ML20211D954

Non-proprietary Probabilistic Analysis of Reduction in Turbine Valve in Test Frequency for Nuclear Plants W/ Westinghouse BB-296 Turbine W/Steam Chests

ML20211D954
Person / Time
Site:	Byron, Braidwood
Issue date:	09/30/1996
From:	Monty B WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML20046D820	List: ML20211D934 ML20211D942 ML20211D949 ML20211D954
References
WCAP-14733, NUDOCS 9709290304
Download: ML20211D954 (64)

Text

"

s

6. '

a ;.g.A.%; L =: 9,s.:;r:.v 4 ,p.klk'.e -, ,c 3a f apAn z.y f.

.r

~

- I ' *'

Qp w :.

< - e

.,w.pW.gl3'M

[.

NQ,9.@g.M'$2:#[%s;o,.j.c.g.v?

2 4 n:s g %Q. '

d, -

. eh,& f. &" My):p :9, .

g! gp.. %

,..?

ig SE . i '*f ig;fy,'. b yj , , . k..c'

.s

.)

j .

gI?y(G y t{.4 MJes.

' N.

ggy( .'g i --

.c :a  ; 1S y . ., 4 x , 7 g . '3t 3

'[ y ', '

~.

', g,. 'A "..

" ,! .I 4y ,[? , if " ,

s. , . .

g.p ,

d-o.- - - -

f .

f .

't.

' m. .

9 (

. '. 3

* ~ i

,f .; ,

^_ cp: M e c~  ?

Y6(l-lp4 w- q .a

' kyd h j .,. l ,' 7 :

s l

?

'- Y;E; ikj.ii1?'s; N "', d>'.1k('{ i$'

j- i !.:

) j, .-

q(?f .

p; y;

( "

j.> "

.f [. , ;; [3..s;;' Fy. k ,. . . .

[ 4.- -

4, ,. . . , . . . -

e

g. . 4 E:. ,

. ',q.  ; 1 !.

--  : ., t..

t 2 '..; }c..

9

-( , p,, , x-

-, . - p- -

- ' &#ng i 4 e . . ,

74

..g yy .

_g

% w; .

[g Wy{. , I l W . S :

_t. ;-

lL{si [.

l q

f'.

a .. . . 1-a r, .

1. C '

..J. 'i ( ) - s 4 .- ,, p p ,. y

~

-* l-

[ l! M i [, ' i. 1 9 .'. . p . 7 L. 3 ,; '; '4 3 f 4 M}

% y [h I h  ;

'I

.i v .

I;:U. p . - -; j .g / 1 MW

~

j ; .h. ..

'l y y; xe a q . . ,  ;.r.

-, y _. n.

k .. -

'.. [ 1 j .

E q yd.

ex jy ;g F;  :  :  ;

%q, g y% . 7 as e 2 1 l 1~ '

-a ,. .

i .~ .% 4 L cm }

3 3 f . -

y)

F

(' , _

n -

3 4 '- ~

't ' -

, 4. : y. g . ..-j '

_ ;> i

':A

q > .' mee

w. .- . ' ' . a  ?'

1

. ;3.- j -

0_ .

s, . .

3 3y r 3+. . u  : . .. . .

C-9709290304 970924 j

' i 4 POR ADOCK 05000454 .! . . ~1 3. J

' gM p PDR g A. . . _ _ _ . _ - - - . . --

.. l*.

- S _-

s a > . . .

+

_m p_ .. 9 1;

hNhh @[

iy h@h' h

Ls y:i e www[ ks phn hh k e jgdHj$$Jgggg,g#@op$j,Q g MehrP45WWBW 4 de 1

%$b$&W@k um t gt%m,y qw e$mWlWSWA m Sh AW y

,m n AA,n e eMMAfwn e mn s@n*@p n dapaws W

. o

.sx M ypy?v3 eNkk w mkhh n

kk b,3u ese/bkNkhhkh m ggemmw,e :ap y m;pwny hb y[k,ghhNhEdNhN sm 4g4 d m + ~

7 Q@y$m%fk.yM$MI$: i w.47ay' M!Mh bweawa h$q$$$m,$lfdhk+s%g,%m hgd$$

g

@mm d my M e& s W qde mN w? @w%m&w$ff[ M mme&M$s@h kvv4ys w m Nh)4@$4 h kh qNmn&MGAsg#gmpisbbhkhhhh pg$h@.fkh.hg#pWN b hgN hhf h e

w a

vg@?+P+g%mbm%ywnuww w

g $p2 gm v e r.w w d% MMg DMOEM mewa LTipdWpp m aet W4

. p$@~>cM m m m&Kg

?s;s ,3+wc a,~4,:

h e%fA h %m y-l,2 #em.

Q i Q;t@k

tfy g;M+ex.jJ g

M v Q3 N

L4W~F p gri Ew Ma n

sv >f w 3,;pe+u p., wd;a~

o g 7 Ogd 70 g m#

7 Q @w+m$ggf v $w 0

mpMw%((gp@ k ms,y g

Q"+ 9 vM myr p erm. ayyynyayh W%

p m

' t

?. f$ dt h

/ r t 4 e 1

% , y",# e.

%(f $uyyyn/

i uf ,

dagd.@jy@ygdem espxa jR Ri4Mysu@t Ep g iflM $h g$ by M 3J l fd

%sbddsmwne %ndMMM%%E fhf jhh @Q )h

&]DI

s Westinghouse Non Proprietary Class 3 WC AP-14733

+ + + +++++

.W*<

Probabilistic Analysis -

of Reduction in .,

Turbine Valve Test l Frequency for Nuclear Plants with Westinghouse BB-296 Turbines with Steam Chests -

s -

Westin g h ouse E n e rgy Syste m s m :s 3

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

WESTINGHOUSE NON PROPRIETARY CLASS 3 WCAP-14733 Probabilistic Analysis of Reduction >

in Turbine Valve Test Frequency for Nuclear Plants with Westinghouse BB-296 Turbines with Steam Chests R. L Haessler D. P. Remlinger J. C. Bellows September 1996 i

Approved by /

l~

/W /

B. S.'Monty, Manager, Ri#ssessment Services l

l Westinghouse Electric Corporation Energy System Business Unit

- P.O. Box 355 I P

ttsburgh, PA 15230-0355 01996 Westinghouse Electric Corporation All Rights Reserved mNn217w.wpf;1t@2596

il LEGAL NOTICE This report was prepared by Westinghouse as an account of work sponsored by the Westinghouse Owners Group (WOG) BB 296 Turbine Valve Test Frequency (TVTF) Mini Group. Neither the WOG BB 296 TVTF Mini Group, any member ol the WOG BB 296 TVTF Mini Group, Westinghouse, nce any peraon acting on behalf of any of them:

