AD-A174 652 FUEL SPRAY IGNITION BY HOT SURFACES AND STABILIZATIOd I/iOF AIRCRAFT FIRES (U) PURDUE UNIV LAFAYETTE INDTHERMAL SCIENCES AND PROPULSION CEN

UNCLASSIFIED A H LEFEBYRES ET AL 15 NOV 85 F/G 2V,2

EllllllEEEl1lllhhEEhIIIhEEEl~hEllllhEEhhhE|/hEEE|hE

|//|I|/|

LI11 L 6

I.'0

IIII 25 11111J.4 N~

MICROCOPY RESOLUTION TEST CHARTNATIONAL BUREAU OF STANDARDS- 1963-A

AD-A 174 652Ill N'(tJ^ ITAtI10N PAGE

Unclassified TI. . None _'_2a L. NA IL '.iFIC A (N1 ELECTE I 1 T1 .1-.1 D .LAS r 11 ''71 AFOSR

:, OrL :,r:,':OW =NOY 0 6 AP 'OVed rot public r*101136;NA 6 distribution ulimttd.

H I, A

IIIA I (t

LUS/AFOSR-COMB/, AFOSR-TR- 8 6 - 0 8 7 4

Put-due University ""....TSPC Air Force Office of Scientific Research

Thermal4Sciences and Propulsion Center Boiling Air Force BaseChaffee Hall/Purdue University Building 410West laf e Jte.y 47907 Washinton, D.C. 20332-

013f;ANI/Z rWN Air Force Office ' :of Scientific Research AFOS AFOSR 82-0107

Bolling Air Force Base . , - -Building 410 I . " ,Washington, D.C. 20332- ,.'__,.. . .. ... . . 2308q A2

Fy n Unclassified-11

A. i. Lefebvre, S N. 13, Murthy and ,I. G. Ski fstad

Annual Technical . 1/15/ 11/14/83! 19 oveber 15 50

None17 /

,Fire Void Space Ignition Combustion

7 i ' -- Jet Instability Fuel Sprays Blowoff Jeloci,

-~ In Task I, the perenttlresearch or-t-h4s--twe-yea--pe-riod"priimarily involvedrefinement of the experimental apparatus, -heinstrumentation, 4n4-themeasureienttechniques -ithe first year'and acquisition of experimental data. n- the second year,Special eforts were made to assure the reliability of the _measurements, including runsmade to examine the process of oxide formation whenjitili zing pure nickel surfaces.Experimental data were acqui red for both liqui d kerosine (Jet-A) and gaseous commercialpropane fuels over a broad range of run conditions. Evaluation of the results in thecontext of existing theories and modifications of the CONCHAS-SPRAY code to model thisexperimental system were also undertakeno the report period.

(Qn Task II, euuri+v] the-first-year)the extensive experimental results on blowoffvelocity, obtained using both conventional Vee-qutter and single-sided flameholders,provided the data base for an analytical study of the factors governing the stabilitycharacteristics of bluff-body flameholders. An equation was derived for predicting ,.

... , .I,, .. X X '.A, ... , ,4,1 . . . .Un(:lassi f ied

Julian H. Tishkoff -\ -1. 1,... ...(202) 767-4935 AFOSR/NA

OD FORM 1473, 33 -PF 8* Unclassified

IblowcFf velocity in terms of flameholder size, flameholder blockage, ambient airpr*3sure and temperature, and laminar flame speed Predictions of blowoff velocitybased on this equation showed excellent agreement with %Oexperimental values . T- -oobtained in this investigation and with the published results of other workers.The measurements of blowoff velocity obtained in the second year for both gaseousand liquid fuels, generally confirmed theoretical predictions in regard to thedependence of peak blowoff velocity on laminar flame speed. They also showed thatgas turbine fuels in the range from Jet A to diesel oil (DF2) exhibit very similarflameholding characteristics, since their laminar flame speeds are virtually thesame.

In Task III, during the first year of effort,.> experimental studies't-,e-been'/ uc,completed and results correlated for ventilation flow from surroundings into acavity with a small interal flow 'tring the second year of effort 4xtensiveflow visualization studiez n undertaken for flow past a protrusion,including the case of a jet through the protrusion. These studiesc~a*& provideddata on formation of vortices adjoining and over the protruberance and the natureof jet flow entrainment into them. Based on an analytical model for low speedflow past a protruberance, the flow field in the vicinity of the protrusion hasbeen computed While the flow fieldl downstream of the protrusion is predictedsatisfactorily, further development are needed in order to predict the upstreamflow field. The test rig for combustion studies is nearing completion.

Accesion ForNTIS CRA&IdDTIC TAB 5U;nanno:,~ced SJustification ..............................

By .................DI A ib ,tio:.

Avialitity Crdx-es

i Avaii :. d orDist po li7)Ap Ia)

:I

TASK I

-2-AFOSR.TR. 99-0874~

BIANNUAL REPORT

AFOSR-8 2-0 107

Period November 15, 1981 to November 14, 1983

TASK I

IGNITION OF FUEL SPRAYS BY HOT SURFACES

Approved fo 0,Public release;

School of Mechanical Engineering

Purdu3 University

W. Lafayette, IN 47906

AIR FCRCE 0FFT.CE G7 SCIFNTIFT C RESEARN1 (A.FSC)NOTI CE CF 'LRAN$5'41 71ALJ -k D TIThis techniotl report riS trc' n 1eviewad qnd Is

Chief, Technical informaltionl Division

_3-

TECHNICAL OBJECTIVE

Acquisition of experimental data for heated boundary-layer

flows of fuel spray/air mixtures under conditions leading to

ignition of the mixture; the data are to be of sufficient detail

and are to be taken over a sufficient range of flow and spray

parameters to enable the development of theoretical models for

ignition which can be used in aircraft fire hazard prediction.

APPROACH

An experimental facility has been developed in which each of

the pertinent parameters of the problem can be independently con-

trolled (boundary layer properties, mean droplet size, fraction

of fuel vaporized, wall conditions, etc.). Optical measurements

(LDV, scattering, image-type spray analysis) and sampling probe

measurements are made both upstream and downstream of the heated

length of duct serving as the test section to obtain the flow

properties in the boundary layer. Those data will serve as a

basis for the development of theoretical analyses of the

processes. Both iinition correlations and a more detailed boun-

dary layer analysis are to be developed in the investigation.

PROGRESS

In the first year of the period covered by this report, most

of the efforts were devoted to refining the measurement methods

and experimental procedures to obtain reliable local measurements

of the two-phase boundary Iayer flow properties to more closely

lI

-4-

relate the ignition phenomena to the relevant physical properties

of the flow. To that end, substantial efforts were devoted to

tailoring the flow profiles and to evaluation and refining of the

individual measurements involved. Runs made with the complete

system initially revealed several minor changes would be neces-

sary in the instrumenta- tion for reliable operation. For exam-

ple, it was found to be essential to install a controller for the

electrically-heated catalytic combustor in the probe/sampling

system to extend its operational lifetime for the range of

operating conditions needed and the run procedures adopted.

Because of the sensitivity of one of the sampling methods used

for determining the fraction of fuel vaporized, it was found to

be critical to assess the isokinetic sampling condition accu-

rately. A procedure employing LDA measuremnts was developed and

refined for that purpose. More accurate wall temperature meas-

urements were also introduced, and better uniformity of the wall

temperature was achieved by improving thermal contacts and by

monitorinq and controlling the sheath temperatures of the electr-

ira1 heater elements.

