Ehhhhhmhmhiu /mEnnEi/IEE/i/ - Defense Technical Information Center

186
jD-R133 350 REPORT OF TESTS OF A COMPRESSOR CONFIGURATION OF DCH i/2 BLADING(U) NAVAE POSTGRADUATE SCHOOL MONTEREY CA S J HIMES JUN 83 UNCLASSIFIED F/G 20/4 NL Ehhhhhmhmhiu mEEEmEmEnnE EEEEEEEEImImI /mEnnEi/IEE/i/ llllI~gggEEEEE EllEllEIhEllEE EllEEEEEEEEEEE

Transcript of Ehhhhhmhmhiu /mEnnEi/IEE/i/ - Defense Technical Information Center

jD-R133 350 REPORT OF TESTS OF A COMPRESSOR CONFIGURATION OF DCH i/2BLADING(U) NAVAE POSTGRADUATE SCHOOL MONTEREY CAS J HIMES JUN 83

UNCLASSIFIED F/G 20/4 NL

EhhhhhmhmhiumEEEmEmEnnEEEEEEEEEImImI/mEnnEi/IEE/i/llllI~gggEEEEEEllEllEIhEllEEEllEEEEEEEEEEE

* . .- ru - -

._ lag-.

18-.

1.25.LA

MICRC-P R O

4634%

IIII II'

MICROCOPY RESOLUTION TEST CHART . " i

NAT W~ S lA OF STANOAHOS -IK3-A + -

'- A

} ... . ... ,...e.... . ,*+p : .- - +o o + _-. :.,,+ - l'l f

+ - t- i" i + e I' . - . +. " " " " " " " " " o

" " "

+ " " "" " +

+ • +' " "",.,, .''.'," - ,t,, ..- ' .".-[ ...-., "-..+' ...-.-. ,++-* ' ' .' ' . ' .. " .* -..- " ... + . " ". ' .. " - -., " '

*. + . ,+, * "- .* ,..,,,A-*. '. '.,.. ,.* .-+, ,. -. - . . .-. - .- . -.-.. - . .. . .. +.. - - - . . + .. . ..+.. . .

" . .m -""", .'"e "" , + """" . '* . " ++ , -"+"".' "•- +'"" +' " . - "+"""- ""+

- ""- "" . "+•A

NAVAL POSTGRADUATE SCHOOLMonterey, California

~OCT 1119 83 .

THESISAREPORT OF TESTS OF A COMPRESSORCONFIGURATION OF DCA BLADING

by -

Stephen J. Himes

June 1983

0...Thesis Advisor: R. P. Shreeve

Approved for public release: distribution unlimited.LA4

8~0 t~OT 034-

UNCLASSIFIEDSuCumtT CLASPIICAYIIS Oir THIS PAGE (ohma a~ _________________

WORTDOCUENTAiON AGEREAD ISTRUCTIONSru. l uuuuuw jL 1 Oovi ACCESSION MO. S. RECIPIENT'S CATALOG NUMBER

4. rgmga ( *tde. TYPE or REPORT a PERIOD COVERED

Report of Tests of a Compressor Master's ThesisConfiguration of DCA Blading June 1983

6. PERFORMING OR11. REPORT NUNBER

7. AU No a) I. CONTRACT OR GRANT NUNIZR(a)

Stephen J. Himes

9- U61F~~k111 *N MANIZA ION MNZ AND ADW 10I. PROGRAM ELEMENT. PROJECT, TASKNaval Postgraduate School AE OKUI UURMonterey, California 93940

I I- CONTROLLING OFFICE MNZ AND ADDRESS 12. REPORT DATE

Naval Postgraduate School June 1983Monterey, California 93940

14L MOITORING AGENCY WNK a AOOReS(if EUUW=# fto C.,hue*u Offic) IS. SECURITY CLASS. (of this "eport)

UnclassifiedS.. ECLSSIICATION/ DOWNGRADING

1i. DISTI111TIO10 SATEMENT (of We Atepmej

Approved for public release: distribution unlimited.

17. OISTRIUUTI@N SIATESMEN? (of * 860~a mftgd #a Week 20. It difwanuIdu Repwet)

I&. SUPPL611EN1TARY NOTES

19. KEYV WORDS6 OAII0 r ~a n~ * r~e aide it~ j, eom old*j Idfl#UF by' Week mmbf)

CascadeDCA Blading

*SL A9SYRAC? CCNES10M 4011 MWO S fmeevide @96 m 011ty Wpbe** emS..)Results of an experimental program to measure the per-

formance of a compressor stator cascade consisting of 20DCA blades of chord 5.01 inches, aspect ratio 2.0 and solid-ity 1.67 under conditions of varying incidence angle andReynolds number are reported. Flow quality and blade per-formance data were obtained using pneumatic probe surveysand surface pressure measurements. Changes in Reynolds

~ Pfu 143 S/fN OP02 L4 014- 661 UNOET CLASSIFIEDS/N 1O2.P.O~.e6I 1SECURITY CLASSIFICATION OF THIS PAGE (11hon Dwsow",t

Unclassified ii

(20. ABSTRACT Continued)

number in the range of 500,000 to 770,000 did not measurablyaffect either flow quality or blade performance. Changesin incidence angle over the range -15 to 10 degrees producedgenerally well behaved blade performance parameters.

.'

S'N 0102- LF. 014. 6601 2 ucasfe

guCURItv CLASSIFICAION OF THIS PAGE~~mbE ae.

~°--

' -. - * -*: -. °

Approved for public release: distribution unlimited.

REPORT OF TESTS OF A COMPRESSOR CONFIGURATION OF DCA BLADING

by

Stephen J. HimesLieutenant Commander, United States Navy

B. S. A. E. United States Naval Academy, 1974

Submitted in partial fulfillment of therequirements for the degree of

MASTER OF SCIENCE IN AERONAUTICAL ENGINEERING

from the

NAVAL POSTGRADUATE SCHOOL

June 1983

Author: _ _ _____

Approved by:/ j.3AcThesis Advisor

Chairman,Department/of Aeronautics

'""e~a1of Science and Engineering

3

: " :*-'& ';< ;":~~~~~~." :.... ."......,. ............ -.....- °.. .................. ::

1.7e

* * F - -- . o..- .°

ABSTRACT

' Results of an experimental program to measure the per-

formance of a compressor stator cascade consisting of 20 d/- , ,r-4r'

(DCA)blades of chord 5.01 inches, aspect ratio 2.0 and solid-

ity 1.67 under conditions of varying incidence angle and

Reynolds number are reported. Flow quality and blade per-

formance data were obtained using pneumatic probe surveys

and surface pressure measurements. Changes in Reynolds

number in the range of 500,000 to 770,000 did not measurably

->. affect either flow quality or blade performance. Changes

in incidence angle over the range -15 to 10 degrees produced

generally well behaved blade performance parameters. , ,:

4

TABLE OF*CONTENTS

I: ',. ,, INTRODUCTION ...... ..... .. '. . ... .* .** ..* ....... .. * * ,.-*, *,13

TEST FACILITY .. ................... ........ .. 15

A. H~IGH REYNOLDS NUMBER CASCADE ............ 15

1 . Wind Tunnxel ...... . ... . .. ....... . .. 1 5

2. PlenumModifications ...... ... .15

3. Cascade North Wall Composition ........ 15

4. Inlet Guide Vane Assembly .... ..... 15

. Test Section.................. ..... 16

B. INSTRUMENTATION ........................... .16

1. Survey Probes............... .. ....... 16

2. Reference Probes ............ ........ 16

3. Wall Pressure Taps........................ 16

4. Acquisition and Reduction System ......... 17

5. Measurement Uncertainty .................. 17

EXPERIMENTAL PROCEDURES .................. 18

A. GENERAL .... .......................... 18

2. Flow Geometry. ...... ............. ... .18

3. Reference Quantities .......... ...

4. Performance Parameter .............. 19

B. SPECIFIC TEST PROCEDURES ................ 19

1. Probe Calibration .............. ....... 19 ..192. Test Section Setup and Adjustment ........ 19

5

.Refereneuantities .. .. ........... ... 19 . .

B. PEC FI TE T ROC DUES .. ... .. . ......... 19 . . .. ..

3. Test Measurements ........................ 20

a. Blade-to-Blade Survey ...... 20

b. Spanwise Survey ....................... 21

c. Instrumented Blade PressureDistribution ............. 21

IV. TEST CASCADE AND PROGRAM OF TESTS ................. 22

A. CASCADE CONFIGURATION ........................ 22

1. Blading .................................. 22

B. TEST PARAMETERS .......................... 22

1. Incidence Angle .......................... 22

2. Reynolds Number ........................... 23

V. RESULTS ................................... 24

A. FLOW QUALITY ................................ 24

B. CASCADE PERFORMANCE ................... 24

C. TEST REPEATABILITY .......................... 24

VI. DISCUSSION ......... .................. 25

A. FLOW QUALITY ................................ 25

1. Uniformity ........... .............. 25

2. Periodicity .............................. 25

3. Pseudo Two-Dimensionality ........ 26

4. Effect of North Wall Composition ....... 26

B. CASCADE PERFORMANCE .................. 27

1. Effect of Reynolds Number ................ 27

C. TEST REPEATABILITY ............ . ........ 27

D. COMPARISON WITH CINA ....................... 28

VII. CONCLUSIONS .................................... 29

6

, , ' _ e~. ...> , '. . . .. . ****.. .............m lie m mmnnmmm Im ~ ~ ia -a ( -----...-... .-. --.'.." ., . ... ,,- ..- -.- ,. _..,.... ,-.... -

VIII. RECOMMENDATIONS ..... ..... .. .. .. .. .. .. .. .. ...... .31

A.PPENIDIX A ................. ....***...**-56

A. FLOW QUALITY DOCUMENTATION . . .5

3. Pseudo Two-Dimnsionalit ...... ..... 56

A. STORAGE OFEXPERIMENTALDTAo~ ......o..170

1. Storage Devices ....o- o o.170

a. Cassette Tape . ....... ... ooo. . . 170

b. File Naming Scheme ... oo_....o..170

2. Thesis Logbook .... ..... o.o. 172

B. PROBE CALIBRATION DATA o...o .......-. 172

APPENDIX C...... ... o.... o.o.oo- . . .. . .. . .. ..... o..174

LIST OF REFERENCES . .o..oooo. . ... .o. .. . .. o ............ 178

INITIAL DISTRIBUTION LIST ... .... ...... .... ..- o...180

7

LIST OF FIGURES

Figure

1. Test Facility Schematic ...... ........... 32

2. McGuire's PlenumModification .................... 33

3a. Cascade Test Section, Plexiglass Wall Installed ..34

3b. Cascade Test Section, Steel Wall Installed ....... 34

4. Test Section Instrumentation and PhysicalDimensions ........ ............... 35

5. Cascade Geometry, and Definition of Angles ....... 36

6. DCA Blade Pressure Tap Locations .............. 37

7. Photograph of Center Instrumented Blade .......... 38

8. AVDR versus Incidence Angle ...................... 39

9. Loss Coefficient versus Incidence Angle .......... 40

10. Deviation Angle versus Incidence Angle ........... 41

11. Static Pressure Coefficient versus IncidenceAngle ............................................ 42

12. AVDR versus Diffusion Factor ..................... 43

13. Loss Coefficient Parameter versus DiffusionFactor ........................................... 44

C.1. Blade Space Control Volume ...................... 174

C.2. Assumed Pressure Function ....................... 176

8

. . .

LIST OF TABLES

Table

I. Measurement Uncertainty .......................... 45

II. Cascade Performance Formulas ............... 46

III. Cascade Configuration Parameters ................ 48

IV. Test Blade Coordinates ........................... 49

V. Variable Test Parameters ......................... 50

.VI Flow Uniformity Summary . ...................... 51

VII. Flow Periodicity and Pseudo Two-Dimensionality Summary .. .... ...... .. ....... . ..... 54

VIII. Cascade Performance Summary .................. 55

B.1. Raw Data Storage ..... ............. ............. . 173

9

-.- ,., .. * -- -. . . . . . . • - . -

LIST OF SYMBOLS

ADVR Axial Velocity--Density RatioC fB Coefficient of force based on surface pressure

integration

C Coefficient of force based on momentum conservation

C Coefficient of static pressure rise

Cx PstatCoefficient of force in the x direction based onxB blade surface pressure integration

C Coefficient of force in the y direction based on-.- CyB-" yblade surface pressure integration

CCoefficient of force in the x direction based on

momentum conservation

CyM Coefficient of force in the y direction based onmomentum conservation

c Blade chord (inches)

D Diffusion factor

h. Spanwise depth of control volume at inlet (i = 1)1 or outlet (i = 2)

i Incidence angle (degrees)

0 [I0 Vcos~dx]/[ orefVrefcosadxl

k2 0V0/ref refC

P Pressure (in H20)

Q Dynamic Pressure (in H20)

s Blade-to-blade spacing (inches)

T Temperature (OR)

V Velocity (ft/sec)

W Relative velocity (ft/sec)

10

- ; . . .. .