(A) Makes any warranty o, ,epresentation whatsoever, express or implied, (1) with respect to the use of any information, apparatus, metnod, process, or similar item disclosed in

• this report, including merchantability and fitness for a particular purpose, (ll) that such use does not infringe on or interfere with privately owned rights, including any party's intellectual property, or (Ill) that this report is suitable to any particular user's circumstance; or (B) Assumes responsibility fu any damages or other liability whatsoever (including any consequential damages, even if the WOG BB 296 TVTF Mini Group or any WOG BB 296 TVTF Mini Group representative has been advised of the possibility of such damages) resulting from any selection or use of this report or any information apparatus, method, arocess, or similar item disclosed in this report.'

l ITL W l?W.Wplit@ 92$96 S0p%3mbef 1996

lii FOREWORD This document contains Westinghouse Electric Corporation proprietary information and data  !

which has been identified by brackets. Coding associated with the brackets sets forth the basis on which the information is considered proprietary. These codes are listed with their meanings in WCAP 7211.

The proprietary information and data contained in this report were obtained at considerable Westinghouse expense and its release could seriously aheet our competitive position. This '

information is to be withheld from public disclosure in accordance with the Rules of Practice 10 CFR 2.790 and the information presented herein be safeguarded in accordance with 10 CFR 2.903. Withholding of this information does not adversely affect the public Interest.

This information has been provided for your internal use only and should not be released to persons or organizations outside the Directorate of Regulation and the ACRS without the express written approval of Westinghouse Electric Corporation. Should it become necessary to release this information to such persons as part of the review procedure, please contact WestinDhouse Electric Corporation, which will make the necessary arrangements required to protect the Corporation's proprietary interests.

The proprietary information is provided in the classified version of this report (WCAP 14732).

l l

m:0217w.wptit@2596 Septemt* 1996

iv ABSTRACT The objective of this program is to provide the probabilistic justification for extending the test intervals of the turbine governor valves and throttle valves. This program applies to nuclear power plants with Westinghouse BB 296 turbines with steam chests, increasing the valve test interval increases the calculated valve failure probability, which is a major contributor to the probability of turbine overspeed. The annual frequency of missile ejection due to turbine overspeed is examined and compared to acceptance criterion from the U.S. Nuclear Regulatory Commission.

mA3217w.wptttH)92596 September 1996

i y

ACKNOWLEDGEMENTS The authors gratefully acknowledge J.D. Campbell, R.G. Thompson, and D.A. Dmach for their 7:!dence and contributions to this project.  ;

)

i I

mMt17w.nyt11HitEOSS aspeemist 1996 e e+ - - y ,9ew--- - ,- .,.w-- r - T- - a-- w-*v

]

vi TABLE OF CONTENTS

1.0 INTRODUCTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 2.0 REVIEW OF PLANT SPECIFIC TURBINE INFORMATION . . . . . . . . . . . . . . . . . 21 l 3.0 REVISION OF BB 296 TURBINE VALVE FAILURE RATES . . . . . . . . . . . . . . . . 31 i

.i 4.0 REVIEW OF DESIGN AND INTERMEDIATE OVERSPEED MISSILE EJECTION PROBABILITIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1 5.0 QUANTIFICATION OF DESTRUCTIVE OVERSPEED . . . . . . . . . . . . . . . . . . . . 51 -

8.0 CALCULATION OF SYSTEM SEPARATION . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 7.0 TURBlNE MIS 0lLE EJECTION FREQUENCY RESULTS AN D CONC LUSION S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1 8.0 R E FE R E NC E S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8- 1 APPENDIX A E. H. FLUID SYSTEM & LUBE DIAGRAMS . . . . . . . . . . . . . . . . . . A 1 APPENDIX B DESTRUCTIVE OVERSPEED FAULT TREES FOR VARIATION 7 AND VARIATION 8 . . . . . . . . . . . . . . . . . . . . . . . . . B 1 5

- s

= . _ _

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

vil UST OF TABLES i

Table 11 WOG BB 296 TVTF Mini Group Members . . . . . . . . . . . . . . . . . . . . . . . 1 2 Table 21 BB 296 TVTF Mini Group Design Variation . . . . . . .. . . . . . . . . . . . . . . . . 2 2 Table 31 Solenoid Valve Failure Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2  ;

Table 51 Conditional Probability of Destructive Overspeed . . . . . . . . . . . . . . . . . . 5 2 Table 5 2 Dominant Contributors to Destructive Overspeed . . . . . . . . . . . . . . . . . . . 5-3 Table 71 Annual Frequency of Missile Ejection ..........................72 t

reL\W17w.wyt:1t>402606 September 1996

,f

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

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

?

I vill  !

l LIST OF MOURES l Figuro 71  :......................................................73 I

l s

b 9

i i

I i

1 l

l-fvtM17tr.vupt:1b000008 September 1906

.-i -

..9 -.e4 +y-7.E. -g-g g y.. .,g-t N w-"'---' w '-*e"- -

11 1 INTRODUCTION Historically, Westinghouse has recommended that turbine valves be tested at periodic intervals. For some plants the technical specifications require wekly testing, for others monthly testing, and for others no technical specification requirement exists. Periodic valve testing requires a temporary power reduction that results in lost electrical generation. In addition, inadvertent reactor trip can become more likely during the transient power reduction and increase.

In recognition of the effects of turbine valve testing on plant equipment and electrical power generation, a Westinghouse Owners Group mini group was established to perform an evaluation of turbine valve test frequency (TVTF) for nuclear power plants with Westinghouse BB 296 turbines. Several early 88 296 units were built without the steam chost design, using BB 95/96 turbine valves. These units are not part of the BB 296 steam chest mini group.

This report contains the results of extending the test interval of turbine valves on the annual probability of turbine missile ejection due to overspeed, using BB 296 turbine throttle and govemor valve failure rates and system separation frequency. The turbine missile ejection frequencies for varying valve test intervals presented in this report were calculated following the applied basic methodology described in the 1987 Westinghouse report WCAP 11525,

'Probabilistic Evaluation of Reduction in Turbine Valve Test Frequency" (Reference 1).