Runs were made over a broad range of parameters to explore

the nature of the data and the ignition process. The boundary

layer was determined to be composed of a vapor/fuel mixture for

most of the runs made under conditions leadina to ignition, cov-

ering a range of bulk equivalence ratios from 0.2 to 2.5. Aver-

age droplet sizes for those runs were varied from below 10

microns to about 50 microns. Significant transport of fuel

jjj'I

-5-

toward the wall was observed, presumably due mostly to droplet

motion. The observed wall tempera- ture at ignition was found to

exhibit the usual minimum for bulk flow condi- tions with

equivalence ratios near unity. Runs were made with both pure

nickel and stainless steel surfaces. Those made with pure nickel

exhibited rather erratic behavior due to the formation of an sub-

stantial oxide layer, primarily under fuel-rich operating condi-

tions.

Two papers relating to the small injectors and the spray

measurement system were presented at the ICLASS '82 meeting in

Madison, WI, in the summer of 1982. One M.S. Thesis dealing with

the LDA measurements was also completed.

In the second year of the period covered by this report,

most of the experimental research was concerned with the acquisi-

tion of data for kerosine (Jet--A) fuel sprays and for a gaseous

fuel (commercial propane). Both nickel and stainless steel sur-

faces were employed for those runs. Effects of free- stream

velocity, boundary layer profile, bulk equivalence ratio, and

inlet air temperature were examined. A number of supplemental

tests were made to check what appeared to be anomolous data under

-.ome run conditions, or to examine certain conditions which might

have led to facility--related results. Most of the questions

raised were resolved satisfactorily, thereby improving the relia-

bility of the information acquired and refininq the operation of

the facility.

AA&..

-6

Further insight was acquired as to the problems encountered

with the pure nickel surface, including a chemical analysis of

the oxide layer and direct observations of the nature of the

ignition in the vicinity of a flaking oxide layer. It was judged

to be virtually impossible to obtain data represen- tative of

strictly clean nickel surfaces under fuel-rich operating condi-

tions. Such measurements would probably have to be made under

more rapid heatinq rates than were possible with the existing

apparatus. The details of the role played by the oxide layer in

the ignition process, apart from its role as a thermal barrier,

remained unclear.

The runs made with propane gas were intended to simulate a

situation corresponding to a bulk mixture of kerosine vapor and

air. Those runs showed the important influence of the initial

boundary layer profile on the wall temp- erature at ignition,

vielding lower wall temperatures at ignition for a thicker ini-

tial boundary layer. Runs were made over the range of flammabil-

ity limits for the mixture.

Along with the experimental investigations conducted in this

research, efforts were made to compare the results with existing

theories, such as that of T.Y. Toong and others, and to initiate

necessary modifications of Lhe CONCHAS-SPRAY code to render it

useful in describing the boundary layer flows of concern here.

-7

TECHNICAL SIGNIFICANCE/RELEVANCE

Means for predicting aircraft fire hazards require relaible

theoretical models for the conditions leading to ignition of

fuel/air mixtures of many types and under a broad range of condi-

tions. The necessary data on which to base such theories must be

acquired in experimental research programs of the type described

here. The development of the imaging-type spray analysis system

for measuring local properties of sprays (number density, size

distribution function) should find applications in a range of

industrial processes involving sprays, as well as for fuel

preparation in gas turbines and other engines. The method has

been developed partly under the AFOSR program and partly under

NASA nponsorship. The miniature airblast injectors developed for

this program have been found to be exceptionally efficient (a

ratio of gas to liquid flow rates of 0.02 as compared to 2.0 for

more conventional injectors). Thesie could find applications both

for fuel preparation and in ertlain industrial processes where

ias/[iquid in-ection is acceptiblfe and where the efficiency is

important (possibly even in roc iproca- ting engines).

EXPRESSIONS ON INTEREST

A number of researchers have inquired as to the methods

employed in this investigation, particularly those related to the

use of the microscopic airblast atomizers.

PUBI, I CAT I ONS

% 0

- 8 -

1. Skifstad, J.G., "An Automated Imaging-Type Spray Analysis

System for Local Spray Properties" presented at the ICLASS-

82 meeting in Madison, WI, 1982.

2. Skifstad, J.G., "Microscopic Airblast Atomizers" presented

at the ICLASS-82 meetinq in Madison, WI, 1982.

3. Skifstad, J.G., "Representation of Functions by Gaussian

Series", paper submitted for publication.

4. Sacksteder, K.R., "A Laser Doppler Anemometer for Measure-

ments in Fuel Sprays," M.S. Thesis, School of Mechanical

Engineering, Purdue University, 1982.

TASK IH

- 10-

BIANNUAL REPORT

AFOSR-82-0107

Period November 15, 1981 to November 14, 1983

TASK II

STABILIZATION OF AIRCRAFT FIRES

Principal Investigator:

A.H. Lefebvre

Thermal Sciences and Propulsion Center

School of Mechanical Engineering

Purdue Univer" ity

W. Lafavette, IN 47907

* ,'

TECHNICAL OBJECTIVES

One objective of the research was to extend the range of

experimental data on the stabilization properties of bluff-body

flameholders to include flameholders of large size (characteris-

tic dimension up to 10 cm) and irregular shape, such as might

arise on the external surface of an aircraft due to structural

damage. Another goal was to derive suitable theoretical rela-

tionships for blowoff velocity for the extended range of flame-

holder sizes and shapes.

APPROACH

The method used to determine the blowoff velocity and other

stability characteristics of bluff-bociy flameholders was the

well-established water in-iection technique, as illustrated

schematically in Fig. 1. With this- ochnique the flameholder

under test is placed near th, exit of a duct supplied with an

airflow containing a water/fuel mixture. Both the water and fuel

(usually Jet-A) are fully vaporized by the time they reach the

flameholder. At the start of a run the equivalence ratio is set

and the flame 1s established with no water injection. The water

flow i. then initiated and increased until flame extinction

S cur s . A plot of the stability loop so obtained (equivalence

ratio versus water/fuel ratio) represents, in effect, a plot of

e:fuivalince ratio versus the reciprocal of pressure. The calcu-

latem relationship between water/fuel ratio and the equivalent

reduct ion in qa'; pressure shows for example, that the injection

- '~ ~%

- 12-

LLJLLLi

FC1)

uii _Ij

Q- CL

H i4oi 4fl

000 000

1

LL- LuU-4

LU

*c:3

- 13 -

of equal weights of water and fuel is equivalent to halving the

pressure. The method has two advantages: it is the only tech-

nique which allows the entire stability loop to be obtained for

large flameholders, and any subatmospheric pressure can be simu-

lated while using fan air at atmospheric pressure.

Stability loops obtained by the method outlined above can be

analyzed to determine the effects of flameholder size, shape, and

blockage on blowoff velocity. The separate influences on blowoff

velocity of variations in approach stream velocity, ambient gas

pressure (altitude) and fuel chemistry, can also be assessed.

PROGRESS

In the first year of the period covered by this report, con-

siderable progress was made, both in the acquisition of extensive

experimental data of high quality, and in the application of

these data to the development of the following new equation for

blowoff 'elocity.

U BO/SL Cs(l - B a ) Re Pr (1)

where U = blow-off velocity

SL = laminar flame speed

C = flameholder shape factor (=Ba /B )

B a = aerodynamic blockage of flameholder

Bg = qeometric blockaqe of flameholder

Pr = Prandtl number

Re = Reynolds number = (SLDcPo/A

Dc = characteristic dimension of flameholder

dC

- 14 -

An alternative form of Eq. (1), which serves to demonstrate

that blowoff velocity is proportional to the square of laminar

flame speed, is the following.