X Velocity, non-dimensionalized by the "limiting"velocity, VT = V2 Cp Tt

x Coordinate in the blade-to-blade direction (inches)

.y Coordinate in the axial direction (inches)

z Coordinate in the. spanwise direction (inches)

- Air angle, measured in the blade-to-blade plane(degrees)

y Stagger angle (degrees)

6 Deviation angle (degrees)

a Solidity (c/s)

Pitch angle (of air flow), measured in thespanwise, blade-to-blade plane

*Blade camber angle (degrees)

[ cos362/2a cos2 $I] Loss coefficient parameter

wLoss coefficient

Subscripts

i Refers to traversing plane; i =1 for inlet,

i = 2 for outlet

p Pressure

plen Plenum (supply)

s Static

t Total

u In the blade-to-blade (x) direction

wz North wall, lower plane

1 Inlet plane

2 Outlet plane

4 11

' ' '"" " " * " ".*.... .' - U', " ' , '"" "" '". "". " - * *- "- "- "- " .'-' . -"

- " - . " "- "

ACKNOWLEDGMENTS

I would like to express my thanks and appreciation to

Dr. R. P. Shreeve, Director, Turbopropulsion Laboratory, for

his efforts in making this study a worthwhile and enjoyable

learning experience. Special notes of thanks go to Jim

Hammer, Kelly Harris, John Morris, Al McGuire and Friedrich

Neuhoff for their prompt and efficient assistance. Finally,

to my wife, Melissa and son, Michael, thank you for your

patience and support during the course of this study.

4..

12

7

I. INTRODUCTION

The on and off-design blade element performance of, and

flow through a double-circular-arc (DCA) compressor cascade

is being examined in a large subsonic cascade wind tunnel.

Tests are to be made subsequently of a controlled-diffusion

(CD) cascade configuration designed to replace it as the

midspan section of the stator of an axial transonic compressor.

Previous phases of the work involving tests with the

present DCA blading were reported by Cina [Ref. 1] and Molloy

[Ref. 2]. Cina's test results were considered to be prelim-

inary because the cascade flow was found not to be periodic

from one blade passage to the next, but was periodic from

blade pair to blade pair. As a result of Cina's work, a

modification to the inlet guide vane (IGV) assembly was made,

decreasing the space between vanes from 2 inches to 1 inch.

Molloy set out to repeat Cina's experiments, but encountered

IGV blade flutter (which resulted in some blade damage) when

adjusting the vanes with the cascade running. Facility modi-

fications were made which included first the installation

of plenum chamber turning vanes. This reduced the large

scale flow fluctuations which had prevented the use of tufts

to determine the cause of the flutter problem. Also, the

vane actuating mechanism was stiffened and new cascade oper-

ating procedures were adopted. Following these improvements,

13

I,

' " "" '"'" " " " ' " " " ". .. . .. ....: . - . ,. Z "i L .. . . . - ,:, ' . ?-

the immediate test objective was to document flow quality

and the performance of the reference DCA cascade. Added

features of the test program were to investigate the possible

effects of Reynolds number and to determine the degree of

repeatability of test data from run to run and also from

cascade build to build. Additionally, an improvement of

the analysis required to derive blade force coefficients

from probe survey data was attempted. In the absence of

wall suction, side wall boundary layer thickening from inlet

to outlet measurement planes causes the control volume to

take on a three-dimensional character. An attempt was made

to account for the additional pressure-area terms on the

control volume surfaces which accompany the streamtube con-

traction by using the Axial Velocity Density Ratio (AVDR)

and an assumed linear distribution of static pressure acting

on the streamtube walls. Details of this analysis are pre-

sented in Appendix C.

The following report documents a program of 20 tests

carried out using the DCA cascade following the facility

improvements. Results are given and are discussed, partic-

ularly from the viewpoint of flow quality and test repeat-

ability. Details of the results are summarized in figures,

tables and appendices. Appendix A contains flow quality

documentation, Appendix B, a description of experimental

data storage and Appendix C, a discussion of blade force

coefficient analysis.

14

. o . . . ..... ... ..

II. TEST FACILITY

i A. HIGH REYNOLDS NUMBER CASCADE1. Wind Tunnel

A schematic of the test facility as originally

configured is shown in Fig. 1. A complete description is

given in [Ref. 3].

2. Plenum Modifications

The plenum chamber (located in the basement of the

cascade building) has undergone successive modifications

to provide acceptably uniform inlet conditions measured at

the inlet guide vane station. Modifications conducted by

Bartocci [Ref. 4] and Moebius [Ref. 51 were supereded by

those of McGuire [Ref. 6]. McGuire's modification to the

plenum is illustrated in Fig. 2. McGuire's plenum modifi-

cation was in place for all tests reported herein.

3. Cascade North Wall Composition

A plexiglass north wall used during cascade tests

enabled McGuire to conduct a concurrent flow visualization

experiment. The regular steel wall was used for two of the

twenty test runs.

4. Inlet Guide Vane Assembly

The inlet guide vane assembly was the same as that

used by Molloy, with vanes at one inch spacing.

15

1 ' :" ;', ,; ,::;;i :;: : ;: ::: ;;;ii.:Li ;;; :: :; :- ;:

:1 -

5. Test Section

Views of the cascade test section are shown in Fig

3a, Fig. 3b, and Fig. 4. Fig. 3a shows the cascade test

section with the plexiglass wall installed and the DCA blade

row in place. Fig. 3b shows the test section with the steel

wall in place. Fig. 4 illustrates instrumentation placement

and physical dimensions.

B. INSTRUMENTATION

1. Survey Probes

Two United Sensor Corporation five-hole probes were

used for surveys at the upper and lower planes. A

DC-125-24-F-22-CD probe, serial number A981-2 was used at

the upper plane and a DA-125 probe, serial A847-1 was used

at the lower plane.

2. Reference Probes

Plenum chamber reference total pressure was sensed

using a pressure tube suspended in the plenum. Reference

static pressure was sensed using a static port located on

the lower portion of the cascade south wall. Redundancy

of reference pressures was provided by a standard pitot-

static probe located approximately 2 inches from the entrance

to the blade row at midspan and aligned with the inlet flow.

3. Wall Pressure Taps

Two rows of static pressure taps were located on

the cascade south wall 16.25 inches ahead and 6.5 inches

p..1

r'

behind mid-chord. Twenty taps per row spaced two inches

apart in the blade-to-blade direction, each connected to

a water manometer board, allowed visual inspection of the

inlet and outlet static pressure distributions.

4. Acquisition and Reduction System

Data was acquired, reduced and plotted using a

modified Hewlett Packard HP-3052A Data Acquisition System

[Ref. 7]. Two 48-port Scanivalves connected to a HP-9845A

desktop computer via a NPS/TPL HG-78K Scanivalve Controller

and a HP-98034A HP-IB Interface Bus allowed all probe and

reference pressures to use a single transducer. Blade sur-

face pressures were sensed by the second Scanivalve transducer.

The desktop computer acted as the system controller and used

acquisition, reduction and plotting software programs docu-

mented in [Ref. 8].

5. Measurement Uncertainty

Table I summarizes measurement uncertainties.

17

* 'a~.- ~. .. . . . . . . . . . . ..... .... . . . .. . .. . . . . . .

III. EXPERIMENTAL PROCEDURES

A. GENERAL

1. Flow Quality

Flow uniformity, periodicity, and pseudo two-

dimensionality criteria (which implies that area change is

Apermitted) must be satisfied before any calculated cascade

performance parameters can be considered meaningful. These

flow quality criteria are discussed extensively in [Ref. 9],

[Ref. 10] and [Ref. 11].

Uniformity generally describes the degree of con-

stancy of measured values in the blade-to-blade and spanwise

directions at the inlet plane and in the spanwise direction

at the outlet plane. Periodicity is indicated by a corre-

spondence of measured values from one blade space to the

next at the outlet plane. The pseudo two-dimensionality

condition is met when the integrated blade surface pressure

distribution gives a blade force vector coincident with the

blade force vector calculated using the principle of conser-

vation of momentum.

2. Flow Geometry

Definition of geometrical parameters describing flow

through a cascade are.given in Fig. 5. Angle measurements

were taken from a vertical reference line, clockwise being

positive.

18

3. Reference Quantities

To remove time dependency of inlet dynamic pressure

caused by atmospheric fluctuations and variations in blower

speed, each pressure measurement was referenced to plenum

conditions, a procedure validated earlier by Duval [Ref. 11].

4. Performance Parameters

Cascade performance parameters were calculated

using the formulas listed in Table II. Note that the effect

of AVDR on loss coefficient, static pressure rise coefficient,

diffusion factor and blade force coefficient derived from

momentum conservation was accounted for in each expression.

B. SPECIFIC TEST PROCEDURES

1. Probe Calibration

Prior to testing, both United Sensor five-hole probes

were calibrated in a free jet flow. The calibration was

represented using Zebner's analytical procedure [Ref. 121

and computer software developed by Neuhoff [Ref. 131.

2. Test Section Setup and Adjustment

Tests were conducted at constant stagger angle while

varying only air inlet angle and velocity magnitude. To

set an inlet condition, before starting the cascade the lower

end walls were set to the desired inlet air angle with end

wall spacings set to 1.5 inches (one-half blade space). The

IGVs were set to deliver the desired inlet air angle follow-

ing a schedule established in preliminary tests conducted

19

......... .. . . . .

T W T 7-. -C. Z.*- 7 . ' '-.Sr r rrr

4without test blades. Finally, upper end walls were posi-

tioned approximately. The cascade was started and the upper

end wall angles were fine-tuned to give a uniform static

pressure distribution at the outlet plane. A check of the

inlet air angle was made with the lower probe placed at mid-

span in the center of the blade-to-blade traverse. If

adjustment of the IGVs was required, the cascade was first

shut down, the IGVs were readjusted, the cascade was restarted

and the inlet air angle was checked again after the cascade

had stabilized at the inlet dynamic pressure required for

the test.1

3. Test Measurements

a. Blade-to-Blade Survey

Blade-to-blade surveys were conducted using the

upper and lower probes simultaneously. Both probes were

located at midspan with the lower probe trailing the upper

by two inches. Measurements were recorded over four adjacent

blade spaces located in the center portion of the cascade

test section. The two innermost blade spaces bracketed the

center blade which was instrumented with 39 pressure taps.

Survey points were spaced 0.25 inches apart over the outer-

moat spaces and at 0.125 inches over the two innermost spaces.

IInlet guide vane flutter similar to that reported by

Molloy [Ref. 21 reoccurred while adjusting the IGV assemblyat relatively high inlet dynamic pressure (about 17 inchesof water). Thus it is essential not to adjust the inletguide vanes while the cascade is operating.

20

.4 n

• oo o,, o, -. .. ....... ...

Closely spaced survey points over two blade passages in-

sured that one full blade wake would be surveyed with good

spatial resolution.

b. Spanwise Survey

Two spanwise surveys were carried out to estab-

lish the spanwise extent of uniform conditions. First,

measurements were taken with the upper probe 1 inch from

the suction side of the centermost blade and the lower probe

1 inch from the pressure side of the centermost blade. Once

the first spanwise traverse was complete, the upper probe

was placed 1 inch from the pressure side, the lower placed

1 inch from the suction side and the spanwise traverse was

repeated.

c. Instrumented Blade Pressure Distribution

After completing the blade-to-blade and spanwise

traverse, one or more sets of blade pressure distributiondata were recorded.

21

-- .-.. . . 6.

IV. TEST CASCADE AND PROGRAM OF TESTS

A. CASCADE CONFIGURATION

Constant parameters for the cascade configuration are

summarized in Table III.

1. Blading

Test blading consisted of twenty constant cross-

section double circular arc (DCA) blades of aspect ratio

2.0 and chord 5.01 inches. Coordinates for the test blades

are given in Table IV. Three of the twenty blades were in-

strumented with static pressure taps at midspan in order

to provide measurements of blade static pressure distribution.