After publishing WCAP 11525, several incidents of sticking of govemor and throttle valves in Westinghouse 08 296 steam chests in late 1987 and in 1988 resulted in the determination that the turbine valve failure rates used in WCAP 11525 for BB 296 steam chests were no longer valid. The failure rates for BB 296 steam chests were recalculated in 1988 and the resultant probabilities of turbino destructive overspeed were sent to all operating plants with BB 296 steam chests in Westinghouse Operations and Maintenance Memo 093 (OMM-093, Reference 2), included in this report are revised failure rates for 88 296 steam chest valves for the operating years since the 1988 study. Six years, 1990-1995, have been used for data collection. Using the most recent six years takes credit for improvements in design and maintenance while retaining adequate time for rare events to occur. The methodology for revising BB 296 steam chest valve failure rates is consistent with the methodology presented in OMM-093 and WCAP 11525.

The methodology developed to calculate the probability of missile ejection due to overspeed events applied in WCAP 11525 is employed in this study. WCAP 11525, Section 5.2, identifies three turbine overspeed events that can result from the failure of turbine valves to close fallowing a system separation or a total loss of load. These overspeed events are design overspeed (approximately 120% of rated turbine speed), intermediate overspeed (approximately 130%), and destructive overspeed (runaway speed in excess of approximately introduction september itse mN1217w.wpf;1b-092596

12 180%). Design overspeed is assumed when a system separation occurs and a turbine trip does not occur at event initiation, one or more governor valves or two or more reheat interceptor valves fail to close immediately, and a successful overspeed trip closes the throttle valves and the reheat stop valves. Intermediate overspeed is ascumed to occur when there is a system separation and one or more alignments of reheat stop valves and reheat interceptor valves fall to close. Destructive overspeed is assumed to occur when a system separation occurs and at least one govemor valve and one throttle valve in the same steam chest fall to close.

The missile ejection frequency results in WCAP 11525, for BB 296 steam chests, indicate that the design and intermediate overspeed failure probabilities are not major contributors to turbine missile ejection probability for BB-296 turbines. Therefore, this study focuses on calculating the probability of destructive overspeed. Generic values for the probability of design and intermediate overspeed are based upon the results for BB 296 steam chest models presented in WCAP 11525.

The analyses reported herein were authorized by the following utilities and are specific to the turbines and valves at their respective nuclear plant sites as indicated in Table 1 1.

Table 11 WOG BB 296 TVTF Mini Group Members Utility Plant Asociacion Nuclear ASCO ASCO Units 1 & 2 Baltimore Electric & Gas Calvert Cliffs Unit 2 Central Nuclear de Almaraz Almaraz Units 1 & 2 Central Nuclear Vandellos 11 Vandellos 2 Commonwealth Edison Byron Units 1 & 2 Braidwood Units 1 h Duke Power McGuire Units 1 & 2 Virginia Power North Anna Units 1 & 2 Washington Pubhc Power Supply System Washington Nuclear Plant 2 This report provides a description of the analysis and data used for WOG BB 296 mini group.

Throughout the report, referenccs are made to the detailed information in WCAP 11525. The methodology described in WCAP-11525 was reviewed and approved by the NRC (Reference 3),

introduction septemt>er 1996 mA3217w.wpt:1tr092596

21 2 REVIEW OF PLANT-SPECIFIC TURBINE INFORMATION bections 4 and 5 of WCAP 11525 describe in detail turbine valves, control systems, and turbino classifications for overspeed analysis. The information in Sections 4 and 5 was compared to plant specific information supplied by BB 296 mini group members to select the appropriato plant variation category and fault tree model.

As illustrated and discussed in WCAP 11525, the BB 296 unit has two steam chests, one on each side of the high pressure turbine. In each steam chest there are two throttle valves with two governor valves downstream. All plants in the BB 296 TVTF mini group have this steam chest valvo arrangement. WCAP 11525 also discusses two types of overspeed control and trip systems; the 300 psi system and variations of the Electrohydraulic (EH) control system.

All plants in the BB 296 TVTF mini group have the EH control system design.

The EH control system has several orders of overspeed and trip redundancy and diversity.

There are two overspeed protection control solenoid dump valves (201 OPC and 20-2 OPC),

oither of which will dump the control emergency trip fluid (ETF) line on overspeed, closing the interceptor valvos and the governor valves. The control oil system has a mechanical ovnrspoed trip valvo and a 20 AST solenoid valve, either of which dumps the autostop oil and initiates a turbino trip at overspoed conditions. The dump of the autostop oil opons an oil-operated interfaco valvo and a 20/ET solenoid valvo, olther of which dumps the governor and throttle valvo omorgency trip fluid.

WCAP 11525 uses a variation number to classify plants according to inlet valve arrangement and control and trip system type. The classification of the BB 296 mini group plants by variation number from WCAP 11525 is given in Table 21. All plants in the mini group are either Variation 7 or 8.

Differences betwoon Variation 7 and 8 concom the electrical overspeed trip mechanisms.

Variation 7 contains an electrical trip mechanism consisting of a solenoid and plunger valvo (20AST 1) that will activato with system separation due to a generator trip signal. The plunger valvo drains the autostop oil which causes the interface valves to open and the turbine valves to close. Throttle, govemor, reheat stop, and reheat interceptor valves close on this signal.

On BB 296 steam chests, the solenoid valvo is also activated by an overspeed signal of approximately 111 percent. Some plants include, for backup, an additional autostop oil solenoid dump valvo (20AST 2) which is redundant to 20AST 1. Instead of a one or two valve 20 AST system, Variation 8 has an electrical trip system consisting of four 20/AST solenoid dump valves. The opening of two of the four solenoid valves results in draining of the emergoney trip headers and the closure of the turbine valves.

Review of Plant Specific Turbine information septemtw 1996 mM217w.wptitro92$96

l I

2 Table 21 BB 296 TVTF Misil Group Design Variation Plant Variation Almaraz Units 1 & 2 7 ASCO Units 1 & 2 7 Braidwood Units 1 and 2 7 Byron Units 1 and 2 7 Calvert Cliffs Unit 2 7 McGuire Units 1 and 2 7 Nodh Anna Units 1 and 2 7 Washington Nuclear Plant 2 7 Vandellos 2 8 For plants in the mini group, variations exist in the number of reheat stop and interceptor valves. Some plants have four reheat stop valves and four interceptor valves leading to two low pressure (LP) turbines. Other plants have six reheat stop valves and six interceptor vaives leading to three low pressure turbines. Variations in the number of reheat stop and interceptor valves affect the analysis for the design and intermediate overspeed events.

Section 4 discusses the treatment of design and intermediate overspeed events. The E. H.