U = C (I - B ) (D S2 /a) (2)BO 8 a c L 0

Equations (1) and (2) were found to predict not only the

experimental values of blowoff velocity obtained in this AFOSR

research program, but also the results obtained by other workers.

This very good agreement is demonstrated in Fig. 2.

Thus at the conclusion of the first year period it was con-

sidered that one of the primary goals, namely, that of developing

an equation for blowoff velocity suitable for large-scale flame-

holders, had been fully met.

During the early part of the second year of the reporting

period, attention was focussed on the influence of fuel type on

flame stability. From inspection of Eqs. (1) and (2) it is clear

that the only way in which a change in fuel type can affect blow-

off velocity is via the laminar flame speed, S However, most

hydrocarbon fuel-air mixtures tend to have very similar values of

S L usually in the range between 0.35 and 0.43 m/s. Thus, at the

outset of the investigation it was anticipated that only slight

variations would be observed in the stability limits for the pro-

posed test fuels, namely gasoline (JP 4), kerosine (Jet A) and

diesel oil (DF2), which had been chosen to represent the range of

fuel types likely to be encountered in aircraft jet engines in

the forseeable future.

- 15-

00 cOD tL LOO (, nO

~rO r C)J qo

LC) t o t

-0-00- 0 E

C) 0

C\J U

U <o 4 mw w 0 0

uLJQ

oo iD

0

CL.CL

0

00i

(G]3flSV3.N)0/ 00 9 0 i O T AJ-

O 1

- 16 -

The results of this research were presented as Paper No.

83-1327 at the AIAA/ASME/SAE 19th Joint Propulsion Conference,

held during June 1983, in Seattle, Washington. Some typical

results for Jet A, propane, and hydrogen are shown in Fig. 3. It

is clear from this figure that the higher flame speed exhibited

by hydrogen leads to a considerable expansion of the stability

loop.

It was found that the stability loops for gasoline, kerosine

and diesel oil are practically the same. The slight observed

differences between these fuels is attributed to differences in

their latent and sensible heat requirements, which cause the

vapor-air mixture for asolirne to be higher than that of kerosine

which, in turn, is higher than that of diesel oil. The practical

significance of this result. is that the threat to aircraft safety

from external fires stabilized either by structural protrusions

or reglions of separated flow on the a]rframe, is virtually the

same for a relatively low volatility fuel, such as diesel oil

(PF2) as it is for a fuel of hiaher volatility, !such as JP 4.

During the latter half of the second year of the reporting

period, work started on the measurements of stability charac-

teristics, notably blowoff vplocity, for flameholders of irregu-

lar shape, as illustrated in Fig. 4. Furthermore, detailed meas-

urement:; of static pressure were carried out both in an around

the flameholder region, in order to establish relationships

between flameholder pressure drop, flameholder size, shape, aero-

dynamic (Iraq, and blowoff velocity.

- 17 -

6 0 00CNE TO=573 K o JET A1.6- B9=0. 175 U0 =65 rn/s u~7PROPANE

0 HYDROGEN

1.4-

1.2

e-1.0-

0.8-v

0.6-0

0.4 1 2 3 4 5 6

WATER/FUEL MASS RATIO

I' ;u' 3. Corn p-It1 it or' c Jot A\, propan,

kkI h'- I1 r9 i

- 18 -

Ist Baseline

2 d Baseline

Pirjr';4. Vdrn~~)~i~r A ~ l~ein irrocqular-shapoI'j (j i ir ,..

- 19 -

The results of this study were incomplete at the end of the

reporting period and it was decided to continue these investiga-

tions during 1984.

REFERENCES

i. Haddock, G.W., Flame-Blowoff Studies of Cylindrical Flame

Holders in Channeled Flow, Progress Report 3-24, May 1951,

Jet Propulsion Laboratory, California Institute of Technol-

ogy, Pasadena, California.

2. De Zubay, E.A., Characteristics of Disk-Controlled Flame,

Aero Digest, Vol. 61, No. 1, pp. 54-56; 102-104, July 1950.

3. Longwell, J.P., Flame Stabilization by Bluff Bodies and Tur-

bulent Flames in Ducts, Fourth Symposium (International) on

Combustion, The Williams and Wilkins Company, Baltimore,

Maryland, pp. 90-97, 1953.

4. Zukowski, E.E. and Marble, F.E., Flame Stability from

Bluff-Body Flameholders, Eighth Symposium (International) on

Combustion, pp. 933-943, 1961.

TECHNICAL SIGNIFICANCE

The development of an equation based on sound scientific

arguments for the prediction of the blowoff velocity of bluff-

body flameholders in terms of all the relevant aeometric, aero-

dynamic, and fuel variables, in considered to be a significant

achievement.

RU

-20 -

Of considerable practical importance is the result obtained

by a combination of theory and experiment which showed that the

threat to aircraft safety from external fires, stabilized either

by structural protrusions or regions of separated flow on the

airframe, is virtually the same for all hydrocarbon fuels of

present or forseeable interest to the U.S. Air Force.

PUBLICATIONS

The references listed below represent the total number of

publications resulting from this program since its inception.

1. Ballal, D.R. and Lefebvre, A.H., "Weak Extinction Limits of

Turbulent Flowing Mixtures," Trans. ASME, J. Eng. Power,

Vol. 101, No. 3, pp. 341-348, 1979.

2. Ballal, D.R. and Lefebvre, A.H., "Weak Extinction Limits of

Turbulent Heterogeneous Fuel/Air Mixtures," Trans. ASME. J.

Eng. Power, Vol. 102, No. 2, pp. 416-421, 1980.

3. Ballal, D.R. and Lefebvre, A.H., "Some Fundamental Aspects

of Flame Stabilization," Fifth International Symposium on

Airbreathing Engines, pp. 48/1-8, 1981.

4. Rao, K.V.L. and Lefebvre, A.H., "Flame Blowoff Studips Using

Larae-Scale Flameholders," Trans. ASME J. Eng. Power, Vol.

104, No. 4, pp. 853-857, 1982.

5. Rizk, N.K. and Lefebvre, A.H., "Influence of Laminar Flame

Speed on the Blowoff Velocity of Bluff-Body Stabilized

- 21 -

Flames," paper presented at the AIAA/ASME/SAE 19th Joint

Propulsion Conference, Seattle, Washington, June 27-29,

1983. This paper has since been accepted for publication in

the AIAA Journal.

PERSONNEL

During the two-year reporting period the key personnel

employed on this program, in addition to the principal investiga-

tor (A.H. Lefebvre) were visitinn scholars Dr. K.V.L. Rao and Dr.

N.K. Rizk, and graduate student R.M. Stwalley.

AWARDS

The paper "Flame Blowoff Studies Using Large-Scale Flame-

holders" received both the 1982 ASME Combustion and Fuels Award

and the 1984 ASME Gas Turbine Award.

-22-

TASK II I

I-- 2 3 -

TASK I- --SUMMARY

1. Reportin3_. Peri od: 11-15-81 to 11-14-83

2. Startin Date: 11-15-81

3. Objective: The objective is the determination of flame structure, propagation

and quenching characteristics in a so-called void space with (a) small air flow,

(b) low pressure fuel injection through a protrusion, (c) small ventilation, and(d) gravitational action normal to both air flow and fuel injection.

4. Appilication: The research project is relevant to (a) aircraft void space fires

and thus aircraft fire safety, (b) gas turbine combustor design with fuel injectors

protruding into the chamber, and (c) ramjet combustor configurations with "blockage".

