

.5.-3 :.

8=

a -.

l.l _ _ _ _ . _ _ - _ _ _- .

.{

) ,

a a = e

.a - .

~

, .: 1 2'

4 .

=

., .e. a,

.E .

l n - a

= 4 s .a u a .

! = g I.

I l l

4 i

l l l l m

2 *- <

.e 4 I e

.3 . - ,

3 .

y

, a.*.1 ..s.

. I'I

. 1 .- 1 i..

2 e s 1 5.I 2* .

As

. ,3 I 2 2 2 jE . ; is 11 e

=t 3 .. : .

t _

1. . . -

. a1 .I. , ,.

~,

~ .

c < ,..

. 3 .

1, r

_ ___ . __ .~ .. .

4 i

l I 9 .

e -,. ,. -.g - - . . - ,,, , , . - - --, , ,s,--y,. , , ,y.. , , . -

t Report No. 02-1040-1095 i

R3 vision 1

__ _ _ - _ _._ : .. Page 58 j e e a a e a a iu . . . . .

a. ,. .

mm

.m. mm - .- i- ._

5

5 .

[

_ gum.____.u. _.m. _m .m..m. .m. . .mm 5

n . n l A

.m. mmm._ _ .mmum. .u. .m..m. m..-

E ..

.. M s

3 m m o e e .a i 3# 9 "

_u._.m..u.

." _.mm.m.. _ .m. eu..m N _ u..u. .m.au. .mm. .m --- m .u. .mu

! 2 - i m

j .

,1 _ M. . o

. . e. e.

I . .

3 *

. (B.s 5 5 W E ,= . . . . .t .

~

. 2 - . .- - - - - - - - - - - - -. - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - -

l 1 -

3= t F

- :4 z<

.. . E t

~ __..-.f.

3 l

__._._____1._

a = . .

i s a

\

- _ _ _ - __- -_ -_ i_ .

l .i *

.l

.a i . .

i

.sa' . .E E .E .E t

_9

- e

~.

3...

1 3

=.

2 .

t . . . .

t .3 3, g

3 :.eI a a a a

! A o d a d 3 Ja 1

38 ... . ,. 5 T 2 2 2 2 2 2 2 2 3 Ja 1 .

m2

- __ _ . 1,1 I- ,,

'. .s .

. 3, 3 .

r __

a. &

. J . .t ., t .

4 J. * .

. J

. . o...

s c

. 4. . ..

. ..i j

. J. . e..

. I .

...... .. - ,n

. - - _ .- _ .... d i

)

i 1

i

,- - - - . . , , .r- - - ~ . , - , - - , - - , - . - , e,. .. an,. ..,,-,,.,n.-~v.., - - - - - - - . , . ~ . , - - , . . , - - .

._ _ - - -. _ _ . _ __- _ _ _ . _. . .--_m._ _ __.- . . _ _ _ __ _ .

F t

1 R:psrt No. 02 1040-1095 ,

)- Revision 1 !

, --------------------------------- Page 59 r

a

..I_____

l

_______ _ _ _ _ l_ _ _ _ _ _I_ . __________l

. . .. . i t . a..

. . . . . l t - . . . . .

i

i 4

l l l i

I i

1 ______ ___ - _ _ _ _ _ _ _ _ _ _

i l

4 l

_.2._ l

_ :* l___:*_l :=*__ l

-l I -

f.- -

~ -

f. ,

1- . . . . .

. ,. = = 1

.I 3> M f

i .

l

_ _ _ .______.=___________ ___________ _____ .

l

r. - P

.. .- 1 1

n . .~

1 .- - -

5 . .

, . .o.

.-t

= .. .

__._ _____ _____ ___ =_ ___=_______ ___ __

. : l l t

1

, ,. 43 __ - =____ ______ ____ _________________

3 l1 1 1 .

4 1 ,.

.s L :

1

___i______.I______I______.l_______ _ _ _ _ _.s______

2

.- r. ..

1 .i

, .. . g

+ .n .

.=t ,-

,. .s.:' -

t -

s i

_________-____ ____________.___________ j.. , ,

i.. .

1 ,l1

,t . . . . .t..! 2 e.

j i I j j j j g J

.g#*

J

==

d' 2

2

. . . . . ]}.. .

E i

i i 8 2 } L; e s

4,.

4 a - .3. . <I,E,1 ,:o !

4 1 .

1 I

..- 1 1 3 .J 1

i . -

i

. - E 1 - ,I 31 .

?. ~3.

i

. {# .I z e. . . -.

o) .2 .

4

.f. , .

. J. .

. .E 2,

3. . v. . J . - t d

.i:

}

.k _J 4._J.. _5

== - _

..==

I __

1 1

i 3

t 1

~ ,__ -.--- __ ,- .

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

,__..__-r..._____--.

R:p3rt M. 02-10.t0 '095 Rsvision 1

________________________ ____________Page 60 1 : j ). ;.

_ _ __ i_ =.______-_-=___._=___l_____j : .:

1,ill:!.25

~-

8-4; 2 21 1.::s J .1 :::4..:: .

>s'i:: !:.

. 3, ; vo 1: .ns, l . -

1

. 3s oi,g 2

, i :j -

l l 1-

_______._______________.l '

____ __________ . 5 : s :3:2 e

Jja.3 .

1 g . ..: i

?. ?. ?. 7. _ ::

._ ?. ?. '4: '*:

- s-.

i s..11.g*

]:.3-. jig y

_ :_ __ __ __ m __ _._ _ _ _._ : : _ _ : = _ _ _ _ _ _ _ _ _ -3.3:j.r.

= ,

=g::::.;.:

_ _ _i_ _ _ _ _.;_ _ _ _i_ _ _ _ _ l__________l

_____i_____

14=

1 l. :e 3

i 3,3, ; -

2. :.

Es

. .k-b b

b b

1 4.t.ti....tt!

.- I.*'.ag s a: - - - - - j ::2 3- lill

s. r . .e.-

r :l ---l----

_ _ _ . _ _ _ _ _ . _ _ _ _ _ . t_

_ _ _ __ _ _ __.l_ _ __ _

_ _ _ _... _ - _\.ii .:: .:.-4 l

__ ii ~::' .--2 C 88 3

t

, 1

. - .1 2 :1

.: sn 4 1.}.

e.,.

333.}.1}:5.,

! i!

A -

___.-_____I___ _ ____.____

, ,5, '

~

s

_ _ _.t_ _ _ _ _I._ _ _ _ _l_ _________________

. _ _ ~l _ _ _ _._ _ _ _ _._ _ _ ;,

l l __

1 :_

1: 3 1

-:t it:

s ! :: .:Ati n.

=.

- .I g

4- ii8:

_ _ _I_ . l

_ _ _ _l_________.--____________ l l

.: -;j:r

!.:.ji}

} j s :.: -

_________________.__.__________ i .: i :n -

) __ _ __i _ _ _ _;_ _ _ _ _t_ _ _ _.I_ _ _ _ _i_ _ _ _ _ _____

1: .Ie,l.

-:' =!

r.

t

. . m

.- 13,= ,1; i

lt

-j 1-

.j 1 l:

-j 2}: :a .

8 1 4:.2 -

I-

2 I: >,21 2 a

_____________._____.____________________ .,i,-: ..

s 1,3 :

i

}.223.:1:l

-ts

-AL.1 1- -

i j f:! *! r J'i13] : h1 i :I :I d

1 i

. U w 1 g

t

5 isr 13' -

.i "

t

.I l .t :i :

t-i 31 -

~;- r t --31 r! }8t!.* l :': j-: 357 1

,s 53 3

it .J : -

.I itl f.-

8-]

..- 33

=

ligigI l- -

t r: I- *:

e ::

si il8 1

p ja>.J.:- ,1,1 < gdj -.[

J

__1. I : .1. -_____1_.__2_1_4___2_,1_I__e_s_:

1 -1: - .i n j i ;# ;is m --- -----

_ _ _ _e _s _ _ _r _e _ _i:-:

_ _e: :: _s _ ::::

Rep:t No. 02-1040 1095 Revision 1

_ _*la_ _ _ _ _,_ l_~ _ _ _l_ _a la l l- la ,3.33,,3

- ____-.___. _ ___._____ j,=}

3 ll.3 3 g 11 3.

? ? ? ? ?

.:J.1-1:.

.g ..- J.:: 1: -

2 : : : : : :

=la r -

I::: I2

. . . . . - - :-i a

. ; ; :- :5 .

1

j

l l l l

_ _ _ _T_ __ _ _ ___ _

.:3 el.'1- ' t-

. - - _ A !

2 e

?

? ?

s

?: ;q :. 4 :. . I ,::I a !!

i .

. . : : : :: == a . - : 111.::: :

i . .

___d____d____ ____e_

. v v :- ~ ~s.L

_ _._1_ _ _ __ _I _ _ _ _ _I _ _ _I__ -._[,_ _ _ _____ _I____ .

ij]'ij:l j-'j 8 -

3., g:.3 3 g,,,:

sr :} :ts ..

si 2

?

~

? ?* :4

J:

il . tI 8 se

,_=.=____=___

___________ ____=___

s ]. :] * : . ' 13.11 -

_ _____ e ,s z l b : v:. :- !.

l l l l

-:4. - [1-1, j :I.1., -.J

,. Art.lp.:tg..rt a.l.s 3 .:.J.:::

t se 5 1' 1

e 6 : } I

,, ; _ _ _.1_ _ _ _ _.I_ _ __ _ _I ._ _ _ I_ _ _ _ _I__.____I_____

I 1 e li :

. T :::t

-ai, II :

. a.:1} :e

___.______________ y 3:--8: : r:

_ :_ _I _ _ _ _,____.I_____L_____i_____ ____.l_____ -

1.t.

:.:j 1 1.!

} j i .: .

1 s11 53 , ;

___.1__ _I______ -I_____I____________ _ 1 ,1

. r.

. = =

=. ;-

e a

2:'l

. s l'

2:

t

]

li.r. s'i

. :: i f .:t f c 4.1.I! I. . i 3

, : : : 1 a;

= .u . _ _ _ _._

_J_ ie_____

_i_ }_ _______

.i. si.

, rt

l ii,.

3 - :-1-a.j.3 3.-

1

! ! 2*. . p.!2 . m h] 2

In ,:g

,. r1- ,1)).,ii.y j j '

1e .p ]

]:

2- . ..

l 1

c

3- j- }d

..,-)

-i -,

3 sg

.1 I.,1..3.0~1].,!; iai si 2

i I ,!