Fluid System & Lube Diagrams for the plants in the mini group are in Appendix A.

f i

Review of Plant Specife Turbine Information september 1996 mA3217w.wptib-o92s96

31 3 REVISION OF BB-296 TURBINE VALVE FAILURE RATES BB 296 failure rates for govemor and throttle valves were updated using the most recent valve failure and operating data. The determination of the failure rates for govemor and throttle valves was based upon the methodology used in WCAP 11525 and OMM 93.

Data for the failure of throttle and govemor valves to close has been collected by Westinghouse. The plant specific data used in this program was reviewed by the utilities in the mini group for accuracy and completeness. The collection period includes the time from January 1,1990 through December 31,1995. This time period provides failure rates based on current valve design and maintenance practices while retalning adequate time for rare events to occur. The data indicates [

)*** failures in 4,266,679 operating hours.

The govemor valve failure rate is calculated as [ )*** failures per hour.

[

)...e

]*** The throttle valve failure rate is then l [ J'** failures per hour. This calculation is also equivalent to the result obtained by using Bayesian statistical methods, [ )***

l There is a known condition of potential thermal binding of BB 296 throttle valves during i presynchronization valve testing. The factors common to all binding incidents are described in l CAL 87-03, titled 'BB296 Throttle Valves,' dated August 24,1987. The recommendations to l minimize the potential for binding are listed in OMM 091, titled ' Maintenance of BB296/0296 I

Throttle Valves', dated November 18,1988, so past thermal binding failures are not included as random failures to close on demand. Users of this study are cautioned to follow the l recommendations in the documents listed above to minimize the potential for binding of the l throttle valves.

Failure rates for other components modeled in the fault trees are the same as those in WCAP 11525, Section 7, with the exception of the failure rates for the 20/ET and 20/OPC solenoid valves. The solenoid valve model incorporates additional common cause failures of the EH trip system 20/ET and 20/OPC solenoid valves compared to the model in WCAP-l 11525. The revised modeling resulted from a review of overspeed events that occurred at the

! Salem and St. Lucie stations in 1991 and 1992. Revised solenoid valve failure rates have also been incorporated into the model. A maximum solenoid failure rate of 1.0E 5 per hour was selected from NUREG/CR 2815 (Reference 7). This is approximately 25 times greater Revision of BB-296 Turbine Ve < 91ure Rates september 1996 mA3217w.wpf.ib 092596

32 than the failure rate used in WCAP 11525. The solenoid valve failure rates used in the revised model sre summarized in Table 31.

Table 31 Solenoid Valve Failure Rates Failure Mode Failure Rate (per hour) 2010PC solenoid valve falls to open (random) 1.0E 5 20 2OPC solenoid valve falls to open (random) 1.0E 5 20/ET solenoid valve falls to open (random) 1.0E 5 2010PC & 20-2/OPC SOVs fall due to common cause 2.0E 6 2010PC & 20/ET SOVs fall due to common cause 2.0E-6 20 2/OPC & 20/ET SOVs fall due to common cause 2.0E-6 2010PC,20 2/OPC and 20/ET SOVs fall due to common 2.0E-6 cause l

Revision of BB-296 Turbine Valve Failure Rates september 1996 mA3217w.wptib-o92596 l

41 4 REVIEW OF DESIGN AND INTERMEDIATE OVERSPEED MISSILE EJECTION PROBABILITIES I l

Probabilities of turbine missile ejection at design and intermediate overspeed are not explicitly calculated for this study. All plants examined in the mini group havo reheat stop and interceptor valves, and design and intermediate overspeeds are 120 and 130 porcent of rated speed, respectively.

Misslie ejection at design or intermediate overspeed can only occur if a crack of sufficient size i is present in the LP rotor discs. The fully integral rotor construction greatly reduces the chance of formation of stress corrosion cracks that can lead to turbine missiles. This results in dosIDn and Intermediate overspeed probabilities of less than [ 7", even with inspection intervals greater than 10 years of operation (Reference 4). Typically, plants in the mini group with shrunk on discs (heavy disc keyplates) perform periodic inspections of the LP rotor discs using ultrasonic methods at 5 year intervals, or in accordance with Westinghouse recommendations. These inspections detect and monitor crack growth, if any, and eliminate the possibility of operation with a disc crack of critical or near critical size. The inspection and monitoring of cracks assures that the probability of missilo ejection at design and intermediate overspeed is sufficiently small for turbines with shrunk on discs that the impacts of these events are small in comparison to the impact of destructive overspeed. Note that Almaraz Units 1 and 2 have replaced their low pressure rotors. An evaluation by Siemens has concluded that the probability for missile generation of the new rotors is bounded by that for the Westinghouso rotors. Therefore, the calculation of the annual probability of destructive overspeed provides a good estimate of the total annual probability of turbine missilo ejection.

An allowance for the turbine overspeed missile ejection probability for d3 sign and intermediate overspeed is based upon the BB-296 steam chest models in WCAP 11525 for heavy disk and keyplate LP rotors. The allowance is discussed further in Section 7.

Review of Design and intermediate Overspeed Missile Ejection Probabilities september 1996 mA3217w.wpt.1b492596

51 5 QUANTIFICATION OF DESTRUCTIVE OVERSPEED The destructive overspeed model, developed in WCAP 11525, assumes that upon a loss of load or system separation, failure to isolate one of the four steam paths to the high pressure turbine is sufficient to cause the destructive overspeed event. Given that destructive +

overspeed occurs, all LP rotor types, including fully integrated rotors, are assumed to experience ductile failure of at least one disc, or disc section and ejection of a turbine missile through the turbine casing.

The St. Lucio EH and control oil diagrams were the basis for developing the destructive overspeed fsult tree for Variation 7 in WCAP 11525. Updates to the EH fluid system dump logic and overspeed protection control solenoid valve (OPC) common cause modeling have been made. These updates are included in the destructive overspeed fault trees used in this study.

Analogouc to Variation 7, the Shearon Harris EH and control oil diagrams were the basis for developing the destructive overspeed fault tree for variation 8 in WCAF s 1525. The Variation 8 destructive overspeed fault tree from WCAP 11525 is used in this study with changes made to the fault tree logic to model valve closure on the dump of EH fluid through the drain path from the valves to the top of the cylinder.

The following items apply to destructive overspeed fault tree quantification:

. Destructive overspeed occurs when one govemor valve and one throttle valve in the same steam chest fall to close after a system separation.

. All destructive overspeed probability results are calculated for BB 296 turbines with two throttle valves and two governor valves per steam chest, with two steam chests per turbine, and an EH control system.

. Calculations were performed for turbine valve test intervals of 1 week,1 month, 3 months,6 months, and 12 months.