The center blade contained 39 pressure taps while each adja-

cent blade had 6 taps. Adjacent blade instrumentation pro-

vided a measure of the uniformity of pressure distribution

from one blade to the next. Pressure tap locations for the

center instrumented blade are illustrated in Fig. 6. Fig. 7

shows a photo of the center instrumented blade.

B. TEST PARAMETERS

The results of 20 tests are reported for which the test

parameters are summarized in Table V.

1. Incidence Angle

Seven test values of incidence angle were chosen.

Five of the seven angles were approximately those reported

22

"S - ... ... .. .." ' . -- . . .'"o°,...-. ..... •. .,. .... ..-.. .•. ... '-..... ... °-•-

by Cina to allow a comparison to be made of corresponding

data sets.

2. Reynolds Number

In order to investigate the possible dependence of

cascade flow quality and performance on Reynolds number,

each incidence angle was tested at at least two Reynolds

numbers. The Reynolds number based on chord was varied be-

tween 491,000 and 773,000.

-I2

23

........

V. RESULTS

A. FLOW QUALITY

Uniformity, periodicity and pseudo two-dimensionality

were evaluated from data presented in Appendix A and are

summarized in Tables VI and VII.

B. CASCADE PERFORMANCE

Cascade performance is summarized in Figs. 8 through

13 and in Table VIII.

C. TEST REPEATABILITY

Run-to-run repeatability of cascade performance para-

meters is indicated in Figs. 8 to 13 by double-headed solid

arrows. The arrow indicates the observed range of the appli-

cable cascade performance parameter. Tunnel build-to-build

repeatability was investigated at the design incidence angle

of 2.2 degrees. Double-headed broken arrows indicate build-

to-build repeatability.

02

24

*0~ . o ' - "' -- - " • ° - q " ° • . .' . ° . " • " . . ~ " -. - .- . ..

VI. DISCUSSION

A. FLOW QUALITY

1. Uniformity

As seen in Table VI, the dynamic pressure, static

pressure, total pressure and non-dimensional velocity were

acceptably uniform over at least twenty percent of blade

span at the inlet and outlet planes, and over the entire

twelve inch blade-to-blade traverse at the inlet plane.

Inlet and outlet air angle were acceptably uniform

in the spanwise direction. Beta 1 was indicated to vary

slightly in the blade-to-blaee direction. The average total

change in Beta 1 from beginning to end of a 12 inch blade-

to-blade traverse was approximately one degree. Since ad-

justment of the inlet guide vane assembly with the cascade

running was prohibited, this variation was accepted. If

the IGV mechanism was stiffened further to allow adjustment

during cascade operation, strict uniformity of inlet air

angle would probably be assured.

2. Periodicity

With the exception of the most negative incidence

angle of -12.5 degrees, good periodicity was obtained over

the three centermost blades. Surface pressures from the

left, center and right blades were in excellent agreement

except at i = -12.5 degrees. Also, all flow parameters

25

.4

4: measured at the outlet plane at midspan were closely periodic

from one blade space to the next with the exception of the

most negative incidence angle.

3. Pseudo Two-Dimensionality

Despite the revision which was made to the calculation

of CfM, (the blade force coefficient derived from probe survey

data; Appendix C), agreement between CfM and CfB (the blade

force coefficient derived from surface pressure measurements),

was achieved in only about 50 percent of the tests (Table VII).

The problem is thought to lie in the calculation of the axial

component of CfM. An order of magnitude comparison of the

terms that make up the axial component CfM (Eqn. C. 14) re-

vealed that the first two terms were of order 1.0, the third

and fourth terms of order 10.0, and the fifth term of order

* 0.01. The difference between the first two pairs of terms,

however, were both less than order 1.0, with the pressure-

area integrals contributing the most. Thus, the value of

CfM was highly sensitive to small inaccuracies in the survey

data and too coarse spacing of data points. The more satis-

factory results obtained for the tangential components of

C fM can be explained because its evaluation does not involve

pressure-area integration terms.

4. Effect of North Wall Composition

At one incidence angle the more rigid and slightly

better-fitting steel wall was used. The resultant spanwise

distribution of flow parameters was generally indistinguishable

26

from the distributions obtained when using the plexiglass

S.' wall. (See Figs. A.51 to A.76 and.Figs. A.139 to A.144).

B. CASCADE PERFORMANCE

1. Effect of Reynolds Number

Figs. 8 to 13, and Appendix A show that varying

Reynolds number from approximately 500,000 to 770,000 had

no measurable effect on the blading performance or on the

cascade flow quality. This can probably be explained as

being due to the controlling influence of a leading edge

separation bubble on the suction side boundary layer transi-

tion [Ref. 6]. Conducting tests significantly above a pre-

scribed threshold Reynolds number is therefore unnecessary

for this blade set. It is noted that when operating at

Reynolds numbers of 500,000, adjustment of the IGV assembly

with the cascade running may be possible in the future after

further stiffening of the actuating mechanism.

C. TEST REPEATABILITY

Run-to-run repeatability of cascade performance para-

meters was satisfactory as was build-to-build repeatability.

The total uncertainties in the measurements of AVDR, loss

coefficient, deviation angle and static pressure rise co-

efficient were about 0.01, 0.005, 0.4 degree, and 0.02

respectively.

27

I

• . . . . -° . • ° " "* . *-°- " - , -' " -

D. COMPARISON WITH CINA

Figs. 8 through 13 each contain a broken line showing

the results of Cina's DCA cascade tests. Differences be-

tween the two data sets are apparent. Given a specific

incidence angle, Cina reported a lower loss coefficient,

deviation angle and AVDR, and higher static pressure rise

coefficient. Three contributing factors can be cited.

First, and most importantly, Cina operated with a different

IGV assembly that did not deliver periodic conditions blade-

passage to blade-passage. Secondly, he used a plenum which

gave less stable, more turbulent supply conditions. Finally,

the upper and lower traverse probes were carefully recalibrated

after Cina completed his tests.

28

. . .%..x. t.C.° C .J. - -- *.*-.**°

VII. CONCLUSIONS

Based on an examination of experimental data the

following conclusions are drawn.

1 1. The cascade flow was satisfactorily uniform.

Uniformity existed at the upper and lower planes over at

least 20% of blade span near midspan and at the lower plane

(midspan) in the blade-to-blade direction for all measured

flow parameters at all tested inlet conditions.

2. The cascade flow was satisfactorily periodic.

Periodicity existed at the upper plane (midspan) in the

blade-to-blade direction for all inlet conditions tested.

3. The cascade flow was not shown consistently to be

strictly pseudo two-dimensional. Accounting for streamtube

contraction due to side wall boundary layer thickening from

lower to upper measuring planes introduced additional small

terms into the equation for the axial force component. How-

ever, the calculation of the axial force is dominated by

the difference between two pressure-area integrals of much

greater magnitude.

4. Cascade performance parameters were not measurably

affected by varying Reynolds number in the range 500,000

to 770,000.

5. Test conditions can be repeated within an acceptable

uncertainty from run-to-run. Build-to-build conditions can

29.. . ,-.,.,..,.. '- - ,., -'.. ,' . ..• .. .,. . . .• .. .,

J" " I ' ' '. . ' - "r " - ; , . 'i ' • , , .: ". .'

also be successfully repeated.

6. Use of the plexiglass north wall instead of a steel

wall had no measurable effect on the results.

7. Inlet guide vanes cannot be safely adjusted with

the cascade operating.

30

VIII. RECOMMENDATIONS

It is recommended that the following actions be

considered.

1. The inlet guide vane assembly should be modified

to allow adjustment of the IGVs while the cascade is running.

2. Yaw balancing manometers should be inclined in order

to provide the experimenter with better inlet and outlet

air angle measurement sensitivity.

3. The motor driven yaw and spanwise positioning systems

mounted on the upper traverse caused a significant bending

moment to be placed on the probe blade-to-blade slide. No-

ticeable wear of the upper traverse mechanism took place

during the course of twenty test runs. Removal of the power

yaw and span drives in favor of a lighter-weight mount similar

to that at the lower plane is recommended.

4. Provide ready data file transfer from the HP-9845A

cassette tapes to the IBM 3033 computer so that advantage

can be taken of the DISSPLA graphics capability available

at the NPS Computer Center.

31

4C.)0

0

39

r

'-4JU I.)

.. E

I.- Ig I

m4.j>' I

332

.41

.. i..Tm

- 4-14t4

4C

0 to

33-

Fig. 3a. Cascade Test Section, Plexiglass WallIns tailed

Fig. 3b. Cascade Test Section, Steel WallInstalled

34

Aii

4'4-CA 1-4 $. (

1- 0 - 4-

9L.8

41 A

5E-

5% 35

0.r

~ 0,

'4.4J

Aa

Ci2

36r

I

i0

0.4

, 0 '.

gf-4

I, a

Ina

go 3U

"o "o % . ° ° % tt • • " • • • ° ° " -. .o • • , - . . . . -. . ..

' : :, - : , ..- - . .. .--o , ..,, .- -, ., ..o. , . . . , ., , , . ,. , - ..., . . .... • . . . . .

raye

01

4

38-

ziCD

w* t1.0r-

o 0 r_2 -

o 0D u--, 0 0

0 0 JW. 0 ) 00WC

0 v3'4 -

o% 0t~9-

4~~ 0 0 *.-

"o 0

4j-

14:

(U 0)

C 10W

.05- cc

r4:G

.9- .9-4

r- ).4 r_; I r4

6::4

0 0 CJ) e-40 0 4JI0 0

0 0) 0N

A 0C 0 C4

-. ~~ a. 3

0 0

o 0 CU co

0 0 3 -

AluTR~zea0 0 I-':IPTrn

04 LAl u

0 '0 (1)

04N. -- 4

£~u~I~.z~IV ur-o-in 0

I 3 r4U4

40

-77.

00 4J )

00 koP-00 v

00 U0

V V UU

0 00 U)

0000 I

'V.-r4 to

I t;

r34U)

~UT~ZUU p~nq-o1pC14~

m 00 W V___a9P 9 9TUV OT-uni 9U

44 41

:SeU

00 v

HOC

0 0

Siu~uae00 C.)qoapTn 0)4A~i-Fvia00u un---n Vr

Lim (3

a,4

4-4

0)-

OH)

-- S

S..A

C14

144

422

o 00 4.) 000 o00 o

v v4 ) 0)

a00

0 0 13 0

0

N 0

0>

u 4

051 0 C10

JI: 13 r-

.10..$41

13 0 0

CD 0

u-I H H H H - H H H

-* 43

co %0

0 0

0

0) 134.

Vz~~eu VIn-3-n -U

000

'.0 >A

414

0 $.4

rz4J.

L0~

0~~'44

a t4-4 4J

a44

44

0n

C)1NA0

7I4 00

4 44

TABLE 1. MEASUREMENT UNCERTAINTY

Item Description Method Uncertainty

x Blade-to-blade displacement Position -t.01 in.x a 0 in. West end p2tentiometerx - 60 in. East end

z Spanwise displacement Position ±.01 in.z - 0 in. North wall potentiometerz - 10 in. South wall on probe mount

t 1 Inlet flow yaw angle Angle vateuti- ±.2 deg.ometer on probemount (handadjustment)

3 2 Outlet flow yaw angle Angle potenti- ±.5 des.ometer on probe

mount (motordriven adj ustment)

P plen Plenum total pressure Tube in ±.01 in. H20ple. plenum chamber gauge

P Ztatic pressure at the Calibrated pneu- -..1 in. H'0test plane matic probe gauge

PWI Static pressure at Static tap on .Q1 in. H170x - 0 in., y - -16.25 in., North wall gaugez -0 in.

P Atmospheric pressure Mercury barometerATM

P Pressure Scanivalve ±.01 in. H20transducer gauge

45

-;, ................... _. . ..

(NN

41 o x ro' g 41 1.0

-x 0 r.c.- 4Nt to.5

41 0 0 -0. 4 ,0 0 '-4 Q.4

to co -u w

m r- a - 4 4

04 41 x x x x vax r-4W1 x x -4 -Ir4r4 N A

4i0 4.4 t44 N40 Q0) CD 11 9

4JJ 4.4

a 44 p44

124 aI0 a

r3,ra '-4

at o4

00

.9. 0 f-4

044

'4. r-4 4'

g-4*r *4 14la,

4.4 4Jo J

IU K0 *eI '.-

E-46

4.. .4 4 . .