.. . i .i g

.. . I 2.l 2 1 1, :

4 -' 3.1 -

3

,ra- : : I ., .$. .44} -

is

., . i. 11 .1

,I, J, a-

. , it 2 ,

) : -

n- - - - - -----

. _ _l_ _ _. _t__y _ E5_. _s _ _ _ _f _i _ _ _f __ ______! _r; __,._t__ II:: ::: ::::-

l Page 61 ,

neport No. 02 1020-1095 R3 vision 1 . g 1.3

I- I I: I I a l I 3,Ill:.23-!

3;.: el-1

'W: 2. i- 21:1 e

? ? ? ? ? ? .' 1.1. 3 :, :: :* 2 '

1-3 I;I : 'i-(;l ti- s51 ag'g.-

I ____ ____ . __ ____ .:1 :J

(

_________ l _ _ _ _ _l l } l ': 1 kin {:

,g!.;3::' !!"II#

1 i

?

2

? ?

?

M1 g 33.:3}:

T**

v' : r l ~ t . l-L .--:*

___.'__________ _________ _____ . }j i.L.i.i: :; :! !,,y

,. g .

_ :_ _i_ _ _ _ _i _ _ _ _I_ _ _ _ ______, _________.I_____ _ .

3 i nsisj;g..ty!:. .2 s I L3-ilI !

3l

- l t. :51!::

___ _ _____._____ ____ ___- t , . 3:,. - .

3 , , 3.

j

_ .__ ____.i _ _ _ _ _ _ _l _ _ _ _ _l _ _ _ __I . _ _ _ _.i 3},3 .

1, :_; _ 21,_ ,,_ _i . _ _

_ _ _1 _ _ _ _ _ .

4 . 20.i3.l}.:.1}:...:

J r ., J. 33

5 j f .i- ,

__:_.____ _i ____l l

I i (; l l l l o

]"

j

.2 T

j34 3

Ii :* ?83*

- L{}:

Il 3 8

l l I3::.:8} f-

) l

.l

l

__.__________ _____ _ ___ ___._____ ,. s, 3 1 .id:1.]

[ j a 1.: .

T

. j!l ,

T___ I .I l3 s: ' '

1 I l _

I

_______________.____ i l g...

.t, g

.: v n

I J. - s[.l ,.

a-* 1 8

' l j ..;4 I j* J.

I:

t

. 6 . ' i .t ., ; ,

__._____ ____ ____ ____ . ..g y3. ],,,,o 1

s ( I i *: 8 '8 .

l h..f:3 :. l.p]

g g 88 sj I -

f,3.

1 l1 3 1 s} 3?p c, i 8 3.i4,ijl!-

3

)) \i!s.

d I;

II g

$8i 3

13 }

u

.3 g*, g{

1..;t2 '1Sl:t

1. 11 t 5,-

v'1 3*s. 1 l til<!'I:g1s <Jij in k* l 3 1' ny ,

j l 'f.

! 2 ]

j!" 3: 3 t}j j<'2 ij a ... .....

_ _1_ _ k _ _} _ _ _: _,___.I..jl

_}_}1 __i.1___a1___

(.. :: ::e :. ::.

Pave 62

i R: port No. 02-1040-1095 f j Revision 1 Page 63

ABLE 4-6 Failure Deperience for Davis-Besse Valves !

l 1 Total valve l l l ,

l Population l Total Demands l l Failure l l In selected l Or operating lMtal Reported l Probability / l lomponent/ Failure Mechanism (Safety Systenal Nours l Failures Rate !

l l l l Istor Operated Valves l l l l l Failure to open/close l 28 l l l en demand l l 54e Demands l e i 1.s x 10-2

l l l l ,

Lir oper:ted valves l l' l l l Failure to open/close l l l l l ca demand l 14 l 182 Demands l 7 l 3.4 x 10-2 '

I I i l

heck valves: l l l l Failure to open on l 43 l l l demand l l 559 Demands l 0 l ---

1 I I I Wage l l1.3 x 106 hours0.00123 days 0.0294 hours 1.752645e-4 weeks 4.0333e-5 months l 3 12 3 x 10*4hr*1 i

r i

?

I R: port No. 02-1040-1095 Revision 1 !

Page 64 l TABLE 4-7 !

railure experience or oevis-nesse Auxiliary Feedwater Ptamps immeber l l Total l l of l Total Demands / l Deported l Failure Failure Mede l Fways l Standby sours Fallures '

Probability / Rate I i i Failure to start l 2 l 84 Demands l 3 l 3 4 x 10-2 (since ocumercial l l l l l operation l l l l .

l l l l l Failure to start l 2 1 52 Demands l 0 l ---

l (after one year of l l l l i operation) l l l l l l l l Failure to continue l 2 14 1 x 104 hrs l 3 l4 9 x 10-5h r*1 i operating l ;,er pump l l [

l l l 1 ,

l i -

f I

t I

l 1 i t 1

i I

l l

. . _ . _ _ . . . .-_m___. . . _ _ . _ _ _ _ . _ _ _ _ _ _ _ _ __ _ _ _ . _ _ . _ _ . _ . . _ _ _ .

l l.

i . t l R: port No. 02-1040-1093 !

Revision 1 l Page 65 l

. ;Ast 2 4-8 l

Failure Experience of Davis-tesse Diesel Generators I r

t INumber of l Ntal l ;otal I l Diesel l Demands / l Asported i Failure !

Failure Mode lGeneratore '

Nours Failures l Probability / Rate !

I I l Failure to Start l 2 184 Demands l 1 l1.2 x 10-2 }

l l 1 i '

Failure to l 2 160595(1) hrs l 3 14 9 x 10-Sh r*1 (1)

Stabiliae l l l l i

l 1 1 I I

Failure to con- l 2 l 767 hre 1 5 16 5 x 10-3hr-1 (2) tinue Operating l l 1 l t l / I i ,

l Failure to Start l l l l3 x 10-2 [

1 and Stabilise l l l l 1 I I I i (1) per hour of standby (2) per hour of operation I

i I-l 1

I l

t I

i

[

l !

i l i 1

+

j R port No. 02-1040-1093 Revision 1 Page 66 TABLE 4-9 i

Initiating Events Challenging AfwS l

l

\

l Offsite Power l Non-Seismic Equip- t Ewne l Availability I ment Availability I I l l Category 1 l l I l l 1 i Issa of main feedwater [ Yes l Yes mall break in ncS l Yes l Yes Steam generator overfill l Yes l Yes Iass of forced RCS Cir- l Yes l Yes culation l l l l Category 2 l l ,

t I I Loss of offsite power l No l Yes Tornado l No l Yes l l Category 3 l l l l Earthquake l No l No l l i

{

l l

I I

i I

,an .

.,u e - - - - + - , - - A 4 .. ___

R port No. 02-1040-1095 Revision 1 TABLE 4-10 Initiating Event Frequency Es*1 mates Frequency (yr*1) l l l Davis-Besse l Recomunended Initiating Event l Reference (5) l Paference (15)l Deperience i Frequency

, I i i i Ioss of Main l 30 l .70 I .67 l . 67 Feedwater l l l l l l l 1 Steam Generator l l .95 l (3) l . 95 overfill l l l 1 l l l 1 small areak LocA l 1.0 x 10-3 I .17 I (4) l . 17 in RCS l l l l l 1 l l Loss of Forced l l .04 l .3 l . 3 BCS Circulation l l l l l 1 l l Loss of Cffsite l . 2 I .32 l .31 l . 32 Power l l l l l l l l Tornado l 1.0 x 10-3 l l6.3 x 10-4 (2)(1) 16.3 x 10-4 i l l l l t Earthquake l 4.3 x 10-3 l l (1) l4.3 x 10-3 1 l l l I

(1) No events reported (2) Davis-Besse FSAR estimate (3) No events reported which activated the A2WS l (4) No events reported after " Burn in" period -

l i

i l

l 1

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

Report No. 02-1040-1093 l Revision 1 Page 68

$k 4-

!s

-4 ii 8 539e kEb yhE" h!i g3 n=<

s'4!*

% 4 b$ h I 1* 8825 1- I~

y I h w

&, f-c

@ ! E*k s a w!T:

g $4

-. kig -

4 % a E

a "4w}h

gj w 1 1 n g !="bu h ;;j4* mojig 8 I E!9 55555 El' 1Elli

! h l

WI .$3

!!l his i ds

!Es!

!ijl! ! !8 iga 1 . . !.i =

R: port No. 02-1040-1095 Revision 1 FIGURE 4-2 Page 69 Symbols Used in Fault Trees

.% nts Circle - Basic Event Diamond -- A f5 ult event that is not resolved any further. Though this is not a basic event, it is considered as if it were one in the analysis since it is not resolved any further, either due to lack of failure data at further resolution, or no further resolution is required for the particular analysis.

Circle within a diamond -- A fault event that is treated like a basic event. The reliability / availability characteristics of this event are calculated separately by a spearate fault tree analysis, and inserted in the main fault tree as if it were a basic event.

Double diamond - An important undeveloped basic event that requires further development.

A House - An event that is normally expected to occur (probability of occurrence = 1), or never to occur (probability of occurrence = 0). It can be used as a " switch" to turn "CN" or "OFF" parts of the tree.

R: port No. 02-1040-1095 FIGURE 4-2 (Cont.)

Revision 1 Page 70 Svmbols Used in Fault Trees Events l Rectangle --

1. An intezzediate event that is resolved further, or I

11. The top event.

I Gates "AND" gate.

"CR" gate.

e e~ hination gate "NOT" -- The small circle indicates "NOT". The bigger dotted circle represents the basic event A which is "NCTe d" . Together they represent the complement of A.

CR gate with N inputs (listed), used in streenlined format of the simplified fault trees.

nour :.

NPDi*Al l

l

R port No. 02-1040-1095 Revision 1 FIGURE 4-2 (Cont.) -

Page .l e

Symbols Used in Fault Trees Other "ransfer in -- The subtree below triangle is drawn elsewhere. (This I is a convenience used in drawing large fault trees. )

Transf er out -- The subtree drawn below the triangle belongs elsewhere.

- This complements the " transfer in" triangle, and an index nu=her within the triangle indicates the correct match.

f I

1 s

7\ Multiple transfers in 1 1 Multiple transfers cut i

I

R port No. 02-1040-1095 Revision 1 Page 72 FIGURE 4-3 System / Event Analysis Matrix ,

l Initiating i AI1tS Configuration l l l Event l 1 1 2 l 3 1 4 l l l 1 l l l l Event 1 Ifrequency lx l frequency 1x l frequency lx ifrequency 1x l l luava11 ability 1 l unavailability 2 lunava11 ability 3 l unavailability 4 l l l l l l l l l l 1 1 I l Event 2 l frequency 2x l frequency 2x l frequency 2x ifrequency 2x l l- junava11 ability 1 l unavailability 2 lunavailabilit, 3 l unavailability 4 l l 1 I I I l l l l l l l l etc. l l l l l l l l l l .

l l l l l l 1

i l l l l l l l 1 I I I l l I I i l l i

lw eal l l l l l

} lrrequency l l l l l lwith which l l [ [ l l A2vs is l l l l l lunavailable l l l l [

Iwkn l l l l l l challenged l l l l l l l 1 I I l e

i

Report No. 02-1040-1095 Revision 1 Page 73 5.0 Results 5.1 Relative Unavailability Ranking of ANS Configuration The unavailability of the APWS for each category of initiating event is summarized in Table 5-1. These results are based on quantitative computer analysis of detailed fault trees for each ANS configuration. 1 In general, the results in Table 5-1 indicate that design and procedures modifications implemented since the mI-2 event have improved the reliability of the ANS by about an order of magnitude. The inclusion of the analysis - based system- t improvements or a diverse third train of AIWS results in about 1

another order of magnitude reduction in APWS unavailability.

