ML17297B259

Requests Info Re long-term Operability of Deep Draft Pumps. Response Required within 60 Days of Request

ML17297B259
Person / Time
Site:	Palo Verde
Issue date:	01/12/1982
From:	Tedesco R Office of Nuclear Reactor Regulation
To:	Van Brunt E ARIZONA PUBLIC SERVICE CO. (FORMERLY ARIZONA NUCLEAR
References
IEB-79-15, NUDOCS 8202030299
Download: ML17297B259 (32)

Text

I p

Docket Nos.60-628 50-529 and 50-630 ter.

E. E.

Yan Brunt, Jr.

Vice President - Nuclear Projects Arizona Public Service Company Pos't Office Box 21666 Phoenix, Arizona 85036

Subject:

Long Term Operability of Deep Draft Pumps-Palo Verde

~ <<y

'~<>> 1~ i"Q Sill~8 1ggp TIVE """"'i DISTRIBUTION Docket NRC PDR L PDR LBA3 Reading TERA NSIC TI C FMirag 1 ia JKerri gan JLee OELD OISE(

ACRS(

IE Bulletin 79-15, dated July ll, 1979, was issued to all licensees and holders of construction permits as a result of deep draft pump deficiencies that were identified at facilities both operating and under construction.

In your response to the bulletin you identified deep draft pumps as being utilized at your facility.

However, your response to the bulletin did not include enough information to demonstrate and assure the long term operability of these pumps.

Enclosed is a document entitled, "Guidelines for Demonstration of Operability of Deep Draft Pumps."

Within 60 days from the issuance date of this leter, you should provide information on all the deep draft pumps identified in your bul,letin response and describe the extent to which your, deep draft pump long term operability assurance program conforms to the various portions of these Guidelines.

Emphasis should be placed on (1) the establishment of installation procedures that are followed each time these pumps are disassembled and reinstalled, and (2) the testing requirements and bearing wear criteria.

The instrumentation called for in the Guidelines should not be considerea a

requirement.

These Guidelines establish an acceptable method of assuring long term operability of deep draft pumps.

They do not necessarily c'onstitute the only method for demonstrating long term operability.

The staff will review the information you submit to determine whether your long term operability assurance program for deep draft pumps is in sufficient conformance with these Guidelines to assure long term operability. If not, the staff will determine whether you have established and utilized other methods and procedures, preferably with the assistance of the pump manufacturer, that also demonstrate and assure that these pumps will perform their intended function for the length of time required.

8202030299 820ii2 PDR ADOCK 05000528 A

PDR DlALBk3 OFFICE)

P o

~ ~ ~ ~ ~ o ~ ~

~ ~ ~ ~ ~ ~

,NAEII JKerrigan:lb 1/ 2 /82 DATE

~ ~ ~ ~ I ~ o grI ~ o Io ~ ~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~

F

~ ~ ~

~

1

~ ~ 0 ~ ~ 0 ~ ~ 0 ~

lia

~ ~ 0 ~ ~ ~

~ ~

,/82

~ ~

~ ~

~ OSIER

~\\ ~i ~ ~i ~

RL esco

~ ~\\ ~ ~ ~i~ ~ ~ ~ ~ ~

/82

~ f~ ~ 0 ~ ~ ~ ~ 0 ~

~ ~

~ ~ ~ 0 ~ 0 ~ ~ ~ 0 ~ ~ 0 ~ ~I~ 0 ~ 0 ~

~ \\ ~ ~ ~

~ ~i ~ ~ ~ ~\\ ~ ~ ~ ~i~ Otto NRC FOAM 318 t10-80) NRCM 0240 OFFICIAL R ECO R D COPY USOPO: 1991~960

~

~

I

~ tf

~0 t

V ll VV

~

'5

~

P r

P

~ H V ~

tI,v 1

4 t

\\

It

~ s<t P

~ VV 1

Hr. E. E.

Van Brunt 2

~

JAN 12 1982 If you have any questions regarding this matter, please contact Janis Kerrigan the Licensing Prospect Manager on (301) 492-7239.

The application/reporting requirements contained in this letter have been approved by the Office of flanagement and Budget; OtS Approval No. 3150-0011.