. The failure probability for the govemor and throttle valves is time-related and is the fraction of time that the component is in a failed state. As discussed in WCAP-11525 Section 7.3, the unavailability is determined using the following formula:

unavailability = 0.5At where A = failure rate in failures per hour t = time Interval between tests in hours Quantihcation of Destructive Overspeed sepomber teos mA3217w.wpt:1b-092S96

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

52 The occurrence of component failures in time is assumed to be random, therefore, it is modeled by a constant failure rate A. Because A is constant, the average or expected downtime of the component is one half of the time interval, thus the coefficient of 0.5 in the unavailability formula.

• Destructive overspeed probabilities are presented as conditional probabilities given that system separation occurs.

• For components in the destructive overspeed fault tree which are assumed to be tested once every refueling outage, a 24 month refueling cycle was assumed. This conservatively bounds shorter refueling cycles.

Probabilities of destructive overspeed, given a system separation has occurred, are presented in Table 51. The results are presented for various turbine valve test intervals. ,7,,,c)

Table 51 Conditional Probability of Destructive Overspeed Typical destructive overspeed probabilities for test intervals of one to three months range from

[ J. Note that because the LP rotor is assumed to eject a missile through the turbine casing when destructive overspeed is reached, the conditional destructive overspeed probabilities in Table 5-1 are also the conditional turbine missile ejection probabilities due to destructive overspeed. The destructive overspeed fault trees used to quantify the overspeed probabilities are in Appendix B.

Quantification results indicate the govemor valve failures are the dominant contributors to destructive overspeed probability and that in comparison to valve failures, the failures of individual elements of EH, control oil, and trip systems havu a smaller impact. The exception to this is the plugging failure of the common drain lines. Table 5 2 lists the combination of failures that contribute significantly to the destructive overspeed event for each variation, as well as their percent contribution to destructive overspeed probabliity (based on the results calculated for a three-month test interval).

l l

Quantification of Destructive Overspeed September 1996 mN3217w.wptit>092596

53 Table 5 2 Dominant Contributors to Destructive Overspeed Percent Contribution to Destructive Overspeed Probability Variation Event (s) (3 Month Valve Test interval) 7 A governor valve f alls to close and 52 the ETF drain line clogs (causing the throttle valves to remain open).

7 A Governor valve falls to close and 22 the autostop oil line clogs (causing the throttle valves to remain open).

8 A govemor valve falls to close and 55 ETF drain line clogs (causing the throttle valves to remain open).

8 A govemor valve falls to close and a 23 throttle valve falls to close (in the same steam chest).

Quantification of Destructive Overspeed september 1996 mV217w.wpf.1t o02596 I

i i _

61 .

6 CALCULATION OF SYSTEM SEPARATION System separation is defined as the sudden and total loss of load on the generator, such as the load loss that is experienced if the generator output breakers opened inadvertently while the plant is at full power.

The sources of generator trip data were the Westinghouse Individual Plant Examination Initiating Event Reactor Trip Database program and the institute for Nuclear Power Operations database containing the Licensing Event Reports (LERs) for generator trip data. Generator trip data was obtained for nuclear power plants with Westinghouse BB 296 steam chests. The data collected represents 24 domestic units and 5 International units. The data covers the time period from January 198(4 through September of 1995. Generator trip data for the applicable intemational plants was supplied by the international utilities.

Generator service time refers to the time that the plant's turbine generator is operating. This parameter is important because system separation frequency (per year) is a function of the time that the generator is in service or operating. NUREG 0020 (Reference 5) contains plant information for generator on line hours. Generator service hours were available for all domestic plants with the BB 296 steam chest design. The hours were totaled from January 1986 through September 1995. Service hours for the intemational plants were supplied by the International utilities.

System Separation Freauency Results l

System separation is calculated using the following formula:

System Separation Frequency = (# generator trips / generator hours)*8760 (hrs /yr)

There were 49 generator trips reported during 1,492,038 service hours. The resulting frequency of system separation for all plants with BB-296 turbines is estimated at .29 l separations per year.

I Calculation of System Separation sepumber 1996 mM217w.wptttr092S96

71 l

7 TURBINE MISSILE EJECTION FREQUENCY RESULTS AND l CONCLUSIONS I The updated BB 293 turbine valve failure rates, system separation frequency, probability of l destructive overspeed, and annual missile ejection probability presented in this report are l applicable to all the BB 296 TVTF mini group plants, i

in verifying the suitability of turbine valve test interval, it is recommended that the general NRC acceptance criteria for turbine missile ejection (Reference 6) be used. The general acceptance criteria states that the annual probability of turbine missile ejection should not exceed 1.0E 05 per year f.*r unfavorably oriented turbines. To be consistent with the methodology of WCAP 11525 and good engineering judgement, it is suggested that an allowance be set aside to account for the fact that a complete analysis of missile ejection at design and intermediate overspeed has not been conducted. In addition, the allowance accounts for the effects of the extraction nonreturn valves. Closure of these valves isolates the extraction lines and feedwater heaters from the turbine and prevents reverse flow of steam through the lines which might cause the turbine to overspeed. Section 8,4 of WCAP 11525 discusses these valves in detail. The suggested allowance is [ )*** per year.

Evaluations performed for this study indicate that the allowance of [ )*** per year conservatively covers the missile ejection probabilities at devgn and intermediate overspeed and provides margin for uncertainties in the model and for the effects of the extraction nonretum valves.

The destructive overspeed model was constructed assuming that a loss of load or system separation had occurred. Section 6 calculates the annual frequency of system separation to be .29 per year. However, a more conservative value of .4 for system separation is used.

Therefore, destructive overspeed probabilities should be multiplied by .4 to obtain the annual probability of destructive overspeed and the frequency of missile ejection per year.

To correctly assess the frequency of missile ejection for a given turbine valve test frequency the following steps are taken:

1) Add the allowance of ( )*** to the destructive overspeed probabilities in Table 51 to obtain the conditional probability of missile ejection from overspeed.

, 2) Multiply the conditional probability of missile ejection by .4, the frequency of system I separation. The result is the frequency of missile ejection per year.

3) Compare the frequency of missile ejection to the acceptance criteria of 1.0E-05.

l Turt>ine Missile Ejection Frequency Results sepiember toes mu217w.wpttt> oo2596

72 Applying steps 1 and 2 to the results presented in Table 51 gives the turbine missile ejection frequency for varying test intervals. The missile ejection frequencies are shown in Table 71.