I 5.1.1 Differences Among Event Categories For a given AIWS configuration, differences among the ANS unavailabilities for different event categories are due to assumptions regarding the availability of systems and

components. The system availability is highest (lowest unavailability) for Category 1 events. With the assumption of loss of offsite power (category 2 events), the potential unavailability of one or both diesel generators contributes additionally to the AIWS unavailability. The large difference between Category 1 and Category 2 events results for the Post-mI configuration is due primarily to the assumed absence of the feed and bleed procedure as a backup means to support the ANS satety function. .( The original emergency procedures for using the startup pump in conjunction with primary coolant gy system feed and bleed apply only to the situation in which offsite power is available). With the additional assumption of loss of all non-seismicsily qualified equipment (Category 3 events), two of three A2WS water supplies are assumed to be unavailable. The potential unavailability of the remaining third water supply increases the AEWS unavailability significantly.

i 5.1.2 Differences Among AIWS Configurations l

As discussed in Section 5.2, the Pre-TMI configuration unavailability is dominated by human factors. Failure of the diesel generators contributes to AFWS unavailability for i Category 2 events. For Category 3 events, the plugging of the i common turbine exhaust line adds to the Pre-MI configuration unavailability.

t

Report No. 02-1040-1095 L Revision 1 Page 74 i Modifications to this exhaust line, implementation of the backup feed and bleed procedure, improvements in the control of other human factors and the design modifications discussed in Section 3.3 have improved the Post-MI configuration availability significantly. Dominant failures in the Post-mI configuration analysis are mechanical failures. This suggests that further improvementa to the AN S should address 1

mechanical failure mechanisms in order to increase the overall system reliability.

The isplementation of a third independent train of AN S is one !

method for addressing these mechanical failures. The Third :

Train configuration examined in this study provides over an order of magnitude improv== ant in ANS reliability. For Category 3 events, however, this Third Train provides no ;

improvement in ANS reliability, since the Third Train itself is assumed not to be seismically qualified.

. The analysis-based design and procedural modifications provide l' an alternative means to address the significant contributors to the Post "JtI AMS unavailability. The dn= hant Post "JtI configuration contributor, the failure of MOV to open/close on demand, becomes relatively unimportant with the analysis-based configuration valve alignment changes. Improved procedures l for locking manual valves in the start-up pump train enhance I the reliability of the feed and bleed procedure as an emergency backup to the ANS. The extension of the feed and bleed procedure to loss of offsite power events, in conjunction with the analysis-based configuration design changes, provides almost two orders of magnitude improvement in ANS reliability for category 2 events. The analysis-based configuration design changes provide some improvement for Category 3 events, as well.

The uncertainty analysis results for each configuration l indicate that the standard deviation is the same order of

} magnitude as the point estimate (mean value) of the unavailability. The confidence in the calculated order of magnitude of the unavailability is, therefore, high. It is t our belief that the unavailability value shown in ".hble 5-1 l are valid for making judgements about the relative ,

j reliabilities of each AMS configuration.

t l

52 Significant contributors to ANS Unavailability Tables 5-2, 5-3, 5-4, 5-5 present results of the importance ranking of events in contributing to the unavailability of the ANS for the Pre "JtI, Post "JtI, Third Train and Analysis-Based 1 configurations respectively. The importance ranking is applied to categories of similar events (e.g.; failure of motor operated valves to open/close on demand). Results are shown for each category of initiating event.

,n., ,v e -e-e,-- ,.e, ,.,,nev-e m n ,- a, e,---v -e~,.r,m -

Report No. 02-1040-1095 Revision 1 Page 75 The dominant failure mechanism for the Pre-TMI configuration is human factors primarily misalignment of locally operated and locally indicated valves. Also, improper manual throttling of the turbine admission valves is a relatively large contributor to the unavailability of this ANS configuration. Failures of motor operated valves and the turbine driven pumps are the major mechanical factors contributing to unavailability. For Category 3 events, the failure of the service water system to supply water to the ANS pumps and the potential plugging of the exhaust line are additional significant mechanical factors.

Implementation of locking procedures on manual valves and motor operated valves in the ANS have reduced the significance of these human factors in the Post-TMI ANS.

Mechanical factors have the greatest importance in this configuration. The A

nant mechanical factor is the failure of motor operated valves to open/close on demand. The failure of the turbine driven pumps to start or to be unavailable due to test / maintenance are also significant. Human factors are still significant for Category 1 events. These are associated with the feed and bleed procedure, as a backup to the two ANS trains. For Category 3 events, the failure of the service water system is an additional significant mechanical factor.

The Third Train configuration has the same failure mechanisms as the Post-TMI configuration with regard to the two ANS trains. The same mechanical factors, therefore, appear in Table 5-4. These A NS failures must occur in conjunction with a failure of the Third Train in order to fail to achieve the ANS safety function. The major fallure mode for the Third Train include both mechanical factors (failure of the startup pump) and human factors (failure to manually initiate feedwater flow via the Third Train).

. Significant contributors to the Analysis-Based configuration l ANS unavailability include the mechanical failure or unavailability due to maintenance of the ANS pumps. The failure of MOVs are significant only for the backup feed and i bleed procedure. For Category 2 events the failure of Diesel Generator No.1, which provides power to HOVs in the letdown line and the makeup pump suction line, becomes a significant contributor to the failure of the feed and bleed procedure.

I i

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

4 Report No. 02-1040-1095 Revision 1 Page 76 It should be noted that additional motor-operated valves exist

which could be used to manually isolate the letdown line.

Closure of these valves is not included in the normal letdown i line isolation procedure. Thas, credit has not been taken for this additional means of letdown line isolation in the overall evaluation, although it is expected that the operator would utilize these valves, if necessary.

i 53 Results of Combined System /Etent Analysis Table 5-6 presents results cf the combined system / event analysis for each of the four MS configurations mained in this study. The Analysis-Based configuration has the lowest figure-of-serit (lowest frequency for MS unavailability when challenged) followed in order by the Third Train configuration, the Post-MI configuration and the Pre-MI y configuration. The Analysis-Based configuration and the Third Wain configuration offer over a factor of ten improvement over the Post-MI configuration. A significant difference in Table 5-6 is between. the Pre-MI configuration and the Post-MI configuration. This is due to the system design modifications already implemented and the feed and bleed procedure which reduces the frequency for loss of the A!WS safety function in Category 1 events.

Category 1 events are estimated to be significantly more frequent than Category 2 or Category 3 events. Although the MS reliability is greatest for Category 1 events, the greater. frequency of such events makes Category 1 the major contributor to the overall figure-of-earit for the Pre-MI, Third Rain and Analysis-Based configurations. Category 2 events are the greatest contributors to the Post-MI configuration's figure-of-earit. This is due primarily to the i

lack of emergency procedures (e.g., the feed and bleed procedure) as a backup to the M S in these events.

l L - --,, .--- - _ , . _ , , - - , _ - - , . , -. ., . . ~ . , - ,

. ,, _.,_,-,.m ., _ . . - ...

)

l l

l Report No. 02-1040-1095 Revision 1 Page 77 i

Table 5-7 illustrates the relative significance of the Pre-D1I

- Post-MI changes. The overall significance of each change is defined as the reduction in the overall frequency with which the ANS is unavailable when challenged (total for all categories) assuming that only that change is made to the Pre-TiI configuration. The most significant changes are attributed to the implementation of administrative procedures to lock in position all manual valves, the utilization of the emergency " feed and bleed" procedure and the dual level controls on the turbine admission valves. The existance of the auxiliary feedwater flow indication in the control room had no effect on the analysis. The significance is, therefore, considered negligible.

5.4 Potential Connnon cause Contributors to ANS Unavailability 1 A single failure or event may cause more than one ccaponent of the ANS to fail. This conunonality in failure may result from shared power supplies, cooling water sources, operator actions, harsh environment and external causes (e.g.: fire, missiles, etc.). Where possible, such common cause contributors are explicitly modeled in the fault tree analysis. Such modeled common cause factors include electric power supplies and cooling water supplies. Other factors are not explicitly modeled. The quantitative impact of these potential common causes on the ANS unavailability is not assessed.

l l

l

R:: port No. 02-1040-1095 R vision 1 TABLE 5-1 Page 78 A N S Unavailability Event l ANS l ANS l Standard Category l Configuration lChavailability l Deviation i i l 1 l Pre-mf l3 3 x 10-2 l1.2 x 10-2 l Post-mI l6 6 x 10-4 l3.3 x 10-4 l Third Train 14.5 x 10-5 12.7 x 10-5 l Analysis-Based 13 3 x 10-5 12.0 x 10-5 7 l l l l l 1 2 l Pre- MI l4.1 x 10*2 l1.4 x 10-2 l Post-mI l5.5 x 10-3 l2 2 x 10-3 l Third Train l1.4 x 10-4 l1.2 x 10-4 ,

l Analysis-Based 19.3 x 10-5 l6.7 x 10-5 g I I l I I I 3 l Pre-MI l8.8 x 10-2 l2 6 x 10-2 l Post-MI l1.9 x 10-2 11.2 x 10-2 IThird Train 11.9 x 10-2 l1 2 r.10-2 l Analysis-Based 11.1 x 10-2 l1,1, x lo-2 l l l 1 4

I I

, - - , - - - *w grwy--waw y 7 y- w w-e-- -y- $ha'"-e- m-

R port No. 02-1040-1093 Revision 1 TABLE 5-2A Page 79 Significant Contributors to AFWS Unavailability PRE-TMI Configuration Category 1 Events Importance Mechanical Factors:

noter operated valves fail to open on demand .42 failure of turbine driven pump to start .10

- failure of turbine driven pump to continue operating .01 turbine driven pump in test / maintenance .08 Human Factors

- valve misalgirunent - recirculation line .81

- valve misaligrunent - test line .02 improper throttling of turbine admission valves .15 1