Sincerely, Enclosure,:

Guidelines for Demonstration of Operability of Deep Draft Pumps cc w/enclosure:

See next page Robert L. Tedesco, Assistant Director'or Licensing Division of Licensing Office of nuclear Reactor Regulation OFFICEI SURNAME/

DATE$

NRC FORM 318 u040) NRCM 0240 OFF) ClAL R ECO R D COPY.

USGPO: 1981~5.060

4 0

J

,f I

~ =

Mr. E. E.

Van Brunt, Jr.

Vice President - Nuclear Projects Arizona Public Service Company P. 0.

Box 21666 Phoenix, Arizona 85036 PALO VERDE CC:

Arthur C. Gehr, Esq.

Snell 8 Wilmer, 3100 Valley Center Phoenix, Arizona 85073 Charles S. Pierson Assistant Attorney General 200 State Capitol 1700 West Washington Phoenix, Arizona 85007 I

Charles R. Kocher, Esq., Assistant Counsel James A. Beoletto, Esq.

Southern California Edison Company P. 0. Box 800

Rosemead, California 91770 Margaret Walker Deputy Director of Energy Programs Economic Planning and Development Office 1700 West Washington Phoenix, Arizona 85007 Mr. Rand L. Greenfield Assistant Attorney General Bataan Memorial Building Santa Fe, New Mexico 87503 Resident Inspector Palo Verde/NPS U.S. Nuclear Regulatory Commission P. 0.

Box 21324 Phoenix, Arizona 85001 Ms. Patricia Lee Hourihan 6413 S. 26th Street

Phoenix, Arizona 85040

0

0 ENCLOSURE GUIDELINES FOR DEMONSTRATION GF OPERABILITY OF DEEP DR~

PUHPS DI SCUSSION I.E. Bvlletin 79-15 dated July 1979, <dentified problems associated xith deep-draft pvmps found at operating facilities and near tern operating licensee facilities.

Deep draft pv-...ps,~hiich are also called vertical turbine pumps, are usvally 30 to GO feet in leng&

xith impellers located in casing boxls at the 3oxest elevation of.-" =

..he pvmp.

The motor (driver) is located at the highest pump elevation xith the discharge nozzle )ust below the no.or.

Bulletin 79-13 xas initiated because several nuclear poxer plant facil'iti s could not demonstrate operability of their pumps.

The pv ps xere experi enc ing.excess i ve v

< bra.ion and bearing xe ar.

rapid bearing xear svggested that

.h se pumps could not perform

.heir re"vire" fvncticns during or folloxing an acciden As a

resu1t of the staff's initial reviex of the resp"nses u IEB 79-I=.

szveral pl ants xere identi i ed as having potenti al prcbl ems

>4.h

.heir de p draft pumps.

These gvidelines are provided for these plan.s so that

.he licensee or appl'.cant involved may have a method acceptable to the staff for demonstrating the operability of deep-draft pumps.

4

~

I

~

~

0 DEEP DRAFT PPisP OPE..AT1NG CHARACTERISTICS In order to better understand the operating characteristics of 1

these pu~s, a rotor dynamics analyses xas perforred to ascer ain the response of the pump rotor under steady state operation.

The analyses considered journal'earing to shaft dynamic response at various eccentricities and fluid viscosities.

The model for

'the analysis depicted a typical deep draft pu~ utilized by the nuclear industry.

The analysis resulted in recorrendations for irproving tho stab/lity of the pump rotor from externally appl.ied inputs and by self-generated inputs.

The conclusions xhich xere derived from the analysis and staff evaluations o

North Arna, Beaver Valley and Sun'acilities xith similar pu,+s include:

1

~ )

Pu ~s xith this type of con igura:ion are prone to bearing xhirl v'braticn prcblew due.o the flexibility of the rotor and casinc s.rvc:ur This phenc~non is accentuated as jo mal "earing cle rance becomes large.

leads to bearing xe r (Journal bearings).

This pheno;.enon 1

lox Head Safety In' ion Purq Ro or Dynamic Analyses',

by

.Frankl(n Rese rch Center, Report FC-'.9S2, da.ed Yizy 1980.

0

2.)

There may be natural frequencfes assocfated wf th the pump assembly whfch occur near the operatfng speed of the pump.

P mp opera.(on xfll drive these freqvencfes and c~n cause bearfng wear.