Fundamental data pertaining to (a) a jet through a protrusion in cross-flow and (b)

flame stability when the jet consists of fuel and there is an ignition source will

be the principal contribution.

5. Status: Analytical-computational studies have been continued on jet flow throu,-

a protrusion in cross-flow when the jet and cross-flow velocities are smuall and of

the same order. Experimental studies have been completed on (a) ventilation flow

into a cavity with small internal flow and (b) visualization of flow in the presenu

of a protrusion, including the case of jet through the protrusion. The test rig fa)

combustion studies with gaseous fuel injection has been set up.

6. Publication: A paper is under preparation for the AIAA Propulsion Conference,

Juen 1984

7. Interactions: In view of the direct-implications to the group dealing withfires at the Wright-Patterson AFB, a seminar-type presentation has been given there

on i six-monthly basis. Discussions have also been held with personnel at the MBS,the Boeing Military Airplane Co. and the Sandia National Laboratories, Livermore,

Ca. Secondly, in view of the implications of the research to the development of

gas turbine and ramjet combustors, discussions have been held with personnel of the

General Electric Co. and the Naval Weapons Center.

t- ' ., . ' < " ' " ' '% "". '1'''' .''.'' ' ''.',. .''.'\'L .' . .''.*.,.*,','', , -..N

rrt t

*i' ve 'o. I' v I V

I ', t, lid .

*1 .j

*t J, I ,

I~~ ,I v.''

LL MI

4..d

- 25 -

(2) with the free stream flow is also shown in the figure. The speed and

pressure of airflow can be adjusted in (1) arid (2); the velocity can be

varied in the range of 10-25 raps in (I) arid 20-100 raps in (2), and the pressure

difference between (1) and (2), 0.5-1.2 psi. Cavity (1) is the actual test

section and (2) represents the "surrounding" or "adjoining external" flow.

Figure 2.1 also indicates the gravitational action direction, noting that

gravity is a consideration in low speed combusting flows.

The test rig has been utilized for (a) flow visualization and (b) mea-

surement of overall flow quantities.

In cavity (1), the test section, a variety of protrusions can be incor-

porated as well as various shapes of walls for attaching the protrusions.

These are illustrated in Fig. 2.2.

By mixing smoke into the low speed jet flow, one can study the jet mixing

and spreading characteristics.

2.2. Test Rig for Combustion Studie~s

The test rig is essentially the same as illustrated in Fig. 2.1. Modi-

fications have been made by (a) replacing (1) and (2) with stainless steel and

steel bodies, respectively, (b) providing methane gas injection into the test

section (1), (c) introducing an ignitor, and (d) locating quartz windows for

optical observation.

Other instrumentation and diagnostics are being examined.

The ignitor design is shown in Fig. 2.3. It can be observed that the

source of ignition can be moved both radially and circumferentially in relation

to the jet through the protrusion and the gravitational direction.

3. Ventilation Flow

Figure 3.1 provides (a) the parameters that were varied during the tests

for establishing ventilating flow into region (1) from the external stream (2);

and (b) the broad conclusions from the study.

Figure 3.2 shows a correlation of flow into (1) utilizing the pressure

difference and flow velocity difference as parameters. The mass flux Ai of airflowing into cavity (1), through a vent of area of cross-section A, when the

density, dynamic head and pressure difference are p, q l and AP1 2 , is shown as

a function of the parameter I.

- 26 -

4. Diagnostics of Flow Past a Protrusion with a Jet

Utilizing several visualization and photographic techniques, the nature

of the flow generated in the vicinity of protrusions has been determined.

Typical details of flow are shown in Figs. 4.1 and 4.2.

The flow field in the jet is basically time-dependent in relation to the

internal structure. The frequency of such structures has been obtained

utilizing a stroboscopic light source. Some typical data have been provided

in Fig. 4.3.

The type of data presented in Figs. 4.1-4.3 illustrate the manner in

which mixing and jet spread can arise (a) upstream, (b) over and (c) downstream

of the three-dimensional protrusion.

5. M.delling and Prediction

The flow past a protrusion with a jet issuing through it is being modelled

in several steps as follows. In all cases, incompressible, viscous flows have

been considered.

(i) Flow past a protrusion when the cavity flow and protrusion are

two-d imensi ona I.(ii) The foregoing case with a two-dimensional (slit-type) jet issuing

through the protrusion.

(iii) Flow past a cylindrical protrusion with a jet issuing out of it,

this case being fully three-dimensional.

5.1. Computational Scheme

A source program, commonly referred to as the PHOENIX, developed by CHAM,Ltd. under the guidance of D.B. Spalding, has been acquired under the Project

and utilized with appropriate modifications.

The main cavity flow is air considered as a perfect gas, low speed,

laminar flow. The jet may consist of any gas; air, carbon-dioxide and methanehave been considered. The ratio of main flow velocity to jet velocity is

either one or three.

The problem has been formulated as i parabolic one with given initial flow

conditions for the cavity and the jet flows.

Although the CYBER 205 is fully operational now at Purdue Univesrity, allof the results presented here have been obtained utilizing the front machine,

namely the CDC 6600. The total nuimier of nodes utilized is between 5,000 and

8,000 in various cases.

- 27 -

5.3. Results

Figures 5.1 and 5.2 present the cavity and protrusion geometry and the

computational domain in the two-dimensional and in the three-dimensional

cases, respectively. The initial conditions for the main cavity flow are

uniform streamwise velocity and uniform pressure. Similarly, the initial

conditions for the jet flow are unifo-m normal velocity and uniform pressure.

It is assumed that the jet exit pressure is the same as the cavity main flow

entry pressure, although the main cavity flow may suffer a slight pressure

loss between entry and the location of the protrusion.

Table 5.1 presents the initial conditions for the various calculations.

Figures 5.3-5.11 present the computer outputs for various cases as

uescribed in Table 5.2.

6. Current Activity and Plans

(a) The test rig for combustion studies is being completed.

(b) The test rig for cold flow studies is being utilized for the balance

of flow visualization studies, which are expected to be completed by March,

1984.

(c) The ignitor assembly is being manufactured. Methane combustion studies

are expected to be undertaken March to September, 1984, principally for obtaining

data on relation of ignitor location and injection and flow parameters to flame

stability.

(d) The three-dimensional flow prediction scheme will be further developed

during 1983-84.

(e) Instrumentation for detailed analysis of combustion processes will be

developed during 1983-84.

- 28 -

LIST OF TABLES

5.1 Initial Conditions for Jet and Cavity Flows(Note: The pressure loss in the free-stream has beenneglected when jet is turned on.)

- 29 -

Velocity Pressure Protrusion Height Re

(m/sec) (atm) (Re Wh/v)

Uniform:freestream 30.0 1.0 252.8

Uni form:jet 30.0 1.0 252.8

Table 5.1. Initial Conditions for Jet and Upstream Flows.(Note: The pressure loss in the free-stream hasbeen neglected when jet is turned on.)

q . -

- 30 -

LIST OF FIGURES

1.1 Test Articles: Damaged Fuel Lines

2.1 Void Space Simulator Tunnel

2.2 Jet Through a Protruberance in a Cross Flow

2.3 Ignitor Locator

3.1 Flow Through Air Vent: Parameters

3.2 Correlations of Flow through Vent

4.1 Flow Visualization

4.2 Flow Patterns Extracted from Flow Visualization

4.3 Jet Interactions

5.1 Side View (a) and Frontal View (b) of Cavity

5.2 Geometry of Protrusion with Jet

5.3 Contour Plot of Streamwise Velocity Component for 2-D Flowwithout Jet: Note that the number 1 represents the lowest fieldvalue and 9, the highest field value.