I I

I i

l l

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

R: port No. 02-1040-1095 Revision 1 TABLE 5-2B Page 80 Significant Contributors to AFWS Unavailability PRE-TMI Configuration 1

Categorf 2 Events I:sportance Mechanical Factors:

- motor operated valves fail to open on demand .37

- failure of turbine driven pump to start .08

- failure of turbine driven pump to continue operating .01 turbine driven pump in test / maintenance .06 Electric Power Factors:

- failure of diesel generator to start .035 failure of diesel generator to continue operating .075 Human Factors:

- valve misalignment - recirculation line .74

- valve misalignment - test line .02 improper throttiing of turbine admAssion valves .13 l

l l

ROport No. 02-1040-1099 Revision 1 TABLE 5-2C Page 81 Sienificant Contributors to AFWS Unavailability PRE-TMI Configuration Category 3 Events Importance 1

Mechanical Factors:

1

- : noter operated valves fail to open on demand .30

- failure of turbine driven ptsap to start .0e

- failure of turbine driven pump to continue operating .004

- turbine driven pump in test / maintenance .04

- failure of service water system .24 plugging of turbine exhaust line .11 I Electric Power Factors:

- failure of diesel generator to start .031

- failur3 of diesel generator to continue operating .067 Human Factors:

! - valve misalignment - recirculation line .45

- valve misalignment - test line .01 improper throttling of turbine admission valves .08 l

l i

R port No. 02-1040-1095 l

TABLE 5-3A Revision 1 Page 82

. Significant Contributors to AMS Unavailability i

Post-TMI Configuration Cat.egory 1 Events Importance

! Mechanical Factors:

actor operated valves fail to open on demand .89

.29

- failure of turbine driven ptnap to start 1 - failure of turbine driven ptany to continue operating .03

- turbine driven pump in test / maintenance .18 4

Startup Pump with Feed and Sleed Mechancial Factors:

1 l ' - motor operated valves fail to open on demand .02 isolation valve on letdown line fails to close .10

on demand air operated valves fall to open on demand .05

- PORV fails to open on damand .10 failure to startup pump to start .02

, - failure to startup pump to continue operating .005

- failure of makeup tank water supply .05 4

Human Factors:

operator falls to isolate letdown line .09 operator falls to start startup pumps .03 operator operates PORY incorrectly .05

- valve misalignments - borated water storage line 0.09 valve misalignments - valve M102 (startup pump train) 0.35 l

i b

4 l

l l

R port No. 02-1040-1095 Revision 1 1

TABLE 5-3B Page 83 j i

Sienificant Contributors to AIWS Unavailabilit*/ )

Post "MI Cenficurstion Catecorv 2 Events Importanc e Mechanical Factors:

motor operated valves f all to open on demand .86 failure of turbine driven pump to start 25 f ailure of tiarbine driven pump to continue operating .03 turbine driven pump in test / maintenance .18 Electric Power Facters:

- failure of diesel generator to start .038 failure of diesel generator to centinue operating .082 O

1 2

Riport No. 02-1040-1095 Revision 1 TABLE 5-3C Page 84 Sienificant Contributors to AFWS Unavailability Post-DtI Conficuration Categorr 3 Events Igortanc e Mechanical Facters:

motor operated valves f all to open on destand .42 failure of turbine driven pung to start .09 f allure of turbine driven psamp to continue operating .01 turbine driven pump in test / maintenance .07 f ailure of service water system .54 Electric Power Facters:

- failure of diesel generator to start .029 failure of diesel generator to continue operating .062 4

i 1

. _ _ - __ -. - - ..-~.

R: port No. 02-1040-1095 Rsvision 1 TABLE 5-4A Page 85 Significant Contributors to AIWS Unavailability Third Train Configuration Category 1 Events Importance j Mechanical Factors:

4 - :notor operated vavles fail to open on demand .89

- failure of turbine driven psamp to start .28

- failure of turbine driven pump to continue operating .01 failure of startup pump to start .29

- failure of startup pump to continue operating .06

- turbine driven pump in test / maintenance .19 Human Factors:

- failure of operator to start startup pump .60 1

1

]

4 1

I i

i I

ir l

I

- , - - - _ _ , 3.. , - , , , . e m ._ ,. .--, , ,-..__ , ,,-- y,- y. .m.. -

y ... , - , _,

R port No. 02-1040-1095 ROvicion 1 TABLE 5-4B Page 86 Significant Contributors to Alvs Unavailability Third "frain Confiauration Category 2 Events Importance Mechanical Factors

- motor operated valves fail to open on demand .80 failure of turbine driven pump to start .19

- failure of turbine driven pump to continue operating .01

- failure of startup pump to start .04 turbine driven pump in test / maintenance .14 Electric Power Factors ,

- failure of diesel generator to start .35 failure of diesel generator to continue operating .52 failure of bus D .04

' Human Factors failure of operator to start startup pump .16 4

I I

R port No. 02-1040-1095 R2 Vision 1 TABLE 5-4C Page 87 Significant Contributors to AFWS Unavailability Third Train Confiauration Category 3 Events Importance Mechanical Factors

actor operated nives fail to open on demand .42 failure of turbine driven pump to start .09

- failure of turbine driven pump to continue operating .01 turbine driven p op in test / maintenance .07

- failure of service water system .54 J

Electric Power Factor::

- failure of diesel generator to start .029

! - failure of diesel generator to continue operating .062 i

i I

R port No. 02-1040-1093 TABLE 5-5A Revision 1 Page 88 Significant contributors to Anis Unavai? ability Analysis-Based Configuration 1

Category 1 Events Importance Mechanical Factors:

motor operated valves fall to open on demand 0.019 failure of turbine driven pump to start 0.69 failure of turbine driven pump to continue operating 0 07

- turbine driven pump in test / maintenance 0.48 ,

Startup Pump with Feed and Bleed Mechanical Factors motor operated valves fail to open on demand 0.03 isolation valve on letdown line fails to close on 0.25 demand air operated valves fail to open on demand 0 06 1 PORY fails to open on demand 0.26 failure of startup pump to start 0.05 '

failure of startup pump to continue operating 0.01 I

Human Factors

- valve misalignments - borated water storage line 0 11 operator fails to isolate letdown line 0.08 operator fails to start startup pumps 0.03 operator operates PORV incorrectly 0.08 J

I

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

R: port No. 02-1040-1095 ev sion 1 TABLE 5-5B Page 89 Significant Contributors to AlifS Unavailability AnalyFiS-Based ConfiquYation Category 2 Events Importance Mechanical Factors:

actor operated valves fail to open on demand 0 01 ,

- failure of turbine driren pump to start 0.77

- failure of turbine driven pump to continue operating 0 08 turbine driven pump in test / maintenance 0 41 Startup Pump with Feed and Bleed i Mechanical Factors:

PORY fails to open on demand 0.08 isolation valve on letdown line fails to close 0.08 on demand g

' - failure of startup pump to start 0.008 Electric Power Factors:

- failure of diesel generator to start 0.24 failure of diesel generator to continue operating 0.52 4

Human Factors:

- operator fails to start startup pump 0.03

- operator fails to isolate letdown line 0.02

- operator operates PORY incorrectly 0.02 J

I

_ ,. . - _ _ . . = . -_ . _ . . _ . - _ - . _ . - - - .

Report No. 02-1040-1096 Revision 1 TAst,E 5-Sc Page 90 Significant Contributors to Afws Unavailability Analysis-Based Configuration L Category 3 Events Deportance

, Mechanical Factors:

- motor operated valves fail to open on demand 0.18

- failure of turbine driven pump to start 0.09

- failure of turbine driven pump to continue operating 0 01

- turbine driven ptmp in test / maintenance. 0 05

- failure of service water system 0.89 1

[

Electri: Power Faators: l

- fallare of diesel generator to start 0.002 failure of diesel generator to continue operating 0 005 4

I I

i 1

i

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

R: port No. 02-1040-1093 Rsvision 1 l

TABLE 5-6 Combined System / Event Analysis Results Page 91 l l l Frequency of A WS l l l l Unavailability (yr*1) l IFrequency l l l l Initiating Event l (yr~l) l Pre-MI l Post-MI l Analysis-Basedl Third Train l l l l l 1 l l 1 Category 1 l l l l l l l l l l

- 1:co of main feedwater l .67 12.2x10-2 l4,4xio-4 12.2x10-5 13 0x10-5

< - cteam generator overfilll .95 l3 1x10-2 l6.3x10-4 13.1x10-5 l4,3xio-5

- small break in acs l .17 15.6x10-3 l1.1x10-4 15.6x10-6 l7.7x10-6

- 1000 of forced RCS l l l l l circulation l .3 l9.9x10-3 12.0x10-4 l9 8x10-6 l1,4xio-5 l I l l I 1 i i l i

-! Category 1 Tbtal l l6.9x10-2 l1,4xio-3 16.8x104 l9.4x10-5 l l l l l l 6 1 1 I i

, category 2 l l l l l l l l l l 1

- 1scs of offsite power l .32 l1 3 10-2 l1 8x10"3 12 9x10-5 14.5x10-5

- tornado 16 3x10~4 !:;.6 10-5 l3.5x10-6 15.8x10-8 l8.8x10-8 I I l I l i i l I i Category 2 75tal l l1 3x10-2 l1.8x10-3 l2.9x10-5 l4,$xto-5 l l l l l I i l l l Category 3 l l l l l l l l l 1

- earthquake l4.3x10-3 l3 8x10~4 18.2x10-5 14.8x10-5 18.2x10-5 l l l l I i l l l l Category 3 Tbtal l l3.8x10-4 18.2x10-5 l4.8x10-5 18.2x10-5 l i I I I

> l i i i i Frequency with which l l l l l ANS 10 unavailable l l l l l when challenged l 18.2x10-2 13.3x10~3 l1 4x10~4 l2.2x10~4 (Tetal for all l l l l l cntegories) l l l l l l l l l l l

1 R: port No. 02-1040-1095 R; vision 1 TABLE 5-7 Page 92 Relative Significance of the Pre-MI - Post-MI Changes i

overall Significance *

1. valves on Train 1 ( AF-360, AT-3870, and the 4.4 x 10~4 I Main Steam Turbine Admission Valve MS-106) are powered off DC power supplies.

j 2. hrbine exhausts are redundant and seismically 4.0 x 10-5 qualified.

3. AMnistrative procedures have been imple- 6.6 x 10-2

+

mented to lock in position all manual valves and local control stations and hand wheels for motor operated valves.