The severity of this condf tfon fs dependent on bearing diametral clearance, rotor unbalance condf tfons and housing flexfbflfty.

As an example, if the wear fn column Journal bearfngs becomes sufffcfently large

( twice he original diametral clearance) so that hese beatfngs are no longer ac fve and 'he undam'ped crf:fcal frequency near the operatfng speed of the pump

's allo~

d to. expand, the additional'uncontrolled bearfng wear will occur.

7>>',s wear can contin>ue n.il the sha',t rubs agai rst the support structure of the bearino and can c e.".t',ally sever the sha 3.)

One acceptable>>e hod for correcting fnstabflf ties fn the pump sha;t fs to utilize a Jo.mal b= r n" exhfbits stable characteristics.

One such des',

n which desian fs the Taper land'bearfng'.

This design fs more st ble than the plafn Journal

bearing, fs less suscep.fble of Ne taper and xfll cause the bearfng to dynamfc fflm qu'ckly during s artup.

to wear because

>om a hydro-

~

~

4.)

St) ffen)ng of the column sect)ons of the pump is advantageous i f there is a column frequency near the operating spec'd of the pump.

The s'hifting of the colur.n.frequency to a higher level will el)mina e

coupl)ng between the pump operating speed and the

. colunn frequency.

Flow )nlet cond) tions to the pumps and scwp des)gns can be )riportant to pump operabil) ty.

Certa)n installations have demons t-,a ed flow character) sties which produced r

vortexing it the bellmouth of the pump.

This vortexing is due to sump design or sump supply line entrance con-ditions.

Th)s condition can contribute to additional pump vibrat)on and wear.

Flow s'rai"htener

devices, re".c.ion of bellm"uth diame.ers, and bottom clearance

-eductions have proven

.o be e, fec ive in el',m)natlng this problem.

6.)

This type of pump has exhibi.ed opera.ional problems due to des)gn and )nst llation def)c)encies.

The h)gh flexib)lity of the shaft and column;ake this design rather forg)v)nq when-it corn s to )nstal]at)on deficien-c)es such as m)salignmen betw n the sha and col~~n,

1

low-precfsfon coupl fng assemibl fes, and. non-perpendfcul ar rountf ng flanges.

This act however, can 1 ead to excess fve bearfng wear without sfgnfffcant noticeable change fn pump operatfng characterfstfcs.

To ensure proper pump opera

fon, proper aligrnent should be established between all mating sur',aces and m asures should be emphasfzed whfch prevent col~mr r

ard sha't eccen'rfcftfes.

These measures can fnclude optfcal.

alfgnm nt of the column segm nts, use of hfgh precision couplirgs and use of-accurate techniques

.o establfsh. tha.

the sump plumb line fs perpendicular to.tne pump mounting f1ange.

The above findings and conclusions hav contrfb<<ted signi =icantly to the Cove) o";.,er~ "f +"ese cuf"elfnes.

The cu i.e'.ines listed below are divided into ins.allatfon and +est r as, The sub'ects to be aCdress.d fn these areas are consfdered to be c, prime

'importance when es.ablishing a

pump operabfl) ty assurance program.

The ex-en to wh(ch each o', the

".ao are'as are implemerted a. z specific facilfty fs dependent on specfffc sy-, ptoms Ai fch have been fdentiffed;xf th.hese pumps while in operatfon and'during servfce perfods.

~

~

~ ~

Impl erne'nting the measures outl ined bel ox, at North Anna 1

h 2 in total, has *been shcxn to provide reasonabl e assurance that the pumps xil1 be operabl e xhen required for their safety function.

These guidel.ines are not intended to repl ace the requirements of P

Standard Revi ex Pl an 3.9.3, Regul a tory Gui de l. 68 or any other requirements presently enforced by the staff

Rather, the

~

~

guidelines are to be used

.as supplementary material for est.blishing I

deep-draft pump operability.

r GUIDELINES FOR OPER'VILITY I MSTALLATIOM 1.0 IMSTALLATIOh PROCEDUR S

Experience has shoxn that these pumps are prone to having operabil ity problems as a result of poor ins allation procedures.

Tho guidelines emphasize those areas of the installation procedure, xhich if implemented, could significantly improve the 1 ikel i hood of an oper'bl e pump.

The procedures ut(l '< -ed should be submi tted to the s off

> or reviex.