(+a.c)

Table 71 Annual Frequency of Missile Ejection Turbine Valve Test Interval Variation 7 days 1 month 3 month 6 months 2 moriths The missile ejection frequencies for both variations are almost the same. Therefore, to represent them in graphical form, the slightly more limiting results for Variation 7 are plotted in Figure 71. The missile ejection frequencies shown in Figure 71 meet the acceptance criterion of 1.0E 5 per year for all test intervals analyzed. These results include conservative values for the system separation frequency and the allowance for design and intermediate overspeed probabilities. The govemor and thro % valve failure rates are based on plant operating experience (primarily monthly tehng). Although extending the valve test interval is not expected to dramatically increase thr. valve failure rates, sufficient failure information at longer tests intervals does not currently exist. It is therefore prudent to conservatively interpret the missile ejection frequenc'/ results as supporting quarterly testing until reasonable failure rate data can be accumulated based on quarterly testing.

The results presented in this report supersede the results in WCAP 11525 and OMM 093.

TVTF mini group plants that currently use WCAP 11525 or OMM-093 as a basis for determining the appropriate turbine valve test intervals should use the information provided in th;s report as the new probabilistic basis for determining the turbine valve test interval.

Turbine Missile Ejection Frequency ResuttS septembw 1996 mM217w wpf.1b-092596

t 73 '

5 6

r

(+a,c) .'

l ,

t 4

t Ti i

t

?

r i

t t

P r

i .

A t,

i

't i

t

-5 4

1 P

i, i

't I-f

l. 1 I

t TurtWne ME)eciton Frequency Assues septemw toes ,

ll - m w inv.ugw1H asses l'

, m ,

i L

2'wm.-,w -+Y- - y + y- w -,- ~<7w.--, , . - , --, e,--e.--**-,--..,-.- .,,-..r---3-m,. .,---h,r-, y .u-w,-.3,s-,..h,.m-~*

81 8 REFERENCES

1. WCAP 11525, 'Probabilistic Evaluation of Reduction in Turbine Valve Test Frequency,'

June 1987.

2. OMM-093, '88296 & BB0296 Destructive Overspeed Protection,' November 1988.

3. Letter from D.C. Dilanni (USNRC) to D.M. Musoif (Northern States Power Co.), Docket Nos. 50 282 and 50 306, ' Amendments Nos. 86 and 79 to Facility Operating Ucenses Nos. DPR 42 and DPR 60: Turbine Valve Test Frequency Reduction (rACS nos.

66867 and 66868),' February 7,1989.

4. WSTG 4 P, ' Analysis of the Probability of the Generation of Missiles from Fully Integrated Low Pressure Rotors," October 1984.

5. NUREG 0200, ' Licensed Operating Reactors Status Summary Report.'

6. Letter from C.E. Rossi of U.S. Nuclear Regulatory Commission to J.A. Martin of Westinghouse Electric Corp., February 2,1987.

7. NUREG/CR 2815, 'Probabilistic Safety Analysis Procedures Guide," Volume 1 Revision 1, Section 5, Appendix C, August 1985.

l l

References sepwnber 1998 mM217w.wpttbee250s

APPENDIX A E.H. FLUID SYSTEM & LUBE DIAGRAMS Electro Hydraulic Fluid System and Lubrication Diagrams Figure Number Plant A1 Almaraz Units 1 & 2 A2 ASCO Units 1 & 2 Braidwood Units 1 & 2 A3 Byron Units 1& 2 A-4 Calvert Cliffs Unit 2 A5 McGuire Units 1 & 2 A-6 North Anna Units 1 & 2 A7 Washington Nuclear Plant 2 A8 Vandellos 2 l

l l

I Apperxfix A september 1996 mV217w wpf:1be92596

APPENDIX B DESTRUCTIVE OVERSPEED FAULT t TREES FOR VARIATION 7 AND VARIATION 8 l

l l

t 1

Apperdx 8 September 1996 ,

m W ty**Pf;1 W M96

l t

DESTRUCTIVE OVERSPEED FAULT TREE FOR VARIATION 7 I

l l

Appendix B September 1996 mm17w.wpf:ft492596

4*u -

a ,4 J- ~ ~ JM .- A we. isim, .e_ w .,,m_h+i4.l. n.+am,, gh-- as . e- # -e.- ,,-t.

h 5 .

l et i,1 i . k i . ti. -

, [i _

b , . - - - ,

lu e HLc_ L he a

i e

LL i

i g 3 L;; u ,

8 n .y lb r i 'dg Iti .

g e

r r g- .

Q.

h!

I '

E-L L e .

Wg y e" af

,,k O  ;

, , - ~

<O v *m _su 4 r , r , r ,

o &$ at g

g_

u}>

.L,;(

d l' ,

l' I

y se g:_

i ,H i 4 G IV m 3 - , r , e- , r , , , e ,

tw r - rg y fg.= ,

1,!l I - f- id , fa by y u

i, , s - - -

r b L <

jfg : I s.

L,-

. E L { eg .L k,b -I ' Y Qii_!v b:

a~ m s.<>.m J .s -.m- u en e-+ i-.mu an-

.A ,a,~ .~ 3 u e .4 ts - a _ -.A {*

I

-i I

-l

- ,9 :

! 1 1

.j =

i ye

't

• ):

1 .

$ h 1

> g

r. m m

=

r l f- t s R'

r- ':

<d 4 & eE' s f =. e.m a m w (

• y ym -

~

1*:

18 at

'E

.t t.

Y 0h m

hbg!

!!$ <]

• ba

,di ,

d . ie , li  !

I* b $sIf !c . E ai!L ds_s C i

% IY

=- I a b

i m

T r:re y

,11 = -

g,\.$

=. b- -

g me l

-r eits  !!! z Eq C M JEs

&yw Jadis dt.s sites, & d

's

'= '

. a '.

] d o

d l

Vigd , ,

g5 E - I 3 - d T - #

2* & &~

z"Yt alsy}s,

!i !5 jy f*

s_iL L h_.L L r , r , r ,

2 .c Ik 1;3;  : f:- - .

e;2 .

s=, n .

ud s-a a Is.: C psj I"a ng nev

$l8 s b d8L Sli li__L ali is c, ,

i!I .m -

Eh a -

55 !ij x ill 'liii. 'Ag .

f;i.

-ql.e g,'

'g , ,

,r a -

ip; -

i;; a

- il its su leie$s 8

, , r ,

allL I I N YY r l=  ! 7.. i!

• p*gl fnL El*3

=

5

ard,.

SL

.Il I } 5

.IS I

SL C II!