4. Mrbine admission valves have automatic dual 1.0 x 10-2 level control, with option for manual control.

5. An emergency procedure has been implemented 5.9 x 10-2 g to manually start and align the main feedwater i startup pump to provide feedwater to the steam generators in the event that both trains of AFWS fail. This also includes the feed and bleed procedure.

i

]

i

The "overall significance" is defined as the reduction in the overall frequency with which the AMS is unavailable when challenged (total for all event categories) assuming that only that change is made to the Pre-MI Q3nfiguration.

i i

a i

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

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

l Report No. 02-1040-1095 J

Revision 1

+ Page 93 i

6.0 conclusions !

1 1

The methodology discussed in this report provides a useful I tool with which to assess the relative reliability of various l potential MS configurations at Davis-Besse Unit No.1. The t results may be used as input to M S design decisions. Such decisions include system level decisions -(i.e.a the relative l benefits offered by the various MS configurations examined) '

and component level decisions (i.e. relative importance of l individual components in contributing to system failure).

1 Cosponent level judgments can be the basis for developing '

igroved m8 designs, not explicitly considered in this j analysis. i Najor conclusions resulting from this study include the l l followings i

! - The Pre-MI WS configuration has a relatively high

] unavailability. This is due largely to potential human 1 factor failure mechanisms.

i

- The design and procedures modifications, originally i planned or implemented at Davis-Besse Unit No. 1 l1 i j subsequent to the 3tI-2 event, effectively address the !

! major failure mechanisms found in the Pre-MI configuration analysis. h Post-TMI configuration i reliability is over a factor of ten greater than the l Pre-MI configuration reliability.

4 i - The major contribution to the Post-MI WS unavailability ,

j is the failure of motor operated valves to open/close on

) demand. Mechanical failures associated with the turbine j driven pumpe are of lesser importance.

1

- The reliability of the Post-MI " feed or bleed" method to 1 provide backg auxiliazy feedwater is not high, by itself, because of human factors. The calculated unavailability on demand is 0.14. However, this backup system does i provide an additional measure of reliability to the 3

already highly reliable ms. .

i

- The inclusion of a third independent train of Ans offers 1

an order of magnitude improvement in 28 reliability.

t

- The reliability of the feed and bleed procedure can be improved through additional administrative controls and explicit instructions for parameter response j verification. The unavailability of the feed and bleed 1-l methrA with these improvements ( Analysis-Based j configuration) is 0.06 for Category 1 events.

2 i

i

, Report No. 02-1040-1099 i

Revision 1 r i

Page 94 1

The AMS design modifications which are part of the Analysis-assed configuration enhance the ANS reliability, by thammelves, nearly as auch as the addition of a third 1 l train. The Analysis-Based design modifications, in ;

conjunction with the procedural changes, enhance the APWs reliability more than does the addition of a third train.

- The overall figure-of-serit for the ANS is dependent not only on the system reliability, but also on the frequency

with which it is challenged. Improvements in the

! figure-of-serit can be achieved through plant design and j procedures modifications which would reduce the frequency of challenges to the ANS (primarily Category 1 events).

i Such improvements may have a greater impact on the ANs

, figure-of-merit than do AN S design modifications. There j have been many such improvements made at Davis-Besse Unit

o.1 since the 13tI-2 event, but their impact on this analysis has not been quantified due to lack of sufficient performance data.

The ese of plant-specific data may have a significant impact on reliability analysis results. Where i conservative, Davis-Besse specific data are used in this l reliabilty analysis. ,

i 1

i l

I i

- _ . . . -- -.,..m ,-.,-r --- ..-m. .- ..~ .-

Report No. 02-104C-1095 Revision 1 Page 95 7.0 References

1. Nuclear Regulatory Cosmission order in the matter ef The Toledo Edison Company and the Cleveland Electric Illtainating Company Davis-Besse Nuclear Power Station, Unit No. 1 dated May 16, 1979.

2. Letter from Harold R. Denton (NRC) to Iowell E. Roe (TEco) dated July 6, 1979.

3. " Auxiliary Feedwater System Diversity Study" Bechtel Associates Professional Corp., August 22, 1980.

4. "The IMPORTANCE Ccmputer Code", Lawrence Livermore Iaboratory, March 14, 1977.

5. " Reactor Safety Study - An Assessment of Accident Risks in U.S. c-reial Nuclear Powr Plants : Appendix III",

WASH-1400, U.S. Nuclear Regulatory Commission, 1975.

6. " Guide for General Principles of Reliability Analysis of Nuclear Power Generating Station Protection System,"

IEEE-352, Institute of Electrical and Electronic Engineers, 1975.

7. " Data Snusmaries of Licensee Event Reports of Valves at U.S. Fr==narcial Nuclear Power Plants", NUREG/CR-1363, volumes 1 - 3, 1980.

8. " Data Summaries of Licensee Event Reports of Pumps at U.S. r-rcial Nuclear Power Plants", NUREG/CR-1205, 1980.

9. " Data Summaries of Licensee Event Reports of Diesel Generators at U.S. F - rcial Nuclear Power Plants",

NUREG/CR-1362, 1980.

10. "IEEE Guide to the Collection and Presentation of Electrical, Electronic and Sensing Component Reliability Data for Nuclear Power Generating Stations', IZZE-500, Institute of Electrical and Electronic Engineers,1977.

11. " Handbook of hamn Reliability Analysis with Emphasis on Nuclear Power Plant Applications", NURIG/CR-1978.

12.- Davis-Besse Unit No. 1 Technical Specifications.

13. " Nuclear Plant Reliability Data System 1978 Annual Reports of Cumulative System and Component Reliability",

NUREG/CR-0942, 1979.

l l

1

Report No. 02-1040-1095 j Revision 1 l Page 96

14. "WAMCUT, A Computer Code for Fault Tree Evaluation",

Electric Power Research Institute Report EPRI NP-803, June, 1978.

15. "A?dS: A Reappraisal, Part III, Frequency of Anticipated Transients", EPRI NP-801, Electric Power Research Institute, July, 1978.

16. EDS Calculations:

Elactric - Power System Fault Tree - Post "2I, 1040-003-002-1, Rev.1 Electric Power System Fault Tree - Pre ':MI, 1040-003-002-2, Rev. O Auxiliary Feedwater' System Fault Tree - Post- MI, 1040-003-003-1, Rev. 1 Auxiliary Feedwater System Fault Tree - Pre-mI, 1040-003-003-2, Rev. O Startup Pump as Third Train, 1040-003-003-3, Rev. O g Main Steam System Fault Tree - Post-MI, 1040-003-004-1, Rev.1 Main Steam System Fault Tree - Pre "MI, 1040-003-004-2, Rev. O Reliability Data - Human Errors, 1040-003-005-1, Rev. O Calculation of Failure Probabilities for Davis Besse Unit 1, 1040-003-005-2, Rev. 1 Failure Data Base, 1040-003-005-3, Rev. O munan Error Probability - Feed and Bleed, 1040-003-005-4, Rev. 0 Event Frequency Data Base, 1040-003-006, Rev. O Derivation of Second Moment, 1040-003-007, Rev. O

\

Report No. 02-1040-1095 Revision 1 Page 97 WMCUT Analysis - A2WS, Post- mI, Category 1, 1040-003-0 08-1, Rev. 0 WAMCUT Analysis - Anis, Post-MI, Category 2, 1040-003-008-2, Rev. 0 WAMCUT Analysis - ADiS, Post- MI, Category 3, 1040-003-008-3, Rev. O WAMCUT Analysis - A2WS, Pre-MI, Category 1, 1040-003-009-1, Rev. O WAMCUT Analysis - Ants, Pre-MI, Category 2, 1040-003-009-2, Rev. O WAMCUT Analysis - A2WS, Pre-MI, Category 3, !

1040-003-009-3, Rev. 0 l l

WAMCUT Analysis - APWS, Third Train, Category 1, 1040-003-010-1, Rev. O l

WAMCUT Analysis - AIWS, Third Train, Category 2, 1040-003-010-2, Re v. O WAMCUT Analysis - Electric Power, Post- MI, Category 1, 1040-003-011-1, Rev. 0 1

WAMCUT Analysis - Electric Power, Post-MI, Category 2, 1040-00 3-011-2, Rev. O WAMCUT Analysis - Electric Power, Pre-MI, Category 1, 1040-003-012-1, Rev. O WAMCUT Analysis - Electric Power, Pre-MI, Category 2 1040-00 3-012-2, Rev. O Importance Calculation, 1040-003-013, Rev. 1 WAMCUT Analysis - Post-MI Start-up Pump with Feed and Bleed, 1040-003-015, Rev. O Startup Plump with Feed and Bleed Fault Tree, 1040-003-016, Rev. 1

Report No. 02-1040-1095 Revision 1 Page 98 Significance of Pre .':MI - Post DiI changes, 1040-003-018, Rev. O WMCUT Analysis - Analysis Based Case, AFWS, Category 1, 1040-003-019 I

WmCUT Anaysis - Analysis Based Case, Start up Pump with Feed and Bleed 1040-003-020 WMCUT Analysis - Analysis Based Case, Category 2, 1040-003-021 W MCUT Analysis - Analysis Based Case, Ants, category 3, 1040-003-022 l

5 I

st h b

~

0 N else 4 i .

h! !N!: l' r

_3h ,. 5* $+!!![{ibs,

c. w I

m =i a a 7 _ kr!

fia 1 .. .

2 @I' :l!

-!!!o ill o

@!!!3

~~

g ih _

A 2 "Il(O '

_ h! qj;i' q! @b ;gg -

11

-He 0 = ano dl Imp tic li 6 I

'l1-i

'e b l

,.I li! !

,li?g e .!!3

' -}fG

,()

i :

-@s _

eiai33 i m , r! &-i!.

@@! l

~

I- -lO c' i !

.h g, s

-@;1* ; gl, d: i = -

-li c

~

Ti

n. a fIIa ip*

14 ; i ia -

l! a h ! ,

n!

ai -

t a

-}h ~!!!

_ h!

-uO

_It rl.!, gi. -

3!g -il O .:

b, - . . . . - - - -.

b @! f,y,e- l.,

'pllphY l;ggas,!

d'I O]e;r@l .J.;.;,@%M sl S;qf"T'l y=j r -

eG if illG - . . .

~

Q A f IFj fa @_!y, 'ly'* gi,ADI @l 5 1., .

lit[

.u , k$

! [j '

-jo T

g! *

~

Q1 '

.;r; -f;o@P-l y '!' l ;

p,ai! Q1 -

e

~ !

r slex @! 2 M

nlib@!3,a .;q &

@:! ille -

, I

@! ("

, -@)f r!o c!s'l - I*fl 4;4! -

'li 3 l'y g g

. , e. ~

$IaAg B .jjpMDI

@l .