1 1

~ 4 P'J.'P

!hSTALLATIOH Determine by measurement that al 1 sha ft segments are straight xithi n tol erances specified by the manufacturer.

1 4

b.

Determine by measurement or provide certi fication that all coupl ings

( or shaft segments X pump to mo.or coupling) are of high precision as specified by the ranufacturer.

c.

De er, ine by measurement

.hat all pump segment flanges are perpendicular to

~1e centerl inc of.he

segment, that the segmen.s are straight and that'ny mating sur aces are Qncen ic to 41 Bs tabl ished datril~

~ 'n'here

>'ournal

7 bearing guides (SPIDERS) are used, establish con-centricity betxeen this asser3ly and its mating sur ace.

d.

Align full pump casing assembly optically to assure

~,

maxirUm. stra( ghtness and concentricity of the assembly.

Any equival n. r) thod is acceptable, as long as the procedure stresses column s raightness and concentricity.

e.

Assure pump to rotor flange perpendicularity and tha.

proper coupling installation is perforred-f.

Assure that all mating surface bolting is properly attached and that manuf acturer torqu ng seguences ar e adhered to.

SU.".?

INS

'L~ ATIOh Assure (xhere used) that su;"/pum>

rating glance is perpendicular to the sur~ purp line.

>r Assur that sump Cesi,"n pr v=n s fluid 'arcr, li s such as vortexirg.or turbulence near the intake

~

p

~ ~ <<<

)

~

~ ~ +L

~

'>'> '<<<<<<q<<

r

~ >J 1

~ >

~ >>h c>>o

> cb I

~ >> v>

~ >

>g y'>p>

~ >)

is not so designed as to allox luid conditions favorable to these ano;alies (i.e.,

sharp bends in piping prior to entrance. into su>~).

c.

Assure tha inter erence does not exist betxe n

the surp and any pu;>p appendage such as a seismic restraint.

2.0 Tes:ine Reeniressents a

The installation procedures are essertial in establishing pump operability.

In addition to careful installation, testing ray

.be required which vill verify proper operation of these purrps.

After completion of the installation checks, licensees or appli-

'ants should evaluate the need for further testing and report the. results of this evaluation together eith the details of any

\\

test plans to the staff.

Should tests be required, an acceptable

'test procedure should include the items listed.~olo~.

The staff 0

recognizes that the instrum ntation and procedures outlined below';

y be di, icult to i plemerst at all facilities and, therefore, the sta,f is emphasizing good installation practices which lead to

.operable c"ycnents.

If tes s demorstrating operability cannc

e."..c".,q as s all the iteias listed below, then alternative procedures s'hould be proposed for evaluation by the sta,f.

The tests should r 'I o ess~

~

> IL ~

I I"=

susl e'ee ~

oI pu.mp Cynamis c ch ra etc;i s

'cs and

~'e r da"?

• "iffere.;: stages of testing, culssina+ing ~ith an ex.rapolation of

~ I vc io i ~si v

I

~

'lJ I I I i a1 3 or tlhe I u 3 ~

2.I Tes.. Instru~enta:ion The, olloaing 'nstrvmentation should be incorporated into the

.es-.

procedure aside fron normal floe measurement, pressure'and vibration 'ns.-J-entati on:

9-a;)

X, Y proximity probes at three axial locations on the pump column, for measvring and recording radial positions of shaft xith respect to the column.

b.)

X, Y, accelerometers (at proximity probe locations) for measuring and recording radial accelerations of the col vmn.

c.)

Dynamic pressure transdvcers

.for measuring fluid pressure a. the, following locations:

1.

Bottom of Column (svction) 2.

Yiid-Column 3.

Top of Column.

d.)

Sha't Rotational speed and dynamic variation instrument.

2.2 PR

-T~ST DATA Kith the pvmp disassembled, measure all Journal bearing 0.0.'s, bearing I.D.'s and calculate bearing diametral clearances.

In addition w'ith pumps fvlly assembled and us ng

.he proximity ~rcbes, obtain the clearance circle" at each of he thr ax' stations by rollirg he shaft section

~3 hin the clearance volv...

of its bearincs and in this ~'ay, es.ahlish proper opera.ion of the probes.