.11

-. 4.1

N.. IcP

C.e,

+Ica pa

ol (A0 0,

4) P

41 -

0

e- 4) W

410UH0

C.-- 01)

414

(-44

4) -4 -4 4> 0-r4 ON4 I-4

IV 4141

0b4 > 410 0

0.447

TABLE III. CASCADE CONFIGURATION PARAMETERS

4

Constants

Blade type DCA

Number of blades 20

Spacing (inches) 3.0

Solidity 1.67

Thickness (% chord) 7.0

Camber angle (degrees) 45.72

Stagger angle (degrees) 14.27

.48

'

°,' 48

o""°°°"• .°° -.. .,'- •• -.- p ° . ° " ,4 ,, . .- .. . . . . .. . . . ... . . . . . .

TABLE IV. TEST BLADE COORDINATES

X-COORD. Y-PRESS. Y-SUCT.

" -0.044 0.000 0.000

-0.021 0.039

-" 0.013 -0.042

0.178 0.007 0.142

0.400 0.067 0.244

0.622 0.120 0.333

0.844 0.164 0.413

1.067 0.207 0.480

1.289 0.242 0.538

1.511 0.271 0.584

1.733 0.293 0.620

1.956 0.309 0.649

2.178 0.320 0.664

2.399 0.324 0.673

2.622 0.324 0.671

2.844 0.318 0.660

3.066 0.304 0.640

3.288 0.284 0.607

3.511 0.260 0.567

3.732 0.229 0.515

3.955 0.191 0.453

4.177 0.147 0.380

4.400 0.098 0.298

4.621 0.040 0.200

4.844 -0.022 0.0914.908 -0.042

J% 4.943 ----- 0.040

4.966 0.000 0.000

b'

49

V.

TABLE V. VARIABLE TEST PARAMETERS

Test B10 0 ReynoldsNumber Number

1 39.02 1.89 773,000

2 38.68 1.55 511,0003* 38.34 1.21 701,000

4 38.89 1.76 491,000

5 42.66 5.54 518,000

6 43.13 5.99. 771,000

7 45.92 8.79 768,000

8 45.89 8.76 493,000

9 31.32 -5.81 560,000

10 31.56 -5.57 503,000

11 27.90 -9.23 502,000

12 27.55 -9.58 626,000

13 39.28 2.15 545,000

14 39.40 2.27 676,000

15 39.16 2.03 527,000

16 39.36 2.23 604,000

17 35.50 -1.59 643,000

18 35.60 -1.50 525,000

19 24.74 -12.39 535,000

20 24.63 -12.5Q 611,000

*Note: Data from run number 3 not plottedbecause data set was judged to betoo coarse.

50

, .,, .,....,..:.- .....-.......... . -.-.-.....................-.................. •. .. .. ...... ... .. . ..

0.0 c 0f- 00 .0 -4 0 0 im C 0

44 0 u, 0 00 c) OD ' r- '00 0, 0 00

00

4.4

co "0 0 co 0

00 %'0 040 c C4 11 r- UL) e 1-4 r- r-l-4>.1 040C)C

4.8

0 4

%-4S

40.cal .0 41

>4 .04 44 8 .. r-4 .. -4 T v- M %T M% r-4 C4 r-4%

w- c' 0 I I

c Q00

0 0. -,)

-P4 > 0

1- V 0 cc .-

I-I 00 W .W -u- C1400 4 Cn C C1 v-4 ~ % -4-00 t-4 04

-.~~~C 44 4. ' 0 0 00a;0'4 ON 0

P4 V00

-J _ _ _ _P4

0

41 0 ) 0 0 r-0 0 r- 0 04 0 c'4r L

"4 0n V-1' 00 4 L O 00 r-4 0 00 r c -

-H41 0*0 -4 M~' v4I t-O C4 N L -Z(3 LA4

V-4 00 0 0 0 0) 014 1 11 11

(U 0~ 0% -.7L 04 C 4.. Cn rf4 04 -4 rLn O'-4

02 >

410)0 0

01 -4 01 r- ~ 4 C4 v-4 C 0C r-4" cCn 40%.D r-4 0

44 >1 N4 00 0 O ; 0 0 0 0 '-

0 0 0110 II I I I I I I

m 01 C'e4 M v4 M~4 On4" 0 4 - T -Nl00 c

-4 0 010 n 0 1 I 'C0c C 0c0

M cy -o8 08

.J 01 0.% N0C - % P N (4

0 C14 48 . 4 C 4w-I 0: ;C 1 ;C C4 "I AI 0 C1-4 ~ 01-a

* "4 ,4 l 0 00 0 00 0 C'052

0.I =1 0!* '-4u Cl!0 00 CNO CV!

0 A

'0 4

4to = II eg 1- 0. i i.

"4k 0 00*4 - %QO uC4 n -0 o F%Q ON~ (Ln i>.8 -.1

(U* w -1 I~s 0.-4 in4I 0.-I 00000000 8 1; 4 ;I.

4.80 1..0 > "q- P

Io

040V2 u

w 0 0 Hco. 4) 10

wae4 0f 0-. ul CV)-I 00. Cf Lr '-0 t0 -.0' rI " ti-I%

4cc C;8688

03- 00 e4 II -T II NO r-- IT CII0 4L T

0_ 8a ('4( 8~O 8 .4 8-'O 0I4 8 'c 8 8 800%.

cc) r-4N'I (.C' ~ (5 C l vI -iI 0

04.

0O 0l 4 O 4 r 4C)% I

- 09 00C!1

ra~r0

.. J OW 53

TABLE VII. FLOW PERIODICITY AND PSEUDO TWO-DIMENSION-ALITY SUMMARY*

Incidence Periodicity Pseudo Two-Dimensionalityangle based on blade based on blade force coef-(degrees) static pressures ficient diagrams

L magnitude(%) Ldirection(

-12.50 imperfect - 2.1 10.00

-12.39 imperfect 13.0 25.00

- 9.58 acceptable - 1.5 3.0

- 9.23 good 41.3 27.Q 0

- 5.81 good -19.1 7,50

- 5.57 good -20.0 3.50

- 1.59 good -25.2 24.50

- 1.50 good 16.5 22-5 °

02.27 good - 4.6 2.5°

2.23 good - 2.4 2.00

5.99 good - 0.7 7.00

5.54 good - 6.1 4.00

8.79 good - 3.4 1.50

8.76 good -16.2 0

*Note: Data from six additional runs at design incidence

angle are not presented.

54

< "'q" '-";<-"-> ".- ''- -. ; " "'-. -;- -';-; .; . . -."". . . . -. " . - - . .

S. = == ,,, .. , .... , .. . ..

0 0 00 00 00 00000000 00 00C00) 00 00 00 00000000 00 00c0 00 c0 00 00000000 00 00

Noc C %091" 'N -0 -0 ITc4 4 0% Ps.V4 '0%a%0 an %c L In a kn r, r. %c %uuu LninsU44lin -U

%AO Go O IT0 -4 C h7 -40 %0 ar.r- 1,0 m4 %00% (aN4 c 22"gn %a wl eq a -tt 4 0 om0%0 %0 %0 0-

OD 0% 21. C N% .-4 m V r, tC %20N C4 C4 12am .-i N.C4-4 Ln m .7 - r& u,%&Ain~ 6nu in %0%

g'-%0 2%C 0 4* Oin 0%N NN -00' 'tdt% -inm5Ln - T 4%-T r a%0 r, % f0 r% .r..Q-T r. r-ir- a00% 0%a47

Ui I I r1 r-.4 1-4 -4 4 -1 4 4 -6 4 -4 -I- 4 -4 -4II I 1 1 1 11 11 1 1 1 1

%QN 't U'1%0 r-..-4 m %T0 r- 4N04 02n w%C 1% ifr, It 1-4 4N T1 m7' 00 i 2wrw'n C4 t U%Tn al 0'0 ur'It .- 4 C4N 0 r 1'0 CIM " 00 %w 0 00.4 %0 0 MOD

En %a0 N4I- %T-Z 0% enC n f1Go 'T P-N C4NO%0 vi .7

Ch0 -40 r-4 00 co '-0 21-Nr4 ,V w u%a W tf20' at %N 00 s-f 0 %0N14-. -4'tw s-b MOw'4000

07 saG 1 40 NO00 L O r-P m 0?-. %0 -?P% r 00 a 0 0 4 14 M M 4 -4-4 sMIs-b s- C4N-4 C4 C4

0 C C0 00 s-I- s-Is- s4 4s s-I4 s-Is- 1;11; ; 1

0 .494 YG L r D000Go00o000P 0 0 OsI4CD 4 P r- r. C4C4 4-4 C404 04 4 - m -4- C-14n

Ni% 2r n4 02 %0 C CsM-If , . 0024- 02 00 0 OT w' IC4 ? 004' M D y m r -4 00 0-r, P 4 0%'t-N~-- C0.

cc0 0 00 00 00 00000000O 00 00r-4 4 4 .4 4 .4 ; 4 . 4

3 c 00 o0 00 00 00000000 00 00

w C, r r co in%0 0 0 0 m %0 -7w 't 0 5L &ca r m.' M -T 4ITN "7. 0N%Nw'O% an0% 0% tm .0 2

05 1-4 0 ; .7. e'n -T 0 w'2M 0s o -4 0 "0 00 o0r- in oNN4 C4 4 v141 -Isr-I r" " r4. .N .s-4f-4 mN C4 T I

.43. 00 00 00 00 00000000 00 00

ItN( 0't co%0 4' 0 0 w wUN"s4 -1 C4 No r- %nr. 0% fri' Nt C4-T o0r- (4 r,1-N4-4 %0 r%0 5% di IG

C4C 00 %0 90 inA o0 0000 0 0 u 00 c

or- r. c r% a. 00 c 0%14 0 0% 00%o0 0 r4g,-i *4 -4-0 V-4Is-I -4 s-IP-b r4 -4 qP

co0 s-40 't2 . s on r % 'D00 en Ns C -4' %T Ow'n%0%s19 n ' W!'! 0! 0 22 I -!W! I'--O P't'%ti 1 '

cm s-f- -440 00 00 4s-4s 4 4 -4 NN Ned4C4C

00% Go C" 6 0%0 0%r4 r- M '2w's0 % 04.* 0"50& w' N 005n L o(4 C4 CNN -40f 0 V a V r- P.

NN c4 a j% w;w . c45( c-s 4Nc4 s 4 w;U w 000 a

C"7 W,4 a . 00%ac N 170 %0 cca0 (n 0 N4QaW04 5(0' 25(4 LM M 02-2 sIO ("Is ITe00M4%0 % %a

- T 't' r4- LAs- v'4 05044400 2N a* ;USaIA ; v c ;V

55

APPENDIX A

A. FLOW QUALITY DOCUMENTATION

1. Uniformity

Uniformity of dynamic pressure (Q/Q), static pres-

sure (APs/Q), total pressure (APt/Q), non dimensional velocity

(X/X), and Beta for spanwise surveys at the inlet and outlet

planes and blade-to-blade surveys at the inlet plane at each

test condition are shown in Figs. A.1 through A.105. Photo-

graphs of upper and lower static pressure distributions for

each incidence angle and Reynolds number comprise Figs. A.106

to A.119.

2. Periodicity

Periodicity of (Q/Q), (APs/Q), (APt/Q), (X/X) and

Beta for blade-to-blade surveys at the outlet plane and

adjacent blade pressure distribution for each test condition

are shown in Figs. A.120 through A.168.

3. Pseudo Two-Dimensionality

Blade force coefficients for each test condition

are plotted in Figs. A.169 to A.182.

4. Notation

The parameters plotted in Figs. A.1 to A.105 and

Figs. A.122 to A.154 are defined as follows:

56

- . 4 - - . . . - - -

- -4i-. . 4~

Parameter Meaning

Q/Q local value of Q 2

APs/Q (Ps - Pwd /5

APt/Q (Pplen - Pt) / 6

X/X local value of X/X

,

,1

2"Barred" quantities apply to the specific measurementplane and are calculated over a prescribed integration in-terval--usually one blade space.

57

Q'/ (upper plane) VS SPAN POSIT: I - -12.5 DCGSPANNISE TRAIVERSE (I on. from suction side)*Re G.IE5 +Re - 5.3E5

1.4

.2

- m m INCHES

Fig. A.1

APs41 (upper plane) VS SPAN POSIT: , - -12.5 DEG*. SPRNNISC TRAVERSE (1 in. from suction side)

*Re 6.IE5 +Re - 5.3E5

.58

',

!-.

i .1, . .