- b@

~

tk41 -

N!! hk !

ii

-@a _- js: i

!o $aabil s ,*j lEt

\

33_hi E

spl'l!.

gi l lc !;iM,

!!o 7: <4 $).i ! %g

-lla o

4 r!! g- a

-y a -id la 1 -

g _!i ag e,,i g,p#4p

=

nO n pl , 3 II m. 1 i - !!I { '

Q k i, s i .,

kl _

~. ~

'E i ,p

-7 b ig e . ; -

!!a. 4 ..

il a g

.i t

- h <!:

~

l 4y he -i!<!r

~

@, hG gqij

!! G

, 110 gi '

It '$' s3

!6 kl '

kla['41

~d, ,

-llia-@'!

m .

So @!

,b ,

i -

lij a b..; @@!-jl,p41@!

~ .

.". @t:i .

~,

4

!!'! I

'-IIe ?!I lb"p3 3 _gjal 11 1 2 ,

31 'il , -]!s{]o2,-Tio 4!

3 1 -

i lil -g3 djjo@d;I,,i M .il 1 i I

E o

@N ii e i

-cHe-l _

- [ 5 r n, 4=.d>c' b

si I

" lI # 1 I g

-!! o to o e -

r 3 r g' !! n .;p a a

jgg

![G y*

nl] O{fil 8 5a ltiIf*et

I '

cg .k. l 3

il.

% .& @b in I"m, r!o .

S3 ~, Qi -

e!7aa P g' a 4'.g' agn ja fa

~

IIGSG r a;-!!a t.

el ,

$'l 'E ,q , ?

1 b

A- A f.

, r id'l

'I'!

(Ofo su !

@!-H Iibbe@l fl

4. e 5~4 i: @l~'

l!O g , .

<l!<s- ,1

!!O lI

~

3 -lll5';!gql' go

.g3 E li, !

~. ati

} g Q- .

- ~

an : gi I

rE S _s'i fi i;

{;iTi .s!

, Y 3

t Oi. i

,i i.

' '! nI I

-pO hl 1

-l O- O O

ii k

@l@l '@

ii, ii -

.iq

@ el iio

.cHe ~

h.,

5 e I

a; -O...

"' Jgia ib p$1 , @i f ~kf S2o@341@l '

A' l '

j y isj 7@iPGi *u n wn d l

4 ll!P ,

<4

! ei& 4 e fM j t

si 10 ~_, ~_n l

{o ei

~

@; e e i ei @ q; e.

@@, @g 1

~

@@ lb<y. iy ,j

~

-IlO 3

-G"44i@lii lip @i,1 .& ii u_

jo a ua -

I" IIl! h!

e

yg i i,i a

>'.yA ~ -

1 i.

..b i :n5 M. j i 1 143M i s

Eh$di'Q :

fE$j ggu EIi?Is1.n !

3 lkh IhI' u [' .. .

1 ei , ,

. , , t -

3 t t a ,11(,

a jg.

g

{i'j!O E '

,gog w@i k- :

i

~

1 .

lj5 i rI ~h Q6 y,I d

h il h$t k

'l k gI qIiIO

[de ,l ;-

so @,

fg.

4 ,3 , e ,t.t 1 -h:uo JiiG! k e

w 1

9 o i ne .. :

' i '

3 '

j 4' *$

L. gip .'

i g@;@ 1i 1F g; ._. - OI j

-h* 0I @I g, .ii. _

-iiO @ m! . @l

-li'g3 ww =

y '

,h !

110 2 u. -4, : i h

Of 8

g - .3 3 i.dO Td i1

{il *

.t u FJ

-1 G fT3

.I' E q

. ul1, @i m.

dis!

%g~ i!j

$i;dIk l

$ \ hb ic ;

f p @t*i , ! hk $ i_...k.

i Tdo .,

')! s Whel el:

O ' .

1 c oq 4 18 .,

il 4 peg' c.qi go q;;

t

@j $

.g

,x, pogi g P

4

~[ '

Ol OI o '

pa(g e a; . M W,@g!a! l t

r @, i m hk,ti

~

4, h)g$P$\

h\au 1

, k1k

,u,

.- W@\ \e\\

il Jo@l n' .

ilij 3h i.

e

l 1 i T ! l'Ib !

TiGN ET

! $$$$... . ;j g )m' i

e s.1 -

e$34$b'- - j q fl, w w-w' o

+n!n f .

'5 t.

! 5is g ..___

.l.f hl, .

~

Q 3

[

hk

@ ., l! -

i a=

p u ,

gen lg 1:

!!! @! I

@i !

Bb Tg e! .'d i li Qi -!!!o f i 'En di @@!;!!!s- &

37,3 - 1 All d -

it 1

-h!

dlI - -

5 g@,

pp@!

k @!. g @@!l,!

me al

h. i,'.f l kiI I

$O l 1

hi 5

.a i g I!!!k f)i$

?P +;.

e +e D me j e r i ii lie tj!! [ij j 'l

~

h1 & i

-}l -

fg , -fl@ gu v\ t th l -

5$

a- gg.[

jt 3 4- -

Y a,"je

~~

OmI 8

ip -

ilO lil 3 '

9 hl -

e' lat

! G "l

i y

ij O Ili

@h_ ; n l

4[fo, ii !' l

! i!.hl w ;i gl~ $ -ff6 -ffG d

r g, p -iEf 'N* @! k4' 0 lii.a10 t !

' _% :,liG,:p,i i , i 4. 4 qll e -}7j_ d,j,i l ' ' -

G _ii,

'!I @! $Gh!

,N EG5 3

n n.

.i i &. ,,. ;

i gj -H qi <p -

-!!e q:u -

_go '

go .

4 gg_tio 3 i

-i}$

~g!!!i

!t a h!

~~

1I @l sj a L -

1 0 ,

hl i

~

. 'l

_!!e

gg 4 a fa/

i 3,;$l j a 4l jjj hf

j! f$h j 4'l

- 1

' si 4o t ma a]*!

go$%,;O,skst .

... ,,@l ;..+

.A. ,

fief _ el I

@l r @l Q' 'a~d"

~ .

'!b, mi '

ga:g n! @l3,gr' d

E' @j@ ii

- @! e

@l -

u, .

s ;-ita -

il l h!ffO r hf -da g;

it o ep ' . -

f*j q i

g f f3 -

ss si Da

.l s('5 'k;lG llIa-&l @' -c

!!a _' p h -lja -

s. .

f .

13 h g'

^

1Ea 41 "

=

. g;

E

!qo i I

'5 De

@! el l!n;;f;!!

<!! M

. e @,el !!NAd fa gg' -

.i ,

i: & @m

@ g@; 4

,15 3

-hO ' il l -UOh!

& fio h-hO ,. h! !!! u -

l'! '

[! @

i OIH @@i I jgo .3!s N!!

p g-A p @1p ,

s %-

,1 l' G o!

~

-!]e; r, i iie

@ b e ,

w--ngs'A q.

5'

-Q -fp fe liiio @!

11 g3 OI illb@l 1b. b @,

z

&si r, a

<!", i '

b3,;

y n=

~

e ib & @$ljl!y? jl ,l g; g" &1 -@l 4 lf,c &l -5*~&,' lllh' n~_u? -

y'tic-@l tge[.* &l '

'il t

,c' l g le @f k"e

- ge P -

t hj *8

,1 ,

on , :1l &l ,

tc 4 ga,,

s ge

,1 l

tuc gg ;b, ;

e@!

,'i. si 5fj O' &! . ",i cO.!

u hs&!

u si '

si si 4*4,.L; "

pf ey,.n e @! G .

'gc &i sNq&.I @ @I O

e e o e $j' f

@j !I .i s'

Ol .,

.vle 0p,f gc&. w9,

, Y fc&.*

4 _Eg &l Q[' $

!E h

& >.m,

>h [ N"&

t

I Or g fe Q E rn h .i i

8

.f.w2) ,

'l epi .3a'n't

$G@I .

j fgj g@iel ! u, ,

G -,

el li,joorn >

jo fjibe!

41 !

br@l g l

-jo d!!

$1 3, 3

- f,

-tilas,8*'@l.

g3 i! @l

-ko g - @l ill @

jo y go 4' il! :-4 g'i gl

@O- -@s iro @;i

~

4 31

M h t

3 kd9 ih Q' .

I, i g p 4} .

Ikkk I i

@ N

@ liepa

, e s fkkd E*@ kbs* 1 e, g 3\

~h* n ma#

..Qs q \ 5 Wo n 9t g@e on N. i L

s c.ib p

31 u!$io .a WP q.'^8II

(

I' 11o@M3% S f M

@e g.1 4,1

.C 3 Q ge 4x:

Q3 hl G~

fg-(;'\\

g 4, 4 %*

\

age io@ ,, i

@g d' ;

,s ai

+ @g ; ,

8

I 5 ~$

l j '

t

!b l b*I n.',

$ I unli ty ;

. rw' il!lG Qf fil jip

-lio

@l y i !-

y s, -

l'I Illl &!

al ql l g ,

09@! l1 l*, y @llll si a

@ ! l Kn{s si p* g, b

?

qi j @l S 4j j

~

@t%! ,',, ',

@l 91 q;

@l. '

b'-

y g,

@l Nt l

4 p 4

M' n WR* !!l in@l

&.f

g q f 6 L- It !!

i yI l

Ll .

I b l i-

@ @O'el , .

pi !

1

@l el be!

@l go . .

.; @! l 1 @i .

e DI l b

J],'o @gl ~

el -

soI e! .

@ p @

@i 4r o afro 4 m y to,,l e .

g, g 4! ei .ei>

el

@ _,nhl mO-chki s gi l ;Il io! I -I!O {; ahl -Ha -ky [i r-113

                                $1                                  4

_g .. j i p .

                                                                                        .g3             p fii
         -go A);                       M a-       4,                .

p' pu ; ei-ro + ' ne _ a e 4,4

         @<,ie@ai.*4 3*

V'.

                                                       @l 1!G 4i          i
                                                                                             $g     "
                                                                                                          ,ic
                                                                                                          ~!O           i
                                               !!o                       -{4j_!!o j,s,                            _-il4l    ';
                                                             '3t                                                               ::
                 -cF                           4                                       !4 V'I                                  E
         $p,e@I $ &-@e@i                                                                     awa          @ e@l                l i
                          ,9                             "i
 !! o 5                 -;;iq1 4                             4;-                        a4                               i
       -Hohl
         +

_g 4p4ln .i 1

                 .=.ii3pmy3
                                                    ,,  ,)             -

ii 3

e.

il a 4

                                                                 .g3                                  4ti, 'i. ,       1
                        .                                         4. y
               -                                       jlO<h4l:

4

                                                                                    @f           '

g <Pg.4, g

t sie u 5 4 il l pI 3 --i l5$

                                              -ho                  _g            h           :

3 ja qql 'I ,i l f i. 4 ' iio l siel-s 4 4I .-in.my,,41 s @-

                                                                   @l
                               ,i e
                       -Fg a      _                  ;
                             <n*g'ijql O,.g    -Ro 3 D '@l Ho@

a o @ sS p C" i' m i-H i ! M

                                 l l                                                             l 3
                )[               1Il

L wE qe t .