~ ~

~

~

~

~, 2 3.1 PHASE 1 Testin (6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> plus start-stop)

R This phase of testing should be comprised of 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> of

'esting (Break-fn) fol 1 oxed by start-s top testing.

Test conditions should simulate as nearly as possible normal torqve required to turn shaft by hand).

Oata should be an'd accident conditions, Parameters to be considered are flox, temperature,

debris, and chemical composition of fluid being pumped.

Static torque 'ests should be-I performed before and after the test

('i.e. measure amount of aken during the six hour test at 1/2 hour in'vals.

A total of 12 start-stop tests'. xill,be performed con-sisting o

a sta" up from zero speed up

.o full-speed, 10-r..invte dwell at full-speed and a shutdo~~

from full speed to zero spec", xi:h recording of all instrvmen <:ion during full cycle cf star.-stop.

Upon comple.ion of Phase 1 :es ing,the '.ol'io~ing 4a:

s".ovid be obtained and recorded:

1.)

Ob ain the 'clearance circles" using the three sew of proximity probes..

~

~

Tests at Nor th 'nna 1

E 2 and "anu actur rs inpu indicates

hat 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> is an adequa.e time interval for bearing brea'k in'eriod..

0 2.)

I'.easure and record the following dir'ensions for each bearing:

t'.

)

Journal O.D.

b. )

Bearing I.D.

c.)

Bearing to Journal diametral clearance d.)

Establish Phase 1 test bearing wear.

THE ACCEPTANCE CRITERIA IS AS FOlLOMS:

'3.

I.) If wear is) 5 mils for any b ar>ng, wear is unacceptab1e-and test -should be terminated.

2.) If wear is/ 5 mils for all bearings

'.)

Reasse."hie the pump

b. )

Obt ai n "cl ea ra nce ci rcles" c.)

Reinstall pump in test loop.-

'.2;'P2 T <his acceptable wear value may be modi ied based on nu ac urers record ndation..

0 h

~

k

~

~

4

' and during. the shutdoxn period.

Data should also be recorded at 1-hour time intervals during the 48 hour5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> test.

The foiloxing>measurerents should be made at the cor>pletjon of Phase 2 of the test:

l.)

Obtain the "clearance circles using the three sets of proximi ty probes, 2.),".easure and rec"rd the folloxing dirZnsions for each boaring:

a.)

Journal O.D.

b.)

Bearing I.D.

c.)

Searing to Journal diametral clearance.

d.)

c.stablish accurulated bearing xe'r.

THE ACCT,PTNCE CRITERIA IS AS FOLLOMS:

If accuse.',ated bearing xear on any bear',ng is 3

P7 mils, x r is un ccetable anC tes. shculd be ter..i na.ed.

2.)

!, accur~lated xe r on all be rings isC7 mils for all bearings'.

a. )

Reasse. Sle puli

b. )

Obtain "clearance ci rcl es" c.)

Reinst ll purq in test loop.

0 5.) ~h3 T <<<

(F6 h

phase 3 testing is to be performed at full systen pressure and temperature and f1uid conditions simulating those expected during acc;Went and nor al operation.

The s?r.e 'procedures should be folloxed as in Phase 2 testing except that data nay be taken xi th less frequency.

The same measurements should be taken at the this phase as xith the o.her phases xith the criteria:

1.) If accumulated bearing xear is) 8 mfls xear is unacceptable and test should he 2.) if a.cci."..ul ated xear is (, 8 mils for all compl e.ion of folloxing acceptance for any bearing, ter ina.ed.

be? r ings, a

3 decision needs to be made o establish:

a.)

the need for ad9tional tes"..ng cr b.)

xhether or not the bearing xear xill be.

cc "tably 1 ox.

The recommended decision process is outlined belch.

P1ot the values of accumulated xear versus tine (H) for each bearing after Phase 2 and Phase 3 tests, namely.

Mear a

H2

~ 54 hour6.25e-4 days <br />0.015 hours <br />8.928571e-5 weeks <br />2.0547e-5 months <br /> Year at H3

~ 150 hours0.00174 days <br />0.0417 hours <br />2.480159e-4 weeks <br />5.7075e-5 months <br />

~

~

~ ~

(

~

V

>>14-Strafght 1 ines are then drawn through the plotted values of wear and ex ended to the right {See example Ffgure 1). If the extension

'ntercepts the maximum acceptable value of wear (8 mfls) at a value H less than the life goal for this pump, addi tfonal testing should be performed.