. . .d " " " - . . . .- . . . . . . •. . : . - ." ' T : 2 . -

APt A (copper piane) VS $PAN POSIT: I * -12.5 DEGSPANIlSl TRAVERSE (I In. from suction side)*Re -6.1E5 *Re -5.3E5

.5

"S--.. ...-.3

INCHES

Fig. A.3

XA.. (upper plane) VS SPAN POSIT: - -12.5 DEGSPRNISE TRAVERSE (1 in. from suction side)*Re - 6.IE5 +Re - 5.3E5

1.4, .

Fig I

S59

".6

:..4

" A e . .. ItCHES

"" Fig. A.4 .5

"" 5

. -.

TRI -6-; 7 %7

-ETA2 VS SPAN POSIT. - -12.5 DEGSPANWISE TRIVERSE (I in. from suction side)

*, - .E5 +Re - 5.3E5

-I

' I-3

.7INS

INCHES

Fig. A.5

6

* 60

", . *. . .- .- . .-.- - - -2 - - . - a. ;-.. --. -. . - :- --- - _ --- ..- \ - -. . . -. ...- .* . . . . ..

Q (lower planel VS SPAN POSIT: I - -12.5 DEG

.... SPRtNISE TRAVERSE ti in. from suction side)

*Re 6.IE5 +Re 5.3E5

1.4

1' INCHES

Fig. A.6

. (lower Plane) VS SPAN POSIT: t - -12.5 DEGSPRNNISE TRAVERSE (I in. from suction side)

*e,6.1E5 +Re 5.3ES.:-4

..

T -INCHES

Fig. A.7

.61

""~~~~ -r -A.I5+R ,

-.

.*INCHES

~ig. A. 7

%., ,-,.~ ,:,:... .... .-... ... .. . . .... .., . . **..~ .... ...- * . ..

4 P/ (lower plane) VS SPAN POSIT: I - 12.5 DEGSPANWISE TRAVERSE (I in. from suction side)Re 6IES . Re , 5.3E5.5

.3

_. INCHES

*Fig. A.8

X/X (lower plane) VS SPAN POSIT: I -12.5 DEGSPANWISE TRAVERSE (1 in. from suction side)*Re - 6.1E5 +Re - 5.3E5

"1.

1.2

4e

.41

-.. 4

4. .

INCHES

Fig. A.962

4'. . ...... . . ....... ...

42

-- t

-I3

INHE

Fig A.10

63L

La - W' . L.~ .. * . - .

* (lower plane) VS B0B POSIT: I - -12.5 DEG

BLADE TO BLADE TRAVERSE tmtdspan)

*Re * 6.|E5 +Re - 5.3E5

1.4

1.2

.5

.6

.4

.2

; TINCHES

Fig. A.11

iPs/; (lower plane) VS ETOB POSIT: I - -12.5 DEG

BLADE TO BLADE TRAVERSE (midspan)*Re - 6.1E5 +Re , 5.3E5

.5

.3

,1

7 "INCHES

Fig. A. 12

64

,, ,~~~~~~~~~~~~~~..,.--.'.,...,....- ,, ... ,...............,... ..... .............. •.....-... ..... ,.... ...-......... . .,S S * * * * .. .. * . . - . : . * x.*.. . .i *L . . . . . . . ., _L,,h. ,3., ,,m,.r.,e ,,,t , .a .' ,.,'

A Pt4 (lower plane VS 9TOB POSIT: - -12.5 DEGBLADE TO BLADE TRAVERSE tmidspan)

- 6.1E5 +Re - 5.3E5

: -"- N ' ?i NCINCHES

Fig. A. 13

XAZ (lower plane) VS BTOB POSIT: I - -12.5 DEGBLADE TO BLADE TRAVERSE (midspan)*Re - 6.IE5 +Re - 5.3E5

1.4

1.2

. 11

.4

.8

' INCHES

Fig. A. 14 65

. . . . .

DCTnl VS 8TOB POS!T: I: -12.5 DEGSLADE TO BLADE TRAVERSE (midspan)

Iq- G.IC5 +Re - 5.3E5

*% I

-Is-

1' 1'INCHES

.[

Fig. A.15

i"6

.L

'S

66

;,%' u-;4: .- ' ,',.1 - .: :.::2, .--..-- - . ...- .....-. -. , .-.. .-.-. . . . . . . . . ....

-q-. I N - - . . .

Q/ (upper plInel VS SPAN POSIT: - -9.5 DEGSPANWISE TRAVERSE (I in. from suction side)*Re 6.3E5 +Re - 5.0E5

" t1.4

1.2

* .8

.4.'.!

INCHES

Fig. A.16

A*pa,, - (upper plane) VS SPRN POSIT: i - -9.5 DEG

SPANWISE TRAVERSE (1 in. from suction side)*Re 6.3E5 +Re 5.0E5

.3

.3

-. S A.

INCHES

Fig. A.1767

,5 ,.: ,' .. '..,. ., ...-- -.- •__ _ _ .- •- . . .- . .

-7,.

APt.,1 (upper plane) VS SPHiN POSIT: 1 -9.5 DEGSPANNISE TRAVERSE (I in. from suction side)*Re = 6.3E5 +Re ".DE5

.3

--.

--.

": -. 5

) INCHES

Fig. A. 18"J

X/R (upper plane) VS SPAN POSIT: i - 9.5 DEGSPANNISE TRAVERSE (I in. from suction side)*Re - 6.3E5 +Re - 5.OE5

1.4

1.2

=.6

.4

.2

INCHES

Fig. A.19 68

- ; f ; ., , .: _ - .- , ...-., , ..... , ..... -... . .• . . . . . . ...

BETR2 VS SPAN POSIT: i - -9.5 DEGSPRNWISE TRAVERSE (i in. from suction side)*Re - 6.3E5 +Re - 5.0E5

IA aI

Fig. 2

N9

-- "-Fig. A. 20

Si.69

S.°

S'°" ,-"* . ." °.'. . " ** " "*''' " .'5 -" - *" " . "-. - ."..'5 ': . .." .. . ' " ', :" * 'i % "

; ' " , ; % : ': ; " - " , - " ." . : . ; " ": " - :. " , -: - - : .' '5 - . '

(O (lower plane) VS SPAN POSIT: * -9.5 DEGSPANWISE TRAVERSE (I in. from suction side)ORe - 6.3E5 +Rc - 5.E

1.4

°2

J4..

.4

.4!

7 I I I N

INCHES

Fig. A. 21

APs/d (lower plane) VS SPAN POSIT: I - -9.5 DEGSPANNISE TRAVERSE (I in. from suction side)

*Re - 6.3E5 +Re - 5.0E5

~.31

T "INCHES

Fig. A.2270

' -~~~.... .........

" -" " ' - ' '-"-". ' " ' "". . -"""", - " "" " -"-' ' '", " " " -"" " """"" '"•' " "-" -" " " "...4 ' -. 1 " ,m£, , ,, - . .. . . e .J , . .. ,. .. ..

APt/d (lower plane) VS SPAN POSIT: 1 -9.5 DEGSPANWISE TRAVERSE (I1 in. from suction side)*Re 6.3E5 +Re 5.0ES.3

.. 4

Fig .23

*R 6.E.+e3.E

3.4.

V 7 INCHES

Fig. A. 247

.,. .-

,, . .. ,,, . , . i, .;, . - -. . . . . . . ..... . .. o - - .. .- . . . . . . .

BETR! VS SPAN POSIT: -9.5 BEG

SPANWSE TRAVERSE k in. from suctIon side)*Re - 6.3E5 +Re . 5.OE5

-23

-25s

oS.

". -27-as

-31

-3

U 9.

'. .INCHES

Fig. A. 25

72

ell',", . ' ,', . .,. ", "- . - . - - . . . . . . .. I

Q/3 (lower plane) VS BSOB POSIT: t , -9.5 DEG

BLADE TO BLADE TRAVERSE (mldspan)*Re , 6.3E5 +Re ,5.0

1.4

'..4

1.2

5%

.2

9' TINCH ES C"

-Fig. A.26

APs/a. (lower plane) VS BTOB POSIT: t - -9.5 DEGBLADE TO BLADE TRAVERSE (midspan)*Re - 6.3E5 +Re - 5.OE5

1"'

!'.3

I 'INCHES

Fig. A.2773

i. ~~... . ... . ..

SAPt/ ' 'lower planel VS DTOB POSIT: I - -9.5 DEGBLADE TO BLADE TRAVERSE (midspan)*Re -*;.3E5 +Re , 5.OES

*R. .3S+e .E

-. 4

1. 91 ? 'INCHES i

Fig. A.28

X'7 (lower plane) VS BOB POSIT: I - -9.5 DEGBLADE TO BLADlE TRA VERSE (midspan)ieg - 5,3E5 +Re - 5.0E5

1.4

.6

.6

.4

.a

Fig. -A. 2974

.i . . . . . .. . " -- .". --- "- . . .. . .... . . . . . . . . ... .- .. .-

. ,. S, . tl . . --_ .. t, . Aj -.. ,,, 4 : .* .1, . T . --. . ,,J'.I. .2' . ... !

BETRI VS DTOB POSIT: I - -9.5 DEGBLADE TO BLADE TPAVERSE (midspan)*Re - 6.3E5 +Re - 5.0E5

•-27

-3

,-t

'INCHES

Fig. A. 30

pp, 75

p' - .- - - ' - 2." " - - - -. . . ....

" 0/a (upper plane) VS SPRN POSIT: t - -5.6 DEGSPRNNISE TRA]VERSE (I in. from suction side)

*Re - 5.6E5 +Re 5.0E5

.4

INCHES

._6

Fig. A.31

APs/ (upper plane) VS SPAN POSIT: I - -5.S DEG

SPRNNISE TRAVERSE (1 in. from suction side)*Re , 5.SE5 +Re - 5.8E5

.3

.2

.t

INCHES

Fig. A.3276

APt/4 (upper plano) VS SPAN POSIT: - -5.6 DEGSPANNISE TRAVERSE (I in. from suction side)*R 5.6E5 +Re 5 S.0E5

-.1

.3

-. 3

S' U/I N, Sl

INCHES

Fig. A.33

X/X (upper plane) VS SPAN POSIT: I - -5.6 DEGSPANWISE TRAVERSE (1 in. from suction side)*Re - 5.SE5 +Re - 3.OE5

1.4

1.2

.5.4

.2.'

% t INCHES% . 4

W.A

Fig.A.3477

.5! ..

"'i

BETR2 VS SPRN POSIT: I , -5.6 DEGSPANIISE TRAVERSE (I in. from suction side)*Re - 5.6E5 +Re., 5.0E5

S

-

!-

4'

INCHES

Fig. A.35

A

.9

78

.. . , ,, *:v,-..,-.- .- v.,,, .- ,'. , . . ". ". . ". .- .... .. : - .. *-,. ". . .-

*T. 7

Q/G (lower plane) VS SPAN POSIT: I -5.6 DEGSPRNW4SE TRAVERSE (I in. from suction side)*Re. 5.6E5 +Re - 5.0ES

1.4

I.'

t

.6

%

*,. .4t

IT 91 ('I ,INCHES

Fig. A.36

* IApsQ (lower plane) VS SPAN POSIT: 1 -5.6 DEGSPANNISE TRAVERSE (1 in. from suction side)*Re = 5.6E5 +Re - 5.OE5

.3

* .3 : - " : : :

-. 1

".3

IT ?INCHES

Fig. A.37

79

APt/a (lower plane) VS SPAN POSIT: t , -5.6 DEGSPANWISE TRAVERSE (1 on. from suction side)

S.6E5 +Re S.QE5

* .5

..1..

%..1

•' 9' • INCHES

Fig. A.38

X/i (lower plane) VS SPAN POSIT: I - -5.6 DEGSPANWISE TRAVERSE (I in. from suction side)*Re - 5.8E5 +Re - 5.0E5

1.4

1.2

.4

.2!

ZNCHES

Fig. A.39 80

,-:,;.-...:, ..... ,,-.-,,-..- ... .- ... .-. . . . .. " . . .. "-. . .... -- _= = = 1 ; I i ii i 4i i i d ~ i ; L . ' : " _, : " " - - . . . * . . .-. ."

•- 71

BETRI VS SPAN POSIT: I - -5.6 DEGSPRN4ISE TRAVERSE (I in. from suction side)*Re - 5.6E5 +Re l 5.0E5

-31

-33

-37T -. INCHES

Fig. A. 40

81

42. . . . . .. . l. . . . .. . .. ' . . . i l e "l ' . . . _ .l . . . . . .