                            ,         @9    ,          e!
                            -c He@l .r!g@l    ! '                           .

41. -

                                               !'lI a POI      -Illa 4
                                                        -cHe          .

j

                                       !!O                       -
                                                                   <        i
                                                       -fl3. ')h ] @ql E                                                     '
            .                              -go            -@     -
                                  .           E            ,,i j        1i.1

me See

                                                                                      !        4         -
                            .                                      .                  fj     jih ! { i 5    di b i T

S 51 3 W

                                                                                        *jl i:n:1

[_ p' !- apa . 'l414i 4 l. g.

                                   -Ho l     k
                          ' !I!

_gogl 4 ,3 - 4 @l@l4!b<n.4;

                  ,4 @,_no                         s                   4i 4                      e                   '-             w
      'iio              4                e 4 -{,o            a>fbfi3@l                     4 E;

ik e *;l >i Jo i Ma g(j,g b el 4 ~ g-4 -

                         $1a             gj@4                        _g3                   .

Jag g g O e -He

                                                   ]3
                                                       -gee.3,@e        .

b i. s

                                                      <g [,

bf;

                                        -E                                '

p@ 4, m l

                                                        +\a a 4l ne          4i 5 y

e agus g g k-j q

                                           .                                     8 ilic                   1
                                                               !!3      ,-.    '.

i .. i!) o -H4 f je fliopp 33g;

                                             ~
                  @l@I            ,'@              g;
 ,$                                      +                           .
                      @l _a$               r Sn a, 4I     i 4

hf .

                      @h .qj @ hl i       ,e $ ew,       wer                            .

4-fi,llh.L i g aa e'

  -                                41          .

s, .. f! a @j et to

 @l       -H+                      -Ho@l.g3@
                      @ g3-e
                 @ge <g gi a                           i M              ,_p+fgj+.                                           .

_!!6 41 b kl

                                      }l To                                     .

i JM;a 2 I til} t p1 Ep

           'g                          {~p if;N
                                                      \  !

hi I i hl _ e dn L.._tel - a @ A p \1il . g 9-6

         ~           ed                W$

t q ga @\ e&fots

               @     %, a\
                                   #ia
           @ i)?gh p4 I

e f, *I g 41n !, {p M" @&V!?gk 4- g, i St g ga le ,, gjl ~- d

&{

F a e .

                                          ;t i

if , d y e pq c . ! 4 1 Il.

~

6 5 3

bi!!E .

                                                          ,  hiIIh!Ii-2 lid  d 1

Q e, ,, G! 4: v y. .n D 'i @i il!!:>41s'~e ei i -igia$!!!

                     ?
                                                    '            3 IP              @l        ti l                            el el                        ..l ESP,;j!!p@' go. e
                                                       <ia y  e!             ii I'

g; < Tc @ ~ el g<>1e W{ wing 'h.:jii I, , jil

                                              ,,                erii to                       4.@!
                            @:                        @l I                                    n @!il
                    ,,                     p M1!!P                     18 c,      e l .'i pl iia                       i r !!ii
                     !(O- ]j' CF J!! rit e!
                                  @hi 11     
                           @                          el sa
                                                    ,  :i -t

     . asus."
                                                    ~

I do g .

                                                                           , dj#ihI t     s t

k}

                                                                             \     E g                                   u      o?g     \

9F g ${

                #            .1   @ <g                                    a-        --
                       'P h i pg                  -

4\ @g

@\

e % . #5

                                       .ae    41 I'                               ,

WW

                                              @             il 4- a  4    g#g{2p$11                       @.

ig ga

                          'p 4            .

zo-- 4; 4 p

                  . 4e    e,                -

e ee -

                            -9*g              p4plo;i.; 4,
                            $ f4
                                                    $6           .
                                                                        ~

go @ l*@@

                            @ se ,\h
                                    ~
                                                            ~

gg@ge

                                                               +
                      .e 3*3 e
                                                                                 &\        -   --

m; v=

                                                     @                [;aIQlll  ,

g 'E*&l'1 @ sI t; 1lgdl i_!!cllllcp7 ne 4@ 1' #, lllic @l 1e g; 0:,1l gIjkc :t

                                                    &l .gc fle
                  -rac           k 2    &l                         g, g
                      'ge&,            .
                                               '            i.h    s ge                                               }c            S b*&I ?s?I b'fe                            ..

g'is @ $ 'oc$- l Qc$l e - fc !!,.: @ " e_ @ se g

g e.>,n ! @ ~ 2 &l ** @ g S gjc

                              @f                    g h.'                                                        @
   &                          @l l
   & I                        &lh,.E; e            4>%.

e nic gc$, o . fc 4 g. . hc ;; 4 . - i ze[ h~lE.fc'llC - Q LI W h.T(,g r llI c

                       >i;
       -3 l,i%~

e g4..

D i id%

Ilkk kk h,llWI I

e! a

                              -goOl                                     '

3 2,*

                                      @l 4           @
                     -ga@1                               l 431 g
                                                @i
                          $2 j       M      .
                                       -$o4LO
      ~1       ago g            .
                                             @l 32 @                                        et                        .
               @        ~

mo 4\ ag-

  .~               k
                                ,   o 41

{l o h ge 1 g k e

                                      .igo                         e g.4                        e.

g?

                                                                \  @
                          -$o@\\_        ,$                     T ga            & y,e g'g

1 p1

                                                                                                              ~          \
 '                                                                      -                     g      .

3 1 -- thactii 5 4. _uo g '

                                                                                        !~h2 5 $
                                                                                         \
                     '             ~~
                                            \-                                            ,

l*iQ'- p s _ so ti'..el 4 g -

E

i@\ ,.

5*hl gi - ' -kose g:

                               *as e
                                                                                       .g                           j ii a                           .

h i,( -tio@ .

                 ~
                                @@S*'                   QGel                     e &

q 4:>$@ ui g@o g hil ! m

!?

u y $o 4 i 3, g : go

                                               & y;;g;         -

p, . _

                                -%e @
                                                                      $* _ge b#'
                            .                      n,,. _

p - ' (

                                               @            {O-klO~h,#{ge                    e                   '
                                               .je        '            b                                         ,

p {e 5* g- o

                                        @_@ gP 4                                                         o e~p,,\
                                                                                 @ p- .gP.      ~

100

\-

n. ng >

                                               !!.                                                                                    r ;l Ig* !n 9 !jA* i ji n,'it                                                                       it *-         i
                                                                                                                        $gp        a T O                                              gf
                                  -hO                                                                                      i y      up3 r                            -}ia                                                         r
                            !!*                                                                -gl
                                          .illo IP                                        -

il O

                                             $ h of 1 ig!I-lllo g,                     -

1

                                                            !! e
                            5 l -IJo 'I' !
                                    -                                b! q,@l                 _

g'3 gi @by!!,-ge .go

r -

   ,                                     -ge                         -e +.                                     q;               ih,
   !,l                                    r--                                                            !I O G               -

d, i

                                         -H*                         -ri l e_
                                                                                                               ~

i4 3 -l!* h! _gg{~ -

                                                                                                                      -hGh!-

i n -go n 4lHe

                                                                                               -       @yg   !
                                                                                                                               -!I in-l        _.

r-

              -!!jo 9T lit o

- -Ika

                                            $ g *i                    _

il tr a- . .

                                          ;I le' h-{
                                                      -!t a

_ijg L II" l

                                                                      !! G h!
                                      @l                         . di                          '
                                         !!e

{ 5 p 11 4!*$ lei -my glNl 4e c as

     @lghlgl            .   .

gj a 4 LHJ _po .

                                     -ik 6 )1 i

4:>eis 1 4!. -jjIq,i e 41

     @l                                      O
                            ~5*_q3hl.g341 g,$1      "1    -

ei s ea. . ~. el _go@l-3*4i ei .

                                    -n il                     -lii4l      ~3*41
            ~

44lb. gi G *g* 3, ria-llio

                                    . n 4        "i3@l Ha _'        4               '

4)l ' E -10

                                    @l 4.!@<p'@               -ll>4j e

                                                                           -Ch@                                                 Ij           .    !f f
                                                         -@a_-hl                                                                'e         ?      lj
                                                                          -go
                                                                                                                                .g       g

_. .. ql i, pL I_i

                                                !!o -c@(                                                                        is!.

ll*; iel

                                         ~                   '-

E ,,i j i i g3 4 .g d@4l 91j p . N,

             .                                                                                                                          ,ii
                                                                                                                                        .i, i
                                               =

ei .

                                              -n>+      ._ 41                                   ,
                                                                                                           -n>+
                                                                                                           $!,4i
                                              $!          ~~

jIdi

                                                                                                                                     -i!4l ile       ,'3                                              il e l',
         .                              E            t                                                                 1 p             @gl4                                                  p @g!

gI

                                              -la                                               !          -

ll ! Ud T - 4! 41 E $1 . 1

                                                                           ,ll a MO -li a   _
                                                                                                                             !! O 3

i-

                                                .      -j!a.-                                       '< 1           .
                                                                                                                          ->jj f; !!i
       ~

4 h ,il i' e

                                              }lG-c p l, 4                          _                      h ,l i! G-
                                                        ~

O'je -cj Ql -

_~

                                                                            $   5   u f !:

t iI us,h-I iia h, 'l,,, lePI sel

                           .s er
                    .                                                           1 tm 1

qqbgsRe@l - g si q@l

  !!ab             y                 g
                                          ,el 4 *!p , l4       <

I b < :Il a Re as' Ql'i!4f@. I . lls

                           &hl                        .

                                                                                    -                      I     1 gl         l
                                                                                                                       )

nh gli ~T 3 t i

                                                                                                      .h 1,\

_il i@! e

        -ni>'_lIIoHo@qi 1

i'q -

            *,I
                                                     @            g3g;
                                                                  ~

g; @l@l' 'd I .

                                                                <!!.4; g3                                                .

i, i @l _iiG ;.r ,

             )1'                   .
                                       @                        -h3E*,j<g!

9 E! D<pl g'OI Dj i

                 ~ '@ @i<hl~{*iI '
                                         ~ (l g.D@                  d l igio                      .

4 ): 4>i' ~

        .44                          I       ?                                     c1
                                       @l>4)l'2 4! @l                                       -IP g;
        '7                                                                                      .

ll: -He Qj ilo

        @l h8             ,                          -iiG@l-jj3
                                       .i,s D i    E         @l

4. i, r {.;<a . .