Sis

- q-I;vb i 1.

l

44-.- 4 46-g- 2Yus 4.i, .m_d_ '4 _4J~J 3 _,A X ,a.2 g A a.9-,. _--M,a 5 -- s A. -- W t

S t

1:

J t:

- ; 8.-

- "8E.

.E8 '

1

! } }=

F M as r 3 -

a q> .

ML

~

., . ....., a - ,;-..... .,

4 2. ed4w. J --.~5 u CmJ v., d.7e= w+A. 4 4- & J 42,ea E)--*==m -- .-McA.-

• ^

l.iti

$E

. li j i'

-!}-}-

c. .,

W r <3' ill F 3 F ,

R (2 (8

. J m, w

.. w,. .

e m r , ,

r . , r r E E5 8 Y

in u

-g at ra

& = & s ~ .R nJk.A &- a 2 -A-- 4&A. p A-kw- - A L~4 - =- 3 t

iI .

8E

!E-il-i e

d F 7 Oe

. i Q:

s 4:

b HL' &

L . fat

T h

1-Jt iI I

'{E.

i *. .

& 4- -

I! (!

4,u

,' y'.

L il, 3 lE.lfil i

! b II T . vu e .

C

~ ~

b a

se akt

!j  !

M!! O L dlsI3hi L all .

~

d . .!jn o .kg i kit

k. Ik T T I Al 8 ir 3 g 3 2. @ bl 3 $_ ~

ka w. ei ~e

F 1

li.

8E

.. g I:.

. Li 2,8 isg i o

}L .

.s 8

k)3 b

I ..

k 8

~

R

~

.}L'ikl!. .bWW a

b$ 3 Ib a.

hl1 3 L d_1 r i fl

$Il .

o l i-l d a .

A!h o .pl y C b

. .8L y W I E5 2 j, t -

3 3. 5 -

I 51 3 IE. 5 o L, u o at

1 T

2 g:

)f

's '!

g

• s . II . . 8!ja qi 5h -

i ii G-l,.I }l T'.[iC(I d<a; i

i

~ .

. Egg

!! I -

lg] ! -

!G  !

141 C

Iiiii.

vg v .- g

,i

- n. ,= g .

L a.-t

! b IW

. II T . d:  : 88

[1 C

~

4

[ -1i *e. fel.

1-dit./ E

!! @ >!j i L -

t ge. 8d- i8 r! -

g!E a

T u

!~

-- <]! ji!I

, <Qt bt s dE <

!yE (!

e I -

I){

I I "g la.:

(b -

y 5ls,!s 5

$js;5.Iv $~L. klL s

, s , , , , ,, , , , , < ,

If y

!ge y . .5"

,3; t-a ra , , lE g . ,

-l

.:. !g= -

g*a,v -

v  : - -

e v-

I;.

III I la

• 1 *- e b b N__;v  !

b; ** Ib!55s L

i, I

5 e:

.sq o -

Il l'

$b

,, , , f ,

o M

dU dii E.

g E E L h,

'S L.

i

,D b

k. i.

te a

na l l f I

}i;_i ;j i g31 <]s l >:

ML . ,

~~

a SW

!;i  ! '

(I (I e-1% Eo L, ?Nm a .

vy vg v gr n-I !_ ,

!e I

I

. L,h, P

,i, m

, Lr, m= 2 Ed

S -. A M>,J t

. [l *

.h lio 1:_

!}-,}

I 7 I l o y p ag[

l* Ca . ' .s g..

d ,

Ji

, - ,i n '

k. <3 h ,,

Ei I

y ws is em. 4iJ s %n 4.- ~*e- 4,e8-n-a4a _s ,, ss4- J .Af- me- +.m .4 J es -m a a L.a eu44,wadtn.. A. _J4 -- ..e.

~.

a

~

lll .

'lt

}. 8

's' t .

.E- ,

q3.

, i n --

li t: .

d c

e

~-

h

• -- &w,1,* s4n 4 s.- 4me$a-- "-r-ub u--. -J+ e i +- u b

a l

. I;

'E

.I l--

1 '. .

ili:

F 7 Ty 7 i

i-e

}- L<]

y.

i!q g

.[c o_&

4, b

t &>

ag 6 5 4 +.4 .ih,4 A m-iJ',a4-.4 e 4 L U,4- 6@ g6g__4 4 J . 4 . p. _ .4 4 4.AJ.,eha +,-,4 .=+.. ..4-.- 4 4 .,p.ps . 5 ,e e

i i

W i

~- T .

~. 4 h e r

E '

qt' s

> < lg-- >

g3 b ;' .

8;-'f <

s-8 4

4 l

1 d

eg. -

f s -

J' r y l

i i. ' gy g f> , gg -

i'

. - - - - - i.- . ..e,_ , , , ..,. . . .,..,_., ,, , ,__ , , , , ,, , , _ _ . _

5.

-- ,t

(=-

i-

%i

• a "E

f-i},f l

w 7

di RiL .

~

C .

({ .

([

Ys 3

, - , m m g ,,e .

E;l v TL. s.e Q. .e . *% e

~

i= I is Q  !) , O t'X, b

• b .

l ' i,

. - - , , - - - - , - - - - , - - - - , - - --------,-----n,w_-,- - - - - -

e:

-E :

4$

%2 5E

'E i*in

! ;:I bE

$!f ml (I

!!5L .

L. . ,

Egg R$

a-eg

)n:

• • I g

s.

Illit ,

h I N . 4Y - 3~ . 8~

C

,-. w--? 4 - ..~ . a N

e:

~

-o-sl 5

g . 5.,

g2 0 .

(I b.

a lg <]s k.

1.

it r, ,

rg ,

2"  %

t s s e

~

fag a m. t.

g. .

c ,) , , ,

ng!

E-l-

I.

u at at

DESTRUCTIVE OVERSPEED FAULT TREE FOR VARIATION 8 l

i-l-

l Appendix B I"

- mM217w.wptib 002506

e m w -E4 4 4 -4 - a f-1:

!l -

I!

! II I I ' is i 1:  ;

i 4 i

!g i jf , ,

b Eb

!i! 31!b 3,b  !

y; i l L,! ;ii. -

a to a

b

. lii r: e s r e .

'tr r

l c . s, i-e .

58 r Me da E

b b

1.

k s  : r i

. g e :gi .

nt5 la a t~i Ef' 8

Jh gi -

C a

el y n I y 3!

g e .

fi

5. -

: - 3 8 .88 (c. .e gg b b b ,

b:

P t .d ~

,!* fI -

LL

t. M. '

l m.

1 L a

o n -

8 d ! .