If the intercept of the line. with wear of 8 mfls exceeds, the. life goal for thfs pump, no addi tfonal

-testing fs requf red and bearfng wear fs acceptable.

If additional testing is deemed necessary it should be don fn a similar manner o that performed during Phase 3 xf th similar acceptance criteria and decfsfon process.

It fs expected that such addi tfonal tes f ng vill either show a stable pump operation xf th no increase fn k

bearing wear or increased bearing wear xfth unaccep hie results.

2.6 Evaluatfon of Pump Acce'ptabflfty If be ring wear {after all t sting phases) fs acceptably low {as per decision proc ss) and f; vfbratfon levels over

-he frequency s-e=trum of 3 cps to 5000 cps are acceptably low and show no unfavorable r nd of fncreasfng magnf ude during the +estfng, the.pump may Le judged accep.able fo fts fntended use.

~

~

I

~

~

~

~ 0

~ t

~ '

~

~

~

~

~

~

~ ~

~

~ ~

~ ~ ~

~

r ~

~

~ ~ ~

I If'

)

~

~

/

I I

.. I..

~

~

~

.: I I

~ I~

I

~t

~

~ ~as I

~ r ~

t

~

~

~ t ~

~

~

~

~

~

~ ~t'

~

~ ~t ~ ~t

~

~ ~ ~ ~

~ ~

~ ~ ~ ~

~

~ ~ I

~

~

~

~ I

~ ~

t at I

~ t

~ ~

~ ~ ~

I

~

~

l

~ ~ I o ~

~

~

~ ~

I

~ ~

~ t

')n

~

~

~ r

~

~

~

~ I

~ tJ,

~ ~

i I'/g}

I" I

l I

~ ~

~

ll

~

~ o

~ I-PJ o

~ ~

~

I

~

~

. ~

~ I ~.~

~ ~

~ p

~t t

~ ~ ~ ~ ~ ~

~ r ~ ~ I

~ ~ ~

~

to

~

~ ~

~ ~ ~ ~

~

~ ~

~

~ ~

I o

~

~

~ ~ ~

)

~ ~

~ I CJ I

~

~

~

~ ~

~

~

~

~ t

~

~ ~

~ a

~

.' I

~o' a

~

~

~ ~

~

~ ~ ~

~ ~

~ ~

~

Q

~

~ ~

~

~t

~

~

IIv..'

r ~,g I

~ I

~ ~

t ~

~ ~

~t ~ tt

~

~ or

~ ~

o

~

H

~ )

~ I ott ~

~

~ ~

g a

~ \\

~to[ at

~ \\ )

~ ~

ot

~

~ ~

~ ~ ~

~ ( ~t

~

~ ~ I t ~

~

~ 'tl

.. I.

~

~

I

~ ~

I

~

.'.).

Il

~

~

~ o

~

~

~

I

~ 1t

~ ~

~

~

~ ~ ~

~t ~

~

~

~

~ ~ ~ ~

~

~

I

~

/)hag 7~~0" t

~

1 t

~

I

~

~ (

~

~

~ ~

~

~ ~

..t...

~ ~ ~

~

I ~"

~ o ~ I

~ ~ o

~ ~ ~ ~

~

~ ~

~

~ to

~ ~ ~ ~

~ ~, ~

I l.

I I

I I

~

I

~

~

~

I ~

~ 4

~

ling'"

~ ~

I

~

i:I

~

I

~

~ ~

~

~I I

I

~ I 1

gL J

J l

..L

~tt 5

~

~

gl t

)

tot I

.I "I

~

~

~

~

~ I

~

~

~

~ ~

QA)..

~ ~

~

~

t at ~

Q~IC 1

~ ~

~

~

~

~

~ os

~

l

~ ~

~ ~tr I'

t ~ I

~ ~

~

~ ~

~ ~ 11

~ a ~

~ ~

~t ~

~

I I

P t!c II'!t(

~t I

I

~

~

~

~

~

at

~ I

~ rao 1~" t C

/)pvtorJ,

~

PVhc

-i' I I 7

~

~

~ ~

~

~ ~

~

~

~

4 v

p

~

~

~