". . , . .. • ". ' ', . . " . , " ,' ', . . " ", . . . . .' ' ., . . . . , , , . , . , " " • , " " .• , ,. " ,

0/6 (lower plane) VS BTOD POSIT: 1 -5.6 DEGBLADE TO BLADE TRAVERSE (midspan)*Re -5.GE5 +Re -5.OES

1.4

.4

INH

Fig. A.41

Ps,' (lower plane) VS BTOB POSIT: =-5.6 DEGBLADE TO BLADE TRAVERSE (midspan)*Re *5.6E5 +Re S .OE5

INHE

Fi.A4

.82

A Pt/, (lower plane) VS 8TOB POSIT: 1 -5.6 DEGBLADE TO BLADE TRAVERSE tmidspan)*Re o 5.6E5 +Re - 5.0E5

.2

°-.

.-.

S'" ? ' INCHES

F1g. A.43

XAZ (lower plane) VS BTOB POSIT: 1- -5.6 DEGBLADE TO BLADE TRAVERSE (midspan)*Re - 5.6E5 +Re - 5.0E5

1.4

1.2

N - '- f

,....'i IIrI N,.

F ig. A.44

BETRI VS BTOB POSIT: -5.6 DEG

BLADE TO BLRDE TRAVERSE (ntdspan)*Re , 5.6ES +Re , 5.OE5

-21

e-31

-33

-35

9' ' INCHES

Fig. A.45

3

8

1.~

:o84

S . . . . . . . . . '* * *'*. ..!* . .-,t. . -

".", (upper plane) VS SPAN POSIT: 1 -1.5 DEGSPRNWISE TRAVERSE (I in. from suction side)*Re , 6.4E5 +Re , 5.2E5

1.4

in 9

.o..

.4

.2

- I i N i

INCHES

Fig. A.46

/ !)/{ (upper" plane) VS SPRN POSIT: I - -1.5 DEG/PRNNISE TRAVERSE (1 in. from suction side)

* SRe - S.4E5 +Re - 5.2E5

,85

M.3.

.1

, .l

-. 1

q . 3 I I I I

(i,..A.47

85

.- ,-, . _ . . . . -..... .. . . . .. . ..... .. a .. .k f

(upper Plane) VS SPAN POSIT: f -1.5 DEGSPANWISE TRAVERSE (I in. from suction side)*Re 6.4E5 +Re -5.2E5

-. 1 ----

INCHES

Fig. A.48

X/X (upper plane) VS SPAN POSIT: i - -1.5 DEGSPANNISE TRAVERSE (I in. from suction side)*Re 6 .4E5 +Re -5.2E5

1.4

- 1.2

.. 4

Fig A.448

BC:TR2 VS SPAN POSIT: - -1.5 DEG

SPRNNISE T!RRVERSE (1 in. from suction side)eRg - 6,4E5 +Re - s.21"5

.5. -3

-4 I I I !

-ig.A- 5

487

-.4-- , , ; - . - , . , . ., ,: o '; . , - . , ' . , . , ,.' , ' . ' v . . . . , . , , . . , , , . . . - • - - . . - . . . ,

(lower plane) VS SPAN POSIT: 1 -1.5 DEGSPANNISE TRAVERSE (I In. from suction side)*Re , 6.4E5 +Re - 5.2E5

.4

.2

N i iI I

INCHES

Fig. A.51

iAPi8 (lower plane) VS SPAN POSIT: i - -1.5 DEGSPRNNISE TRAVERSE" (1 in. from suetion stde)

- 6.4E'5 +Re - 5.2E5

i I I ____

7 INCHES. .. .Fig. A. 52 8

,,, - . ',- ,-. •.,, ",• .,;,,,', ,,.' ." ,, ".. '- "..". " ' -. .•.".. . . .-.-.•...-.... ....... ,. .• ,.. . .-. .•... . . . . . . . . . ,,

Pt,' (lower plane) VS SPAN POSIT: 1 -1.5 DEG

SPRNWISE TRAVERSE (I in. from suction side)*Re , 6.4E5 +Re - 5.2E5

.2

-. 3

-• -°"N

,I INCHES

Fig. A.53

X/7 (lower plane) VS SPRN POSIT: I - 1.5 DEGSPRNWISE TRAVERSE (I in. from suction side)*Re - 6.4E5 +Re - 5.2E5

1.4

1.2

.11

.6

',€" I

INCHES

Fig. A.54 89

BETAI VS SPAN POSIT: I - -1.5 DEG

SPRNNISE TRAVERSE k| in. from suction side)*Re 6 6.4E5 +Re - 5.2E5

-33

4-4l 11 , , I a

-- 4 a " INCHEStT

Fig. A.55

4

90

7-- ..'. .,

7: 7

Q0/ (lower plartel VS BTOB POSIT: t , -1.5 DEGBLADE TO BLADE TRAVERSE (midspan)*Re 6.4E5 +Re , 5.2E5

1.4

3.01

.3

.4

*I .,

e~. 'INCHES

Fig. A.56

BLRDE TO BLADE TRAVERSE (mldspan)

*Re =6.4E5 +Re - 5.2E5

.2

.3

p..,

-. 2

Fig. A.57 9

- * -91

APt/'3 (lower- plan.) VS BTOD POSIT: -1.5 DEGBLADE TO BLADE TRAVERSE tmidspan)*eg 6.4E5 +Re S .ZES

.3

II~ I~ INCHES

Fig. A.58

X..Z (lower plans) VS BTOB POSIT: -1.5 DEGBLADE TO BLADE TRAVERSE (midspan)*eg - 6.E5 +Re -5.2E5

1.4

1.2

00 I~INCHes

Fig. A.59 92

BETRI VS BTOB POSIT: - -1.5 DEGBLADE TO BLADE TRSVERSE (mldspan)*Re * 6.4E5 +Re * 5.2E5

-34

-36

-39

* -421Ir4'INCHES 0

Fig. A.60

.9

-

'"

93

/ (upper plane) VS SPAN POSITi 1 2.2 DEGSPANWISE TRAVERSE (I in. from suction side)

* sRe-S.7E5(pl*-.ig1&ss wall) *Rs-6.0E5(sstee wall).

1.4

1.2

.2

J I

INCHES

Fig. A.61

APs/Q (upper plane) VS SPAN POSIT: i 2.2 DEGSPANWISE TRVRE(I in rmsuction side)*Re-6.?E5(plexiglass wall) +ReS6.0E5(steel wall)

.5

4. .3

-. 3

- Ut INCHES

Fig. A.62 94

V .. a . . . 7*7* ~ . . . ...... ' -.. . . . . . . .--

"Pt /4 (upper plane) VS SPAN POSIT: 1 2.2 DEGSPRNWISE TRAVERSE (1 in. from suction side)*Re,,G.ES(piexiglas$ wall) +Re,6.0ES(steel wall):3.5

.2

-. 3

INCHES

Fig. A.63

Xfx (upper plane) VS SPAN POSIT: I-2.2 DEGSPANWISE TRAVERSE (I in. from suction side)*Re-6.?E5(plexiglass wall) +Re6S.0E5(steel wall)

1.4

.4

. -INCHES

Fig. A.64

-. :X/ (uppr pane)VS PRN OSI: i 2. 9E

" SRNISETRVERE * i. romsutio sde.................................................................................

................. .......................................

FiD-i33 350 REPORT OF TESTS OF H C04PRESSOR CONFIGUJRATION OF DCA 2/2GLRDING(U) NAVAL POSTGRADUATE SCHOOL MONTEREY CRS J HINES JUN 83

UNCLASSIFIED F/G 28/4 NLEIIIEIIIIEIEI///////ll///l

IIIIEEEEIII

smmhmmhhmhhhulllllllll-lI

.. 1.,- lmh ,I,,, it. ,a .. ,. -a, -. .,, -,.,t -.. . _ . I' . , ;,.. . . . , ,... .... . . .

.*. A N .. t.a..4"s.°'..

-7.

1'-* -.

III&=.6 L~m140 12.0

11111 -q

1.5 I.

MVICOCOY RESOLUTION TEST CHARTJHAmMM KPIME of STANDMNM-196-A

2.2:

II .,, * % ,*, % 4,*, *,_,. .* - *,- .,. ,," ,3. .".. .",.'.... . . . . .-.... , ._... N,.". ." • •

, r c .. r r . ., v t *.* '.** ....%%

DETA2 VS SPAN POSIT: f 2 .2 DEGSPANWISE TRAVERSE (I in. from suction side)*R I 6.7ES(plexiglass wall) +Re.6.0E5(steel wall)

- INCHES

Fig. A.65

96

* - . . . . -. --- o -' 4- -* *° - -o o. .. ° *.

Q/ (lower plane) VS SPAN POSIT: 1 2.2 DEGSPANNISE TRAVERSE (U in. from suction side)SRe,-,.7E(plexiglass wall) +ReG.O'S(steel wall)

1.2

.6I

*q2 .6

:.'

"I.4

-.. .

V T INCHES

Fig. A.66

APOA (lower plane, VS SPAN POSIT: I - 2.2 DEGSPRNWISE TRAVERSE (I in. from suction side)*Re-6.?ES(plexiglass wall) *Ren6.3E5(steel wall)

.5I

.3

.t

-. 3

INCHES

1g. A.67 97

APt/! (lower plane) VS SPAN POSIT: t - 2.2 DGSPANNISE TRAVERSE (I tn. from suction side)*Re 6.E?5(plexiglass wall) Re-S.0E5(steel wall)

T INCHES '

Fig. A. 68

" X"R (lower plane) VS SPAIN POSIT: -2.2 DEGspAINNr TRRqVERSE (I in. from suction side)m~~Re-6.7E5(plexlglass wall) +Re-6.0E'5(steel wall I)

1.4

,.4

7 ?INCHESSFig. A.69 98

4, __________,_____,_____,________',____,__,_""__.,,'.-----.---_. -

BIETRI VS SPAN POSIT: 1 3.2 DEGSPANI4ISE TRRVERSE (I on. from suction side)*R 6j,.70E(plexiglass wall) Re-G.OErStsleel wall)

°27 -

-39

-41

-43

-45

T INCHE9

Fig. A.70

'I

1 99

'26,

Q41 (lower plane) VS 9'09 POSIT: 2 ,.2 DEGBLADC TO BLADE TRAVERSE (mtdspan)*Re,,$.lE53(pta~glass wall) *Re,6.0E(steel wall)

t.4

t.0

.4

* ,4

f. . . . _ . . . . .

I NCHES

Fig. A. 71

6Peo/Q (lower plane) VS 9TO POSIT: i " Z.2 DEGBLADE TO BLADE TRAVERSE (midspan)* Re=,6.?7(plexlglaus wall) +Re,,6.0 5(steel wall)

.5I

iTSNHO

Fig. A. 72 100

I,:',,,:_, ,,",,-....".",- :;.:;7 - .. "-. .,: .- .- -": - -. : . : :. : :. : :. , : . 4 .......

&Pt/5 (lower plans) VS STOB POSIT: 1 , 2.2 D]EG

BLRD TO BLADE TRAVERSE (midspan)6R,,.70r(plexiglas$ wualll Re-6.0E5(*teel wall)

.3.

.3

0-A

Si INCHES

Fig. A.73

X/R (lower plane) VS BTOB POSITt I - 2.2 DEGBLADE TO BLADE TRAVERSE (midepan)*Re-6.7ES(plexfglass wall) +Ro-G.OE5(steel wall)

1.4

1.2

t t.6

101

BETRI VS BTOB POSIT: I - 3.2 DEGBLADE TO BLRDE TRAVERSE (medspan)*Rei6.ME(plexiglass wall) RE-6.0ES(steel wall)

JA

-43

-451

1' ' INCHES

Fig. A.75

102

SQI (upper plane) VS SPAN POSITt - 5.6 DEGSPANWISE TRAVERSE tl In. from suction side)*Re 7.?ES +Re - 5.2E5

Fig A. 7. 6

A'S/ (uprpae.SSANPST6 . E

-.3

INCHES

Fig. A. 76

APs'O (upper plane) VS SPAN POSIT: I - 5.6 EGSPAI::SE TRAVERSE (1 in. from suction side)

- .71"5 +Re - 5.2E5

.1

4'. --.3TCH.