5.4 Contour Plot of Normal VeloLity Component for 2-D Flow without Jet.

5.5 Contour Plot of Pressure Field for 2-0 Flow without Jet.

5.6 Contour Plot of Streamwise Velocity Component for 2-0 Flow withJet of 10 m/sec.

5.7 Contour Plot of Normal Velocity Component for 2-D Flow with Jetof 10 m/sec.

5.8 Contour Plot of Pressure Field for 2-D Flow with Jet of 10 m/sec.

5.9 Contour Plot of Streamwise Velocity Component for 2-D Flow withJet of 30 m/sec.

5.10 Contour Plot of Normal Velocity Component for 2-D Flow with Jetof 30 m/sec.

5.11 Contour Plot of Pressure Field for 2-D Flow with Jet of 30 m/sec.

R 11 1 l II I

- 31 -

AIR FILLED PIPE WATER FILLED PIPE

AVTAG FILLFD PIPE

DAMAGE TO PIPES (38mm DIA) BY AN INCENDIARY ROUNn

, ' 4 . "" 5 -

.2 p.- :," . J- J, r2""

REAR PLATE BEHIND AIR FILLED PIPE REAR PLATE BEHIND WATER FILLED PIPE

FIG. 1.1 TEST ARTICLES: DAMAGED FUEL LINES

'].~

- 32 -

0 7

'.[ ® "i ~i 1 I®

1. BLOWER *TEST SECTION: 15XIOX60 cms2. SETTLING CHAMBER FREE AIR FLOW VELOCITY:

3. FLOW DIVIDER SECTION 20 -100 rn/s

4. FLOW STRAIGHTENER SECTION

5. TEST SECTION: VOID SPACE SIMULATOR6. FLOW DISCHARGE SECTION

7 EXHAUST

FG. 2.1 VOID SPACE SIMULATOR TUNNEL

-33 -

JET THROUGH A PROTRUBERANCEIN A CROSS FLOW

* PARAMETERS

Test V DSection

tc) d)

Sec-o A-A

R= Wall Surface jet FlowSeto -

Curvature

Jet in a Crossf low Through a Protruberance

PARAMETER RANGE

Dp 0.5 in.

hi 0.125-Q.5 in.

Dj/D0.125-0.75 in.

R co + and -

V 1 10-50 fps

ReL. Itft 50- 250 X10 3

F IG. 2.2 JET THROUGH A PROTRUBERANCE IN ACROSS FLOW

-34-

VODTET1/4" 5 /32"