Sh I. .

n$g s r=-

be i t-

.k! s c , , , , ,,

Jba-b b E . .

j .

. . . b8 ng te t i:

L,i r3 - Me fiiL EL dL = L s_1

" h44, 6sa -Aa---46 5 J-4s, --k4-*- >

-Ma+ ,-,-~ < L- &L6--t= 4-*-oe ,4M --

- I e

o a- p l -

I 3

45 5E

.E

. I *. ~

i l1 1

6

- _ _.__4 m 4 . j _ m a .

4 4

=

lt

.; e -

'I i li i *-

!}.}

i ,E i

(

r r,

. ,2 1 ,p ,

lg 3

, ,E. .~ - ., , , , s -- - - - -

ea4*aa4 J.i 04mg, .4/ m.M4 .#,2 L 4,hde 4 man _4 4 M A a. =4,4 5 J.ye.44ae-.gaida.ma..~. hA.* 4 4 .4 a ..h,. % Jd L 4 -

?

{

.  ?

.w j'3 m s:

.. j g:

s. t :

m 69 I I'-

-4 t.

4 k

J f

4

[? M 1 :.

l W l.

S i

r ,' r _,

l- ..

E

e a . - . ..___,,,4. , _ - 4 .# _ ._._, .. A _m .

ll

}t

(!

-i

.! v .

1<1 -

l .

ail ao 4L dL h 2 @ .j @ .

i'I!@u@ ml u m ,,

A e

a ri Sk1 E{3 u

f L

• • C e' T, s c eb.

l. .h.s. L F ' '

, , . . . q> n;g .

su a b n n ii!

i i

' in '

=I L L= Lr SL -

SdLL

A e a u 4.kmn-, s e.- . b k'

E

~?

I

1:  %

i8 5I I)- '

g*

!}.}

Wi I

e.

l n $

N

&; ta;

, y t- 1 d

w w*  ?* -- e%

a .- a -c. .- . arm-.4 , .e M ,. h 4..#_.

4

~I lol yil

!ii 9

L}

s

- 8 -

E- 3 LatsLage

• a -4r2 1 k 4 -4 yaR m 4 4 1-e@.f.L4 .._ama__d. .4 S3$_)E-,, J m. ,a #464 4.6- - d A .J-+4A 4f J8 J4 *+W 4% e M8J- 34 -- W-4--8444.4E--S664-4-4W8 -Adv%- B4ed4~.46.-h-5-4 #- e d- c

^

4

. . .4 4 -

g

.l

.e e- ,

. ,g.

-u g; -

.g'

( .;

3 .

I 4

a g:

4 4

{ ,

l k .

. t 4

N ,

W 3 g_t 3 .

p q .-W-* **e-ai 1--' - d1- g <rt

a a E

I g' :

u I

tr.

i .1

<)s r

bi al

<]!

F g ,

=s s

3 II r , r , r ,

k b d 3

% i n!!  :

b .k.8-IIiN. I a g  !

5 l, r ,

L' n_

. I 11 oh 5 ,. lau S.ae .iv i -

k.

r. , ,

r, q- ,

r, )

b ,

p -_% 42 -a , . -

3 {

2 1

II

?t as

^ 5 .. ,

ilt NF 5 6:11 QB. bd <35 his, h, 5 .

i "h&!L $![

,5 I

l 3 N,E A B '

lld

<CE IE_1v L'lt 2

c

!!l' -

b!

r c.

i d ,, ,,

IIll a b o

- E E I  ! -

(E di ,

d!!

. h.

en E L u L mt

l l

i ,

I

):

it , , - ,

,s ll I. L.I i. . It g - g.= 3 p,![:L Ic: ?8 E I, E

[! C

!;i diL h[L Al di m m m, lil At at L 3 ,  ! i; iL  !

a3 itt Ll  ;:i. i ii.i !

a

, , m , , , ,

,O e  ;.

'/""."'- i tle di gv! F bi -

68 e t_ r F.i - 5 NLp b, ,

Jb m

EEL .

d 1 I -

555

, 3 m dii il

,I L u g

b; o C . h.

, < , m < ,

<l I ti;-

1g-

}- li b ;I' Ie e- le.

ey  %

L j'd 3 i.e E

Edl a

, ,7 E L '*  !  !

Lh.,,l D

'* k! .E a1 8

i

< , <, , , r ,,

3 E

8- y 5E  ! -

5 i -

big 1 -

35

• sa *1 ;  !!9 5 Id r t*Ie

&'i 4 cL e e! v C

ss r , < , ,

<3

!5)!- 'r-"' 'l

' =

[; d S -

'!l.

!s; -

!.E B -

62 e <!

48 L 1. ge h3e d31L 'S C L,1 =* dL .

,I,

.f.

1 it

'l

'E

.: n Nh IEl ns c "gII df v m r ,

y f3I C hI g!ve 6:C C sae -

i)!

<]:

h~ w.s

4 i

it

'l

'E l'

,e sci d

b, S li!,

crut, m m b, * - b I ,

b ,

. D . I $ = b g $

g_r i .- i i!L tit in, git d

. b,y 3 -

by :. c;. b ,, , . b ,,,, 3 I- li sal >C I gsm . ta!ri Igr e{I t h.

I. ,

I1

2 1

5t

• E 1

,os b Il b

L, she , .

h2-. .

F ,

g I!!.

b, e, i G e '

i[i y

1. ^

Q  ! s O

i Iii I y

'E tl 5

E i

F l t,

& rd E -.

r g3 3

S??L

1:!

v B jt l1 t il1 a'

e6 Eka I

!!,\ 4 '

{fj GI' kl d' HL i sL . at .

F Wd R I. R I'I S

'frie 1<.v-V E-

, <1 -

3e s.:

I

1

)2 E

• E

[ *e. s

}. a, l I 6

I

$5 r!,

Elbg V

a k

L 3_t

I

t

• I

• E.

I*a

,e I I!

m m lH

• ti <]n in! <]s

$ S III! i k

i,kE I!!'l w'

lig <]i hs r

Y

.[hi ,

I

• e.

o i. gt

e avg b b<s ,

7 15 1 !' T ii, l ili_ i dL L to an

e i

li f

8

• ${

g s .,

<t i,

8[

• 9 , <]i b:l <]s I,,r

. EV. , rafE e 43 -

(Ide girg i

r is

{d 2 ,I

(!

iLL

!L ,

k , , rY k ^!

I g Es f5 E '

{ -k ~

$ I

$Lb bl E.

vg r . sge

! ~

!. I '

I!!, I hL L" n hL ant

..