3 .g3 4 u ,, q.; 3!.p i, fPb f@ili a i!o oj , b .I.l--

                                               ;         si I
                                                                                           ?l

t i 1 I 1

-)!

i

       .                                                                              *!!,i i

d:: '2 3 ,;}! . i INf

                     <'H4                                    pp Atya,I!aiia-i>4l
            %                                      _g3g;
                                                    ~
           .,,      @ I- @ i     4,'Ha -
                                                             ,g..

g3 <ge 3; e,. ii a 3 . Di

                      @!                        -II3 lal    .
                        @i#pi ~{ ~ , E*pp-84lO!                                               pl           l i

(!4 "

           @i f c fFei em>/p!
                    @j gl 4!
                                                            -Ha i

E"4;

  ~ic                                                  -            @,

rP Megi ca-

  @! -@s                              -yo@l_jj3 O!
                             -g3                      ii . .- ei je                     -

tl ;

                                     -clle
     ~

hO-lka- . 5 -g c 2aP !@= l ,, ll 3 aia g ic,i lil 3.3 Tio ,, b

                              ..          .!..al
                                               .  'i, .-                         Eii

9

dI g i

                                                                            -S.a-
                                                                                -                                                VI                    rdis          xj j
                                                                                          .il.I                                                         335           s             e
                                                                                                                                                                      ?-

4I -il G4y!.p!. : g gy r I

                                                                                                                             ,                                        4
                                                                                                                                         .                s
                                                                           @                                                     @i                   f[ pet,l  jiel WsWo gofa--go--ge                 -
                  ~
                                                                           -@j               i@                        @
                                                                           @-Ho@l b! .r,                         yja j y i 9 i -!!!a     -

i Ie {c ,; j go i {jei r 4+ 1 i I,ip!3 <a 1- as IIlla- g

                          -     3 ilaG-ta-g
                                                                      ~

g.g;

                   .                                                                   il ,              -Nob                          i!!!!!i
                                                                                         -g                -             -                       .

[d,4' e M $ !!'l lil@!!! ne i,j e

                                                                                                                                       .     -go ,et @.

L !!! G p - - - e!, i El 2- C--],O .h,C- - " UI o-;.p -Me _k, O,,,j

                                                                          .ge 4
                                                                                                                                             -ii a-(ig) @                           -

Mo

@,)

Sc el

_ _ _ - - - - - - ~ ' 1 g 1 8 I *E5 " i 4l15* h Elijil 1 I;ii*

                                                                                                        !!!jiI' l.

T

                                                                                                      ~i' ici@!

{e-hc E4l1 ee 4

+1-

[{

                                                        -lie           n.1 t !
                                             \

G ' M ' at' hG- O- @~3 g _

                                                 -hO~h@
                                                   -      -                        f
                             -h,e-)~lO
                              ' -~
       -@]

_ _ga @} teo

       .I,
                 ~

l ~@l III

-g

@ il @l

                           ~
                                                                 ..                                'olo!

no - t } Wi d'I '

                                                                                               , @i@}

                                         @l                                                               i;,. ar  .

I'l 'l ! -

                                                                                                          $_llij u;-

i l li OO @ $i t!!j[gII m ;g;, ofeie-se a ff t' .

                                -nla 41                  -

gf . ' gg bl' klfI,s e r

             .                         4r ga %.                                                         -

10 9,1

         !!G-ho-{aj-                                                                     _g3g; l
                                       @MI                                         1 4,'.        43  '
                                       +1' M
             +!              ;                                                          -gefl 41             4;
       .. ei              .
                                @#e      _
             ~

f d -k3 - O' G- -hOfG- O DpIe - n a -a

                   }g                                 ,

MM $ e .

                       -Bs q                                                                  -l
                                        -q Ea n o- -a     i
                                                                                               ,5l @ jo ?!

5G$e jjjI.

             @@                   4: j,        ,

[a-ja-jog,d m S

                 - -_________y-m

                                                    -          %.. _   'w            ~
                                                                                    .I fg![

5 G

                                                                            ;!y??!! !'
                                           ~
                                               @l                         !!ll]fyI E'hf'I,'..!*@i u,

i" 'of

           -litia s;41 g..
                      @l g                l ta                             bi        .
 ~

l'I [ @S . ira : {wgj j'~ l

                                      @ ! agg 1Da*E'!*@l@i g @frl a                            Ps p,

n6@l b @@i E*q, l

                                           '              1, ;

i ."

                                      @l Q)
       @!<; f
       ~

{ %,%, E ,#r@ l i sia% cf / i N '-' ~- .

gi' . i,! il u HL P i 'la,

            -li!La@!!!    til el                                                                                   l!!!a i       .

fpgj @! bih!h$I q; oe! . lr"ie! j!!!

           @@@!pd!i iil                                                                                      .

W <q un

4 ..

                                                                                   'l -
                                                                                                          @l
                               @                                                        <a   .g 3 ik                ei' T                                                                                          O!

g@a&s MDal41. ei b Of

                  'ji'
                                @                                                          Y         l    @l               -

p @i!ei ge er{o ,

                                                                  @l
           @                   ~'5"< !,1                                                              .
                              @l si                                                          -                               -
         &<ji!*
                              -lifPb@.,l 1

1 bel go l'1

                              . !' l ei                                                                                       .

g sh 1!yp 3 Gri; gi t:pp_x ci!: b@! - e .

                                                                                                       ! $f,,!

ej gc gg. f ~r w ! qa a 4%g,' lj!! i

                                    @l li iii @!                                                        g.

b fo < i!! fi

                             ]e.         ,

ei n

h -f3 , 2a ,: b@ goU"@l

                                         ;                                                   Oi
                                       >i                                                      i

$!. 41 p

          -ll!I,G
                                     @l gg O
                .                                        l                                   @!                -

fFrlbb g, @.i i

                     -Ii!!sei                         l'                                     @l
                                    @l
    &                         P
                                    <j.i,
 #@  -!>j            WYOI@l 1hel -5. I'       w! ;

i*

pg. ; el -

                                   .                            l   *l
      @                    @l                          fii        I l

i

w .*!el ![ hlt

     @*i@e
    @l
                          @i eg         .

gg

    @@l           3      @@l-                       4;;
            @i    m m           ei                         .

g . @l

@l

{o@*lglli " HI @* p.jj;.E* !!!iI

            @l                @l          3
                                                    @i
            @l                @l          -
                                               -$ a ge

{ # @l I D si

            @@l l!              @l                    4;
    ~
           @l.                el            @          '

41@l 41 E*

            ,    ei
                 @i 3.e . -ne*l
    ~"
                 @!         ll       @l             41 41 @i                    ei
  ~

OI -

b' 'il' ll>& dl ' i ;lj

                       @l               @!

[nbjil

                                                                             '!, Pi'I gog.                                       if f, " '#1
                                        @l
                                                      <il Jil 3 lll G el@@4
                              @                       je b

i, i b d ii l i

                                              -<f>l nG                           {a#ge     '

til 4j -h* a p '

8) .

I'laOI ~53

l il g 5 il 41p&<;; lij db fn E* O /s

              ~
                              @l             1)ll           .
                                       @l pg.                                 .

41

                                              ~ ~ ' ' - - ~ _ _ _
                                                                                           !)

s r ! ily a4, E , h l l'fg}Y'l

             &                     si                                                Lz %l
                 *i e N*                .

q, g,,

     >       g,                                                    Qf 39e
       *     &f                '                                   el El                  ira'
             @!                              N*@i 4 lr'                                '
                  @s b                         4 Ra ;@

s $*4,# @l

                                                                  @l @l lll>.

Ba @l g~ l

                        @":s                                             @l b-gag,@@l bf' s c lVlI q;ll l~I 41!    g
                                                                  @l @l SI                                                  q tag,V,lil.                                                  !
                      *i $

liig gdi g,

                                                                  @@l A"l            gg                                                  g,.
        'fp s

l.g <] .

                                                                                                                            , p=
           .                               $jli!                                                              hI! b       k sl !

i

                              -IIG-((}-}{4l J.' -   -t fp.lho!nll)
                                                                                                              .A g  5
                                                                                                                      ;g se- j
                              -E3-IIG@!                                                                             '
                               .       <H4 14                                                               -

id

                                                                                                       -h4l 4,b
             &. a 1-ae gi
                -II&                                                                                   @l' 4        g@@l! @ii                                                  .

lii'poj6 o-go-g'o 4.. te gy g - gq,i _gdgl I"i hl -lue ,"; ! q"a - - - it)l HO .. .

              .        !                                                                               4ill
                                                                                                       -h4                      ,
                                                                                                                  ~
                                       <g441 b-Hee!

N MWWW WWWWE

                               !! O- !! >- il 4i
                                       -                           ~~

4 . 8 A; aj 2

G g. b

                                                                     , =       3  .
                                                                            ,1 8  4 1  ! !

g 1 3 -T lll 's@l G

                                 .                t 3 -f!<@!

n; -i t!Il li o- !! a-gi @i I

                                          @i
         !!jo-{ija                                     .
          &                   ei 4!            q;
                   -ne e.i I

0

1 . 4 . . g

                                         -!!O{l,ilij                 !!.!gfii jh-1 1 I

sii'!d!iipj

                                                   @l                      iiig'
              -ya--Ea--Ba--@a--ga--He M -lG@I                   .

)4l-

ge_ 0 !8 $ _v a-[a-jo--ja-H$ -!ia t? l ~ - -((e -- @ @lfo-Ho--ple _ a --J b.!G Aj. ~ . ~!!C h! l {f s b: ^!!!!![5. BiTtt ! i <!; i g' !!!; 5lll5l[-h l ijjjhiTi j y g. .. -i :gl -!sta p;;o-j o ji hsl . i 1 @l @l . . ~ @l ,M ' -rsa r,l a 1. ~ n . . - ,l.,. ;. <! It ! aj , y, 4ia-tw-ga -- - &i -go-ga- g - g'i fia-g3 - Sa- @ -tea -h3 q.; @ 4 '; . Ol 4!o al e $j,!!fij dlP.I aa-ria- - ilo ih-l 7 E ' N @,E Na9 ell% . ! l af 5$ \ O OI @ l $lhl$$If Nji soiegege 4; . /Q. e k, yip" Gi c @l sa 3' g OI 4' io ti S o f o f o j -q g3g; $! il e 41 e e + +: p a,; E .biaiajogog _g, 4Mija l *<an ee 33 , . e 'i!.! ! 1I go.go.g, 1* @ 41 .1 4l @ ja ?I a. ~90'Ibo \j;!Il lu w y2 M@:

t

!!G5G5Gf*e . 7 I ~ - _ _ _ _}}

ML20129D548

Contents

Text

Title:

SUMMARY

2.1 Background

2,;

Navigation menu

ML20129D548

Text

Title:

SUMMARY

2.1 Background

2,;

Navigation menu

Search