3/4__ AI 8 @ INSERT TOt FUELPASS IGNITOR LEADS

3 S6 I/THROUGH

1/4 -AIR VENT

- ~~~AIR FLOW .-----

FUEL INJECTION

PARAMETER:* LOCATION OF IGNITION SOURCE

0 RADIUS* ANGLE RELATIVE TO 'go DIRECTION*PROTRUSION RELATIVE TO WALL

FIG. 2.3 IGNITOR LOCATOR

* .- ~* ~** ~ ~ %

- 35 -

FLOW THROUGH AIR VENT*PARAMETERS

© K Wallv -- Tfhickness

( VOID SPACE FLOW

*REGIMES WITH AP012 >O

o 12 =P -P2 t A P12 'P- P2I. VI > V(AP 12 /2p) and > V2

I,

2. V, < and > V2

3. VI < "and < V2

* CONCLUSIONSI. VORTEX FORMATION IN CASE 2

2. FLOW BIDIRECTIONAL IN CASE 23. FLOW NONSTEADY IN CASES 2 & 34. FLOW REDUCED IN CASE 3

COMPARED TO CASE I

-BASED ON CONSTANT D & D/t

FIG. 3.1 FLOW THROUGH AIR VENT: PARAMETERS

' " .. .- , , , . . ., .4' . ' ." ',< ' " '.'-' .; ' . " .. .. ., -"; , .-.-"..::

- 36 -

FLOW THROUGH AIR VENT (CONTINUED)*RANGES OF PARAMETERS

D,-.,O.25 -0.50 in.AP12 ''.5-1.50 psi

Vl, V2 '- 3 0 -75 fpm

Bi:0.7- v

10.6- Vl< V2- vIv 2

0.5-

0.4-

0.3-

0.2-

0.t-

-,0 1, ,1 2 3 10 20 30

I -[I + AP12 /q] ; II rn/[A PI' 112]

FIG. 3.2 CORRELATIONS OF FLOW THROUGH VENT

_.,',f v{~;t~ ~ * * ~\

- 37 -

FLOW VISUALIZATION" FOG JUICE SMOKE* WOOL TUFTSeCARBON-CRISCO PAINT

FIG. 4.1 FLOW VISUALIZATION

-38-

*EXTRACTED FROM FLOW VISUALIZATION

FIG. 4.2 FLOW PATTERNS EXTRACTED FROM FLOWVISUALIZATION

I 1111,1,11,11 -j

39-

JET INTERACTIONSDj/Dp h

*CIRCULAR CYLINDER 0.250 0.125 300.500 20

0.750 0.125 40

0.500 20

SCUBOID 0.250 0.125 250.500 20

0.750 0125 45

4t

*CLOSED STREAMSURFACES DO NOT FORM

*TOTAL NUMBER OF SEP. & ATTACH.

"TOTAL NUMBER OF NODES AND SADDLES

FIG. 4.3 JET INTERACTIONS

-40 -

Y, v

P60 cm

30(m/sec)5c

1.5 cm

M TI.5m Z.w(a) Side View

y, V

.15 cm15 cm

1.5 cm

(b) Frontal View

Figure 5.1. Side View (a) and Frontal View (b) ofCavity Geometry

- 41 -

V jt= 30(m/sec)

11.5 cm

1.5 cm- 1

Figure 5.2. Geometry of Protrusion with Jet

42 ' . ..-- -- -- -- -- -- -- -- -- - -- -- - - .-- - - -- - - - - -- - - - - -

I III r c

11 /i I cI :I jI r. c

1 III 1 1 .CI - III S c

I .ju, Il C ¢ir! A! I CcI C It ci

SU III C ciI E ( li S vII ' K ILI C- Ia .. t0 ) ul t Ci

C90

IXE. l .

It c

w . 0 X /it c I ,cc C P ~ €00 XI E Ito c I|

0 0Ca Ilj CciIll c ci,0 cn ."a o1it/ -

w E .. - (0 iO (; cT

T; & >o . It.o oI c i r1 1 q c I

ri ill q cfLL-III ii LL -5a r

T 0LII 10iLI S.lii , C I

Li .C

it.AAAArAC.Af F F&,,, rAC..46h4AC.CAh'A,a. dlii r C Iq

I 4C¢, .gt.4Ac .A.AC. AC- CC.V..C1.AC isc1 1

iCr.A4A, cA ¢.A4 AA.A, 16. AA 'crv.C."j..C.

TA.AAAAAAAA.AA.C.AC.,ACIs6 J4J I

JAAAAAAAC4,ACA4A(,.A I q

i~t~,~r~.C. A i6 ;S [

r.,"A , ,A I . 6

AAAAAACA . I .C

IAAC.,AA.A ACr.~ (({,i 11v I. F .€ lllIlI~ tl

A .A r. A C5A C. I TVV

If, r r.C. rc e, 1I4C v

i AAAA . I,r A Ac r. Cc 1.C,

Y r A,r.rA C. . . (

1, fC.A A eIt. ICC

T r. . . .A

6rcAc a~. AAA

rt r. A A AA' , r, (, AC I,A .A r, nG rC. e, A

TAc ,r, A. .r. , A cc.q ,f, ,.C. A, r. r A A C. A 6 t

,.,' A, ,tA , * CITT t TTTITT II I II

S, 'TU',£ ',,VI Tl T T TI{t I1 9rrcc.cccc A 1 Arr: f i I I I T IT ItIT'Ifl It I

I .... ..A T .I. i .t.C .;% ix I I f I ll t llt I

i'I,.r .C cIcC. ,' ' I. II V%%' TtIIIIrTirti iiv lTti liiltl II C.A . I..(¢. ..,. #.....AeC. , . . -,. I *. I Ir. V I T I I tillIi I It i fll T I

I for.; tr.r.cr fi. ,,..,. ., T L.r. ,c ., C. Ii . ', . , I 1 v t I lilt f IT ill IIII .r ' C '.'.(,A , .. . t..6. AC d'. . i C. I7t . 1 V T T ; t I tt f I IV f(II I

IC C" (.,AC..(. ( ,, .~. .... ,.C.C.C.r,., .'A'tl icc,,,,,..i i '. c. , Evi iit iI it ITT!! iy

T ... . ,. I I I T I I f I I I I T I I I

S .b -I I , . / I -: . , .r T I' C¢t¢. IIVTCIVCIVV TC

I I '' I I I ,,I,, ( Vi' Ii , /;. -. . %CS2q- *

/ II' *. ,t ,,. cs.';CC2 * rl,'t'% 't% -l.;. % .1 I ' p.I '/ . . C ' ,;,;:c'CCC .; :,; C ~ lt(:(; , ;+

I flI I u,'!I * .,, ., C s:Ac:C 'lts,;r";,.ACCC, r;rtCCCCCC;,,. S

-----------------------------_. .....---.- .. "-.-------'--------------------. ..

*It. Agi- tiiei Ii i l~,

I arl I¢ :! u ol t ui'tiwtl"I; * " lI I ' 1P4W hl- v

I ,C9 1 c0 1 I I lfl')I1

hl=~-n 1VD1 ,- , .. . .i:...

- 43 -

t r:

44------------------44-

----- ----- ----- ----- ----- -------------.

i-i , ririri; .t-r .- c. . . . .-- e iir.rsc~~~ii~~

T- 5~ I.:, 1~i 1. C. C I r.iit1.ic I

C. C . 4ft- C.-r~ .C.rrr '

.rc V

I ~ ~ c i.c :c--. c cc.i.c .i, ;(- ictir.i.v :7cC V c-i- vvj c(cC;rt --- i. c-:iic~-rr(ccc-

T ifi.lCrc~ irc (ir~~tt~c- .. .scC r.e;vqi-clrccc.

1: r.- C,; rc 1______t:__c__,;______cI C,;

I. :Cc C T1 V T,;? T II; q.;VV VT VE V "..,rr rr

I~~rrr C. fT'TTIYI?t, 'C' II'VTIFTTT I

I IT IVI.TTTVTT rv ICrr t'I IVIT Ie.Ccv~t~c

I~~~ ~~ C. C.Y~rT rV' CVIITVIV T! VIIVTT'ITTII ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 'tc IIIVITIVVIITIVITII!IIIITIIII

V iii ti~~tI~ri~tTy XITI~ITTTIrITIVVI~tT~ tyyryrrIIV

1,1 Fit IIvr~fVV IT'lltT!I I ilIVVV!C!tIrVIV

tt I ITrTrTfVl,,I rTT,TiVTVVII li I ITTVIV!ITIIVIIIIr~rt~..tIIIr~ I!TTrVITIVTTItITTTTTjil

I ~~~~~ E-......... TiIt? T ;IVV!I.TVIVVTIlIVIIIII T I II' iIV,. I IfI Tf II ! ~TTII IIII I TIVI (1r