'Pi: ig. A.77 103

S..,

Apt/4 (upper plane) VS SPAN POSIT" I 5.6 DEGSPANWIS TRAVERSE k! In. from suction side)

- ?.E5 +Re 5.2E5

-A .1 prpae SSANPST . E

'a

,.1

1 .2

I - - I i ,I I

idn

6#6 INCHES

Fig. A.78

• X/X (upper plane) VS SPRN POSIT: I - 5.6 DEG~SPRNNISE TRAVERSE (1 in. from suction side)

rarea - 7ES +Re - 5,2E5

1.4

N .6

2.4

I1 a - S N , S

, ,, . INCHES

." PFig. A. 79..... 104

:, ....-....." ," i.....,"..:."..-........ .. .. . .. . . . . ...... ,.'.. ........ . . . . . .

BETA2 VS SPRN POSIT: i., 5.6 DEGSPANWISE TRRVERSE (I im. from suction side)*Re , ?.70C +Re . 5.2E5

4

'-4

I "NCHES

Fig. A.80

I1

.

~105a'

.. .IJ ] . . . .. .. _ , . . .

O/ri (lower plane) VS SPAN POSIT: 1 5.6 DEGSPAN ISE TRAVERSE (I in. from suction side)aRe - ?.?ES +Re - 5.2E5

1.4

..4

.3

* . (

.4

.2

A

INCHES

Fig. A.81

AP.'1 (lower plane) VS SPAN POSIT: I 5.6 DEGSPANNISE TRAVECRSE (1 in. from suc' tion side)aRe - 7.7E5 +Re ,, 5.2E5

.5

.3

.1

",S I I I I I-i -

T 9 • INCHES

' Fig. A.8210

'. 106

APt-l (lower plane) VS SPAN POSIT: 1 5 .6 BEGSPANWISE TRAVERSE (I tn. from suction side)

7R .7ES +Re 5 .2E5

%I2

INHE

Fig. A.83

XX (lower plane) VS SPAN POSIT: i 5.6 DEGSPANNISE TRAVERSE (I in. from suction side)

a *Re 7.7E5 +Re 5.2E5

1.4

INCHE

Fi. A8 0

BETA! VS. SPRN POSIT: - .6 DEGSPANWISE TRAVERSE (I in. from suction side)

* ?.7£5 ES c - 5.2E

II

-32

-41t

-43

4-4

-47CI

T INCHES

Fig. A.85

108

JI4'

-*,*

a/c slower plans) VS BTOB POSZT: 1 5.6 DEG

BLADE TO BLADE TRAVERSE tmidspan)*Re - ?.?E5 +Re - 5.2E5

1.4

.4

.INCHES

Fig. A.86

PS6' (lowe, plane) VS DTOB POSIT: 1 = 5.6 DEGBLADE TO BLADE TRAVERSE (midspan)*Re - 7.?E5 +Re 5.2E5

Fi.A

.109

- . -a-.I A'.-87-

-109

1

l - | , . , , +, + . r . . +. . , . " . + -.. 3 ,

Apt./d (lower plane) VS STOB POSITt I - 5.6 DEGBLADE TO BLADE TRAVERSE (midspan)*Re 7.7ES R.. 5.2E5

,. 1

Fig. A.88

X/ (lower plane) VS BTOB POSIT: I - 5.6 DEGBLADE TO BLADE TRAVERSE (midspan)*Re - 7.7E5 +Re- 5.2E5

1.4

*1.2

.3

.$

.4

.2

:'p.. Fig. A.89:," 110

..

-:,. .., ..'..: *,, .. .....,,..-. .......-.... ,-.. ... .-.. .*. . . *.. ..-. . . . .. ... . .. .... . . .. . . . .. -. .. .. . . .

-. KTRI VS BTOB POSIT: o 5.6 DEGBLADE To BLADE TRAVERSE (m~dspen)

OR 7.?E0 *Re -5.2ES

-4W

-43

-45

-471-4, I INCHES

Fig. A.90

V4 (ppr p~lan) VS SPAN POSIT 8 .9 DEG

*eg ? .?E5 +Re 4*S55

8.4

.4

L INCHES

Fig. A.91

IdIeg/5 (upper plane) VS SPAN POSIT: 8 .8 DEGSPANWISE TRAVERSE (1 in. from suction side)

7 .7E5 +Re -4.3E5

.3

.11

-.*T .# .. rL .''a i i "° . -s-- -- -

*1S.7

APt d (upper plane) VS SPAN POSIT: i , 9.8 DEGSPANUISE TRAVERSE (I in. from suction side)

7 ?,.7E5 +Re , 4.9E5

.A2

%%

C.,. -,t1

-.3

INCHES

Fig. A.93

XAZ (upper plane) VS SPAN POSIT: I - 8.8 DEG

;' SPRNNISE TRAVERSE (I in. from suction side)*Re - ?.7E5 +Re - 4.9ES

1.4

INCHES

113

/ t ; 'r ° i -,- " '

' J -, " ". ".'.'.' ' ' ,.... .' " . ."."'-" .' -. L".- "-

BETR2 VS SPRN POSIT: t - 8.8 DEGSPANW4ISE TRAVERSE (1 in. from suction side)

- 7.7E5 +Re 4.SES

-3.4..*°

INCHES

Fig. A.95

4114

:4

" " . .. '.S--.- ' ''' - . ' . '' - .' . . - -. + -. . ." -,-. "+ .- '' -. -

9, f* % 9,* .5%•*4o% . .• .•. • . -.- .+ - : - - ,

QI4 (lower plane) VS SPAN POSIT: I , 6.6 DEGSPANNISE TRRVERSE (I in. from suction side)*Re e ?.1ES +Re - 4.9E5

1.4

1.2

INCHES

Fig. A.96

APe/6 (lower plane) VS SPAN POSIT: I - . EGSPRNNISE TRAVERSE (I in. from suction side)

- ?.7E5 +Re - 4.SE5

.2

.1

-. A1

-.3

SLN* INCHES

Fig. A.9711

* " *. " . - % . . '. " " . . . *. . .'. . . .. .. -. '.." . .. . " . '"". "'. . " . . .5. " "

APt/d (lower plane) VS SPAN POSIT: * 8.8 DEGSPANWISE TRAVERSE (I in. from suction side)*R - ?.?E5 +Re - 4.9SC5

.. ,.3 -.

.-. 3

INCHES

Fig. A.98

X/X - (lower plane) VS SPAN POSIT: I - 8.8 DEGSPRNWISE TRAVERSE (I in. from suction side)*Re - ?.TES +Re - 4.9E5

1.4

1.2

.. 4

.6

.4,

.2

INCHES

L Fig. A.99 116

6 1w l l . . '. .4. * .. * . ..... - = ... . " - .116

DETRI VS SPRN POSIT: I - .8 DEGSPANNISE TRAVERSE (I in. from suction side)*Re ?.75 +Re , 4.9E5

-44

-40

4, -4u

,4" - - - '-'- '1

INCHES

- Fig. A. 100

1176I 1L *

0,'/ (lower plane) VS STOB POSIT: I , 8.8 DEGDLRDE TO BLADE TRAVERSE (midspan)*Re 7 7.?E5 +Re 4.9E5

1.4

s.a

.6

'P ? "INCHES

Fig. A. 101

Aps/3 (lower plane) VS STOB POSIT: -8.8 DEG

BLADE TO BLADE TRAVERSE (midspan)

*tRe= 7.7ES +Re -4.9E5

.$

.4 .

e.4

* '? INCHES

Fig. A. 102

.11

Apt/d (lower~ plane) VS BTOB POSIT: 0.U DEGSLADE TO BLADE TRAVERSE (midspan)

a7.7E5 +Re -4.9E5

1.4

-1

1 * INCHOS

Fig. A.104 1

I..:

BETRI VS DTOB POSIT: 0 ,.8 DEGBLADE TO BLADE TRAVERSE (midspan)*Re - 7.E5 +Re - 4.9E5

-40

-.3a

-44

-46

"I , -46, i I. i

9'- INCHES

Fig. A.105

.12a'

p..

m -.........." "' .' .•... .• . ' : .. .. . . . . . . . . . . . . . . . . . . . . ..

1 -12.50 0 Re =611,000

uper wall static lower wall Statizpressure distribution pressure distrbuition..

Fig. A.106

'S 01 -12.39 ,Re =535,000

I alsai oe alattt

Fig. A.107

121

-. 80 Re 626,000

Fig. A. 108

I=-9.23, Re -502,000

Fig. A.109122

A.-5.81 0 Re =560,000

Fig. A.110

j ~.557,R 503,000

Fig. A.111123

1 -1.59 0 Re =643,000

~ .;~ lower wall s tatic~ tppr v11staicpressuire distribution.

pressure distribution

Fig. A.112

1 = 1 . 5 0 0 Re 525,000

IOW..a,. ~ *..-

. . . . . . . . . . . . . . . . . . . . .

pesr diti.to

Fig. A.113

124

i2.27 0Re 676,000

Fig. A.114

I .23 0 Re 604,000

44%

val taf

. . . . . . . ..... . . . . . . . . . .

Fig. A.115125

5.99*g0, Re -771,000

Fig. A.116

j 50,R 518,000

5.5 R

r-t

pressurig.dA.117

126

I-8.79 0 Re -768,000

upper wall static

resr isrbto

Fig. A.118

12

0/a (upper planei VS BTOB POSIT: - -12.5 DEGBLADE TO BLADE TRAVERSE (mtdspan).Re * 6.1E5 +Re - 5.3E5

1.4

t.1)

J'i..4

..

9'9''INCH S

Fig. A. 120

APs/Q (upper plane) VS BTOB POSIT: f t2.5 DEG

BLADE TO BLADE TRAVERSE (midspan)* 6.1E5 +Re - 5.3E5

,%F

ep t , I NCHOS =

Fig. A. 121 128

-.- 1

_~~ ~ ~ ~~~~ W- , . . /.% .,%..%" o%" ./

*.of

APt/d (upper plane' VS BTOB POSIT: I . -12.5 DEG

BLADE TO BLADE TRAVERSE (mdspn)

" "* 6.IE5 +Re 5.3E5

.3

.-.

'INCHES

X,3z (upper plane) VS BTOB POSIT: a--12.5 DEGBLADE TO BLADE TRAVERSE (midspan)*Re G .1ES +oRe -5.3E5

1.4 -

1.2

-

.4

a ka

, ;23 'INCHES '

Fig. A.12

"-12 XT

it-BRET L ETRVRE ri~a

ICTA2 VS BTOB POSIT: -1-2.5 DEGBLADE TO BLADC TRAVERSE (midspan)oR1 S.Ics +Re 5.3CS

0130

Q/d (upper plane' VS BTOB P')SIT: - -9.5 DEGBLADE TO BLADE TRAVERSE (midspan)*Re - 6.3C5 +Re - 5.0E5

1.4

* 1.2

~ 4 7 INCHES

Fig. A. 125

APS/3 (upper plane) VS BTOB POSIT: |--9.5 DEG

BLADE TO BLADE TRAVERSE (midspan)*Re - .3E5 +Re -00 E

.5

• .6

,.4

.. ;

9' ? 'INCHES

Fig. A.126 131

• ...... .

. .. . .. ...... :.. .. . .. .~ ~-.T: & -: . .. . . .

iPt/5 (upper plane) VS STOB POSIT: - -9.5 BEG

BLADE TO BLADE TRVERSE (mtdspan)

*Re - 6.3E5 +Re , 5.0E5.5

.3l

:'-.*

' -. 3

'.... ... .-.s

4' INCHES

Fig. A.127

X/R (upper plane) VS BTOB POSIT: - -9.5 DEG,.-. BLADE TO BLADE TRAVERSE (midspan)

*Re - 6.3E5 +Re - 5.0E5

1.4

1.2

~..

...

4I' 'INCHES

Fig. A.128 132

-132

BETR2 VS DTOB POSIT: t-9.S5 DEGBLADE TO BLADE TRAVERSE (midspan)*Re 6.3E5 +-Re -5.0E5

-- pl

66,

OAS (upper plane) VS BTOB POSIT: t -5.6 DEGBLADE TO BLADE TRAVERSE tmidspan)*Re S S.ES +Re - 5.OE5

.'

.8

.4!

9' INCHES

Fig. A.130

APs/3 - (upper plane) VS BTOB POSIT: - -5.6 DEGBLADE TO BLADE TRAVERSE (midspan)*Re - 5.6E5 +Re - 5.0 5.5

.3"

-.4

F'ig. A.131134

S

. .. . .

d .", . ..,/ ,. ,_ . .".,., - , , -. , .-. . .,, - ..,. - . . . . ..

,.; , . ,. ..-,,= . l.L -~

.. , , .i .' ,,, ,.- - . . . " - ".'.'""'