TV IYV I

ID Ij IU Ii IE. w( in I- 1 7 11 11 711 1 1111 ITII

eI 4j 10 1~(ic'.~ T I 1;ItvI

11)~~~ ~ ~ ~ 1 f~ tn I. Il~ I .III Itj IIt

- 45 -

*~~~~c - -.

...... 1.., . *... . . >. . .. . . ... .,..........,...,'... ,,....,..... . ....... ... ... ......... -... --.-C --

ol AI LI I .

ff .4 4 .4'.'.4 ,+ . , 1..1.,, 4+. , * l. . , , ,,4 44 . . . . . .* • . * 4 .. ,, , ,,, . .. , *.. ..* . 1. / LI,

.O ... . r- .. ..... (( ...,', , '' 4 , . , , , , . . 4. 4 . . r 44 * .4.. 1.. ,'4 .L L ...

A, .4. . 1.4,1,.1, 4,... .4 .4.4 4.,.,',.,.,'... ., r.,.. . . . ,. . .1¢. A .- A,: .; /% c e x 9[f-, . ,. ,..(. , ,/".1..1.,. .4. 444-4,.. . 444, 444 ,. * ,4*4, l l l l l ~ l l, ,.4., ''4. k ; '. 4 A .(.L(-. 1 (.* Jill I.

.4 4".,4 .(,14.,4. .. . *. .*' .. . , . , 44 '4 . 4 44. *' ,4 (4 LIL L 4';

[~~~~1 III/11 .LG

...44IIIIIIII., . .I.. . . . ,. . . . .. ,., 4 . . (..14.,1 LI '

Z 4.4..(fi.4.l4l4il 44 .(I

. V£4 .. 4 '1...11 I

.4I .ir . 4 4 r", . ( , 4 G LI

I41f .4.f.'4. ... .. 4.,.' . , (.,,,4. 4,.,(t /.. , 4 . ,q t 44. .l, ,,,41..4..4 ,41, 11

f44.1 1 .A.1r. .. ( A A C,,A ,'. 1.,, " . . . *1 I..4Ac-l . I I

ZIJ. . ".-U,'. -. C, A A, K,,,.I 4 A .1 ILL

I _LI/l / ,,,.,,,, I ,s1..v.A,,,. , ... 1141...,.e.,,ALI 4

"-' " " AI . ' A' A A At"

II 4KS, L.,,14.14 I'~I I. I.,4.,4.. 1, 11 *1 111

., ..',1,1,,,4441,,,'144,,j, :,1 ,,.,.(.-,---(,,.. ...'4,(:4-4/ It-4

I111.,1.44',S'..'~.4.4A41.4'.., S111.. ... 4.,4,....'4' -.,( ........ '4

T4 r. 11. c q I4 T4T(TT',:. "'--''';" ' ." .... .. , ,, ....

1.1,. ¢¢..1.1rTT1.(4? ," ,Ar1.41,. ;..T .. .

-114.,.1.1.r14. 41 i . . . .. .

E , r. ( (. 11 * +' qk q q C, 4;Z+ L 4; . '[ T . - '

c ' - .

5 ' "

J1(4'1414,.. 44411 I C 44 ____ ________________________,,____; + e *

:

A.1.1,A A .4,1 A 4 4 A..1 1, AC A A

o .. ..4... ' 4 Tt,,t4',4-,I

f.. 0.; 1,,., c -- rit t rAr f r r .f t ?r v lf, A o-C -r,'(I 1 .44~4 1 1 , . , .1;rr rr , r I f , I , . ,I I ,; ..,.. . -1-

I ,A tr ,el . L I '.

'"' II/II 4'4 I¢, * ''' - tt l

0kr A r. r. rAA-4 c, r, r. ,.. i. r AtJf.t, 6

1, r,. . , , r..,l l t +' * # € € ;€I(

A A I f, A' -tr Ac

r, e, 4, : A rA r r. (t 4 r ., . , ,1 +

c

e I I U.. 4 -..... l l i I .-e. ............. ..... p Ie.. . . . . il l l

11144l j

Ittt 47e OA "a. " " r f'U- V (111 vi

* .4,,.tj ;.'C€ I ;Vli

- - - - - - - - - - - - - - - - - - - - - - - - - --...............

cvI&t-~1fu L 4 lv

01v.citI 4 4~4-

- 46 -

I C

U 444 . . * . .. .. .C I r1 1q f* ,cc, , ,.. ., L .. +. ' . + '' 4" . ' , . ;f. .C, .. .. . +. .. P

, , . f f,, , f '4 .* ,-i I+II, ,', , ,,++ + 4, f+,

I O k I' .... .. cf. . 44 4. 4 ..f c;4.

iz ' ......I 6 . . .44 . 4'. 4~.~

0 ;c r.c , . ; .c.4 C err:'..C -.''.

2 . . .. cC r.. c 1; (*.' 4

' c.. . C4 .. * 4c..f C;cr4."c-.,.Cj I U * aIDr , 44*c LL 6" C, C C. C C .:C%' t

l/Ji lll 0 : . 4.4. 4 fl 'fcc" r ,,. 4. CA7'. 4 fc . ,.4 Ccf r..f'.

4.

" v" +itl p i. . Oi h~l ll l l. l 4u. l -ll -1 00L C 4 . ;.."

. , ,

I. III~l, , II#1 11, ,4.4-. .€ +. .+ +f, .. ,clf 2

hid. cc.jciiCiti ,I~ccccI~~ 7iZC Ci rcccccicriccci.:

ciiill lll l~lrl , C r .4 C C .., ,- ' c-c

Sc/ll/I iI/i i II'c .c~ii~.I'/l II /I JILL L

. . . I .I I.. I I I I I 1 4 Ir. rL r.v : I, A

T;;;;III IIIII. :, IiI

I'll I Ic.* J ¢I /.,

Li iii, ill/il Ar,,,,,c LJ11 ,L. LIJI.II,¢

TI[ I-+{ I., I I I I , I II I II IIIt L

4 i i i/ l/ . ,r. ¢ .+., r ; r . ,. L

, .€.€,€,€. '. € : C. C.r . €

C.l1 C .Q7.,iIL

I II/ l l ll / l l I I I J I iIIIII I1 4 Ittll l ltllL

r (.4.4.4.4t'i III 'I I t . ,111 I-ILI.II 4 ;4..I I. I I I I I I I/I I I I I LI .

E~r I// / / L / /.IL LL I I I L I I ILIL L I II I LIL r c; ILrI

r 1-.;'. r.'. 1.'..s;4 .. . 4c

ill/cI /1lI]I// ////IL I// if/il i .' /l L 'ii l/ILILIIIL I[

7/ /4.l/ l l II L

II l/I/t/i/ti I. L,) il/1-:,, " --i(l/ i/i I./-hl/.I t.LLIILLLLlt LI.~t

,; .. ,7 ,; CIv / L 4/LL/ 'irlILLi.LlLILIIILlLILLL

IC '".4.

I 4

C4.4 . c4 -

',c:, '.4..rC;.'.' .;' hJ/hdlI1L/tLiIL li'II/IILILilIL/iLI .4 ;q.c-c tr.,',. ,r .r.c'4, .. + ' ,r:,:i'' I Ic /c,,/ , / / , I ',/'It ILI LI /II/

Jr.,-r44.- ,;*¢.-4 ( --4 ; , , .-- if.., 4~. JZ i//////iLI// I.iI/i/LlLiILiLi/.

Ic (4.,+. ; cc,4 .4-¢...-,. ;: :c ;,,;.4 .,. ;(;,,( . 1 L I I / I L I i I I L 4. I

144.4'' *'4c..-'.',-"! iLI/lci

44.4.'4 ,'','t r c4 c rf:cc:;.. IL/L'rf 4-4+(.,., ., 4;f~+- ,,; s's

4, I ' 4.'C" 4.44;q C'.."L ; .'"

I . .- .c .. .. .. .. . ... ; . ... ... ... . . .... . . . . - -.-

,If . c r',.

¢ +, . , 5* ," ,c . ,f ,5 r ,47. 4 .. ,++" €.. , ',',.c.s'"€444 ,r . ,. ",'+ ,.'..4'+"7 €f ,*P.42c

f4;.4;,',r +: c + ' ... \' .. . . . . " . . ,4.7 +'. . ', *-;I. 4.+ ".4 ro... .. ;C .. ,. '''e , C '.44

' (

2,p 1 I ,0 t t I .r 0 14WN

,1 'A I 2 M11 ;C 1I O o If'i d 1

-LIMIU IhAM311A 1",1 VSi.V Ivi-W

- 47-

11 r r'.rrQC rrrrCC.,r r,

I ~r. (I r.

. . . . . C.CC ;",(;

YL Qc.CC. r.rf?:( c

r. s , z. Z: c ;r.

rccrc 1: rn CC'.''c c : zCc

ir JiH'II'',,j' H IrC C~r I _-c C.;;~_r ccI IcCI/IH~j cC C~CCCCqqcCrqr.:~''

IIIII'IIIII I / t i I I / L Cr ( / Irr

f. A' AAC.~ Ir~ r.CC,.'-~ r ;rrcr Irrc1 ;cI C.Cr~QrC.crIlIcl-. .Q-' clcc:. -

I ~ ~~~~~~~~ C. C ~ ITIT VVI TTI

!I ~ rrr~CrcC Qr CCCrCC.Q It!TIIl"C r.CCC C .I _c rTT

I C~v.~ecccC~ llC I (itI r cc vci~~rloI u

I ~ r ICccQ~v TTTrryTT V iI c r r 1;. cC IC C VC 'T TTTTTIrT V .

rI- C (C.CC illT T I IT TT - -

j Ir fI I IT!'rTT t 0 x E I I II IItIf I TT, T 0 T Is I~

c' CC c C( . rc~ I 1 I 1I V3

C_.. t, fO C v1

I c ' cc cc. C- c%; rc I~j 1 11r '

cjc cC I; CV- T. I T~ T T I E I

- 48-

*I I

*~~~~~~I ItL /ii lr,/:IIL L

E in;ii i:I 11£I

I ~ ~~~ - liIII,-Iii

I Im '11.1 LLllI111 11 1 i L

T Ui IsiiijitL I/Z lrr :ckcc'rr' I C rc.I Ii I du IL S/Li1lt , IL

ii /t li i ,lti l LL IIILI I IL %.

I.r r r r!,cc : Z!. r. f C c r.C L.1iti ..CJ L 1* C(~C~ C.CC te LidiLi ll.it LllI~~c .4cc~.cc- .cc . L

CI LI cf Lii r. I r-ccvc (; L11111 1c -1 V C~~C.~ V C

T1 1

i liliIli ii -,fj4~ LI crr. ~ c CVV' CC1 :

III II iI qLJIJI c.~~ v'~ 11ccC%- ,CVCC ,VC CCC^C;CC

I Lil ~. i C crrc r

iti I: r c~c CCCcCcc(C 5*eci C CCvCqV;C'iv

Iiii iii. cc ,;c*cc

I ACI~~/~A.4, I LI I v e ve C vvfIc 4 ,S .'fr . .A ,;f III /I '.( ,C C ~ . V

A r.(i( t., frf .Ak .Ac C ceCICf!

I C, LI 'c-, v vLL 1 cSl/~ 7i C'".C ,Vs %.. k

A A A A 4 Lil q c1 7CCC V cC ViVqfIr r, ;.'' fL " G !1

A f r, . III v cc %C

1~J 44 ((.cf //

I '"f, / it

r:~rr. I

11. is,~ r. r.c 1 .,c

cIlfrcc~~. I.',..C c

.... ,.. ...... ..... I....... .. . . .. . . .. . . . . . . . . . . . .. . . . . . . . ... *....c.c.,,c- I

AtftI 'j . %IY miif 1414 An i l e1*41u

tf'' * n f ot I s( I TIK

-49-

~C 1; :':ca', E N af

C.~~~L xe.a ~""r~ r X E c~

k; r.t~ .C ;'r eraa a'.. '.aaC' E 0

I~~;aa:~~.C'~ca ,e-e- -ca...~"C.;- J

C..rIr~ r (r1c.'.;. . . -11.. ,~aa .C~aC .C CC~ .(.. E,~. ~ II c r' c ~ c r. r 1 .:

c . -.t e'.:.~a'..r-c.CaC a.

I c; Cr, 1:"c.c.r. o CLU.:

I (rr C C" C ' ' C*:'a a

-1. l Q',a'-.'ccC.. (:-. zr~

c(Ia' ccc :C~a cC.LC4C.CCcC

a' .C.CC CC'

II111il lillil L4

I ~ ~ ~ ~ ~ r Aiip~i AC il '~ A r A A rC

lli jIll,. P/ l A .'~~4 a.C -. r '*C cca',,.r.C C cqq(Zc C. ~ii6 Il/l (+6Cjff.CC'A. .'c cCICC c C4C C C

I/f/0 1 f r,11 .. A (,.aa. p l.,C ~ '''''J C Cr~ (..crCC r-rc lC.r1I 1 J 11 1 I I Y'r/ i,.'.. .A CCa.AI'I /H* U *

4: t;/ L. Il.I c ,.'A.C...d..' C .1 ,Vr'' " ;QrrqC C ,C ;C0; 4111 AA a .4/ AlP I A;' A Ai c~ / c" 'r I . (C

A I f. r, A I rI f, I I, P, I. I I I P. tI c/ 1. r p ~ /P I .I I I

I1/ PPi 1//IiPPa I I I I L I IlI ,.P/i1I I l I I CorfcqC

I / ~~~~~~' /~~~ I L I I I I I I I I I I I / /I/ i / 1 1IP iI I I /1 11CC'~ ' ' ' IP/II P 11/I I i/ I, / P / p1 p//IL

* I/ill //Ii/Ilill IlL~li ipij/jiip ptCCi/PP ~~~~ ~~~ I11 /I//IPlPdIlL

I LII.' ~ ~ ~ tl~ I111 Ip/ /11llI~I/. I' cLII rrcr ft II/lP/P'IP 1 PP illl/I/ PIII PJ L.1111ct'1..C

ea-eC~~.-.r:C""C'.' rCC(eC / ' a I pPIPi Ii I 1111/ P P PI I I I P PII' Cr C ''ra.'aC".-'.'C~',. ti;- C ' lDIP Ia/lp 11Ia i ,a

1/ ~iaiPP I a/i//P Il/i/' /PIPLcaI ,"a:''.a.- 'a ,~**,* II al/i/I/I 1/P/// 11:/I P/II'~~~~~~ ~ ~~~~~~~~~~~ C. ': ~ .. C.. ''I/pP 1. I/lIP/ui'II

I~ ~~~~ VQ V.'r. r.* .. '.* CC'a r: s' ,,~ I '/I/ii/Pi'P

.*... ......................... .........../--------//----i----

1)v ~ ~ ~ ~ ~ ~ C Il- lyr I fIllAtI :-'Ift, : 1 l C1 t0

Cmr" I'ltll CCCC('.C-ill q %ul.rJra.'CC.L,:C'

- 50 -

a Q

31 ED E-0 '11 E

- -0 .

i a, e :,6,.. 1UI

I

.1L

f1I T~~ f I~ TV I lf0 t I4- T1 -TT T

1 g2

14 lI iIIIIm

[ II t -LIIII IL/Il .AA~f~AAAAGGA LSL/ II IIIIi ILItIiIILL /11III1II/IIL

I A, .&A5~ A iii/L //l/l /ll I P1//SlPlLglljllj llj ll/ l lI A;~AAAAAGG.(. LIIiILILLL/ ILI iI/I *1111/IILL/i IlL/I F t L

I './AAAAAGG LLIILIII IIILLLL[LDL~LI/II iii /11/I IILILILIAAAAAAGA iLL lLILILLLLLL/ILLSUL/IILLILILL/ILLLI

ni/I/,u, lll/ / llllllAAAAGGIVt#llll. A

|11111111!111I~jI/i I 11/ AILIIItIIIIIIII~

IiIIIl ,llll i//l // 4/4 11. I///IIIIIIIIIIIlJIIA4

|jIiIIII////lI /IIIII /I IIIIIIIIIIeII

, , , ,, , , ,, . , zl'''"'"""

|I ¢€€ € € €€,c €€T€€€;€€€ C€ :€'€cc¢C

1

C-. CCCCCF~C'f ?I I T T-TT TT'[ CTTCCCTCrcrCFCCTT~EtTfI TcC Cl~aTTTTTTTTTTcctT cc l

ITTT Ccrt~tTTTTTTcTTTcct TTtrCcF, CrTTTC VVCCICT~tVT _TT_____ _____T__T

I TTTTc TVcT,tgI Tvt ,Itc ~ cfCc Ccccccecc F F cc t-c V ccT' cT c ccv Tcc~ Tcc cccvTT cc Tc

*T TcITcvcTTecvcTceccvTcivvvcTccvT.,ctTTc TcvcccTITIIccece c IcTTTcerTT cccvTIccIcc TITGTt'Ctf TcltlcFlvV,ccITTI+T ccc 'fTcTvTvTcTTvcvc vITTITTccc ccccv cv cvccc cc T I rF, lll ITcclclf Tfce ceIITcIT,

|T cccceccTccccTcIvv-'.-c TcTccTecTIcc cTc~i. IT*

|I f f l~ l~lllllll llet lltltfftfeltllffT fl.

l~ffl~llff f~fT ItITTTx[IIt u 11ltl 1TlTT iittlTT lIIT IIT~Y t~lfTT ttlttfit~litIt tlTfft frftllrT

*1 It lYItlllltllllltttiltl t ii lt t t ltllti 11117 tl Il~ lfl7? VI Fi~tl'ferii V 'T ljt|flt?t ll t FVtt ll? tlt T

I ilrlrtlffffrrl l? iU Ff~ll 9fll tllltti Fflllifl tlytll~

*11l1f1t 11tl l Tlli fit?'lt l it til tttt tlll|Ttf t?,f, etlt T|

------------------------------------..... . . . . . . . . . . . . . . . . . . . . . . . ...... ..... et t--l IiI *Ii4.h(Fn ... * SIe-e-e e f e

t lh

e e 4 V

eI

eFJ

eIIV

I f m l iI*I lfel11t! 11tv1ltt ,t,'vt¢

F il4 l i ,11MEOWx

1%R