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

APt/o (upper plane) VS STOB POSIT: - -5.6 DEGBLADE TO BLADE TRAVEPSE (midspan)*Re 5.6E5 +Re - S.0E5

.3

- -.:

INCHES

Fig. A.132

X/7 (upper plane) VS BTOB POSIT: - -5.6 DEGBLADE TO BLADE TRAVERSE (midspan)*Re - 5.6E5 +Re - 5.0E5

1.4

1.2

.4

.2

9, ' 'INCHES

a:' #ig A.133'135

BETR2 VS STOB POSIT: -- 5.6 DEGBLADE TO BLADE TRAVERSE (midspan)

* *5.6E5 +Re 5.0E5

-33

o/d tupper plane) VS BOB POSITt - -1.5 DEGBLADE TO BLADE TRAVERSE (mldspan)*Re - 6.4E5 +Re -5,2E

I* 1.4

. . 135

*R .4

..

S " ? INCHES

Fig. A.135

tPs/1 (upper plane) VS 7TOB POSIT: I - -1.5 DEGBLADE TO BLADE TRAVERSE (midspan)~R. - B.4E'5 +Re - 5.2E:5

.$

I .2

.- I

~-.I,

I,.| • , I I NI I IS

"'Fig. A. 13613

" 137.

APt./d (upper plane) VS DTOB POSIT: I - -1.5 DEGBLADE TO BLADE TRAVERSE (ml dspan)*Re * 6.4E5 +Re. 5.2E5

.3

9' 9 ' INCHES

Fig. A.137

X/X (upper plane) vs BTOB POSIT: - -1.5 DEGBLADE TO BLADE TRAVERSE (midspan)*Re - 6.4E5 +Re - 5.2E5

1.4

12

.$

.4

9' ',;, ' TINCHES i

Fig. A.138 138pt*b.***l* . ..

BETR2 VS BTOB POSIT: 1 -1.5 DEG'BLADE TO BLADE TRAVERSE 'midspan)

6.4E5 +Re - 5.2E5

19

"o I

-3

%-°

.- Fig. A. 139

.0,,

139

4i.4

- ; , , * _' '_ " , 44, w '. . '" . J ', , '. - , '- . ', . '. '. .' , . ' ", ' . ., ' . ' ' . -. .- "

n0/ (upper plane) VS BTOB POSIT: - 2.2 DEGBLADE TO BLADE TRAVERSE (midspan)*Re,6.;'05(plexlglass wall) +R..6.0ES(stol wall)

.4

.2

9'~ 'INCHES

Fig. A.140

APs/ (upper plane) VS 9TOB POSIT: 2.2 DEGBLADE TO BLADE TRAVERSE (mtdspan)*Rj 6.?E5(plexiglass wall) +Re-6.0ES(steel wall)

.3

-.3

o..

i' 9 ' 'INCHES

Fig. A.141140

A Pt (upper plano) VS 9TOB POSIT: 3 * 2.2 DEGI BLADE TO BLADE TRAVERSE tmidspan)*R 6.7 plexiglass wall) ReS.OES(steel wall)

.. 3

;%'9

Fig. A. 142

'."X/X (upper plIan@) VS BTOB POSIT: !-2.2 DEG' BLADE TO BLADE: TRRVERSE (midspan)

*IRe-(;.27(plexiglass wall) +Re,,6.00:'($seel wall)1.4

1.2

...

--.3'INCHOS

Fig. A.1434141

. ... ,5" ,.'., . . : . . . . . . . . . . - ' : .

BCTA2 VS 9TOB POSIT: 1- .2 DEGBLADE TO BLADE TRRVERSE (midspan)

6Ro,,.70(pl qlas I wall) +R.W6.0E(st.el wall)

*1 4

-:-a

N -3

. --- I . . . . ,NC, , , ,S

-.. Fig. A. 144

.4

.4.

,. 142

S. %

BILAD:E: TO BLADE TRAVERSE (midspan)': *Re 7, .7E5 +Re ,,5 .2 rS

1.4

*"-q,

1.2

q 7 INCHIFS

Fig. A.145

QPS/6 (upper plane) VS BTOB POSIT: * 5.6 DEGBLADE TO BLRDE TRAVERSE (midspan)

. *Re 7 7.7E5 +Re - 5.2ES

mI

e.3

Fig. A. 146 143

.4

_*..-,-,. ,..,- . .- - . ,..--- ., .. -.-- - . . .- -.- -. .

APt'o (upper plane) VS 9TOB POSIT: - 5.6 DEGBLADE TO BLADE TRAVERSE (midspan)

S,?.5 +Re , 5.2E5

II

':

-. INCHES

Fig. A. 147

X;X (upper plane) VS BTOB POSIT: i -5.6 DEG

BLADE TO BLADE TRAVERSE (midspan)*Re -7.705 +Re -5.2ES

.4

4.4

.11

.44

-i

A.148144

- -. -,.. , . . . .- .- - .- .- .- - . -. 3-t ., . ., , .-. • . . . .

BETRZ VS BTOB POSIT: I - 5.6 DEGBLADC TO BLADE TRAVERSE (midspan)

* ?cES +Re - 5.2E5

S

4 4

*I I I I I I I I

-3 wii

Fig. A. 149

14

145

,1;'+,' ,)." '"," - ." , ., ." -"." -. .' -. ' ,-.-.',. ' .' . +- ,. -' . ,". -- . , -. ' . + ."2 -:- : : :- -: -'.-

G'Q (upper plane) VS BTOB POSIT: 0 6.9 DEGBLADE TO BLADE TRAVERSE (midspan)*Re 7 .7lE5 +Re , 4.9E5

1.4

.2

.4-

9L INCHES

Fig. A.150

APe/t (upper plane) VS BTOB POSIT: - 9.8 DEGBLADE TO BLADE TRRVERSE (midspan)*Re - 7.7E5 + Re - 4.9E5

.. 4

-S.,,

; ? ' T INCHE'S

i. A.15114

'44

}.2,:N.,. 4.,' ,2 ; .T2 .- !. .-..i.i . :" - "' "' ' " " • • " " .. . -"' -

APt/i (upper plane) VS 9708 POSIT; 8.9 DEGBLADE TO BLADE TRAVERSE kmidspan)

*Re 7 .?E5 +Re 4.9E5

41 7 T INCHES L--A

Fig. A.152

X/X (upper plane) VS 9708 POSIT: 8 .8 DEGBLADE TO BLADE TRAVERSE (midspan)*Re ?.?E5 +Re 4.9E5

.'1.4

*7 1.

. ..4 . . . .

Pig. A.153 'INCHES

147

BETR2 VS BTOB POSIT: 8 -. 8 DEGBLRDC TO BLRDE TRVERSE (midspan*Re 7.70. eRg - 4.9E5

*,--I

.

s-s

9' TINCHES

Fig. A.154

148

Cp VS X/C: i - -12.50Re = 530,000R = Right Adjacent BladeL = Left Adjacent Blade0 = Pressure Side, Center Blade

1*0 Suction Side, Center Blade

o 0 00 X,'CA0 R 0.2 0.4 0.6 0.8 A 1 0

0oL , A A A

-1.0 0

100

Fig. A.155

l0

Cp VS X/C: i =-12.5°

Re = 610,000

1.0-

'Op

F^.. . o, .op. /

0 A 0 6

A IAA

Fig. A.156149

' 2 "'"" " " ";-".'"-"- -'' ' " "" . - '""-" -:.. . . . .,. ,....,- ....-. ... o. ,. ..:. . .. ... .. :., .--..- : .-. :: -. : . .-.-. .-.-. . -..

Cp VS X/C: i = -9.50Re = 500,000R = Right Adjacent BladeL = Left Adjacent Blade

1.0- 0 = Pressure Side, Center BladeA4 Suction. Side- Center Blade

CL A n .0 ' " X/C!501a 6iO O0'2 0.4 0.6 0.8 AlI. 0

"" 4

Fig. A.157

* .Cp VS X/C: i = -9.50Re = 630,000

1.0-

%A

9-Ci0 0 0O0f 0 /

o(.2 0.4 0.6 0.8

-1.0-0 A00_ _

. 0

Fig. A. 158150

Cp VS X/C: i = -5.60Re = 500,000R = Right Adjacent Blade

.. L = Left Adjacent Blade• .0 -Pressure Side, Center Blade1.0- = Suction Side, Center Blade

-~~ 0Q 0- 0 I o 0 0 Q0O. O- -O O O O Sx/c

U . 0.2 0.4 0.6 0.f aA 10

Fig. A. 159

.. o " z .

Cp VS X/C: i =-5.60

Re - 560,000

1.0-

"'0 0 " 0 0 r/c

00.2 0.4 .6 0. 1 Al.0

A A A A

Fig. A. 160

151

[ . .... .E b ;:-'. .. .-. . . ?-... 9 *.\'~% . .. .... . ." "-. .- - , -' - - "~ - - • " ...Il i l ~ mB am d ~ ~ m ,j. .di a *',... . *._ . . * .*, ..-. .. ,,- . • . ,. . .-. , . . .. ,. .

Cp VS X/C: i -1.50Re = 520,000R = Right Adjacent BladeL = Left Adjacent Blade

1.0) 0 = Pressure side, Center BladeA= Suc-tion- Side, Center Blade

eQ0 ) 0 00 o o 0 o 0 0k o ox/c000.2 0.4 0.6 A A 10

Fig. A. 161

Cp VS X/C: i -15=Re = 640,000

1.0

oof -0 A' 0/C0 00"J 0.2 0 .4 0.6 1 .D

AA

Fig. A.162152

Cp vs X/C: i 2.2°Re = 600,000R = Right Adjacent BladeL = Left Adjacent Blade

1.0- 0 = Pressure Side, Center Blade= Suction Side, Center Blade

co 0 0 00 0 0 0 a OR 00o

0.2 0.4 ,o.6, A 0.8 1.0

At

-1.0-

Fig. A. 163

Cp VS X/C: i =2.20Re = 670,000

1.0-

000 0 0 0 0 0 0

0 0.2 O.4 0.6

A A

-1.0"

Fig. A. 164153

*Cp VS X/C: 1 5.6 0Re - 520,000R = Right Adjacent BladeL - Left Adjacent Blade

10 0 = Pressure Side, Center Blade= Suction Side, Center Blade

CO 0 o 0 0 0 00 10 0 0

A A A A x/CU 0.2 0.4 0.6 A 0. 1.0

* Fig. A. 165

Cp VS X/C: i 5.6 0

Re = 770,000

1.0-

CO04~ 0 0 0 0 0 0 0 0Q000

00.2 0.4 0 *A 0.8 1.0

Fig. A.166154

Cp VS X/C: i = 8.80Re = 490,000R = Right Adjacent BladeL = Left Adjacent Blade

1.0- 0 = Pressure Side, Center BladeA = Suction Side, Center Blade

0 0 0 0 a 0 go 0 0 ,

A 41 A'5/C

.4

Fig. A. 167

. Cp VS X/C: i = 8.80

1.0.]Re = 770,000

"1.0-

0 0 0 0 0 0 0 0 0

0 0.2 0.4 A8 1.0

A ~A~0.8 I

-1.0-

A

A

Fig. A. 168155

, • .. - . , .. S. .. . . . .. . . . . --. . . . .

-

-

i- tan 1 + tan o2 )

1 = -1.2.5 d 2gres

Re = 611,000

* o.. -f

-I.L-

.*1

";; BI

Fig. A.169

156

,, .% ...,..-. -.- . .-... -... -. .-... . .... .. .. -. ..

* . . . - .

.. - tan- 1 (tan +tan

i = -12.3.9degreesRe = 535,000

1..0

II

I. 0 t \ Cfm.,.

CfB

,t

!NwN

i Fig. A. 170

157

FI. y

8a~=tan(.~ 23.

j=-9.58degreus

Re 626,000

1.0

1.0

Fi.A.7

* 158

4

+ tan 2tan- 2

i ==-9.23degreesRe 503,000

1.0

C fm1.0CB

4%

-4i

*'.

".

;- Fig. A. 172

159

.. . -.. ... ,

= tan- (LIIK Lan0 2 )2

i = -5.8lcdegrees -5

Re = 560,000

-1 0

,1.0

-f.Q.CfB

, -- C

Iy

Fig. A.173

160

.9

. .. . . - 7 - - -- -- r 7 . ,-. - • Y -. . - .

= a n + tan

o" - i = -5.57degreesRe = 5G3,000() "

1.00,:-1.1.

II

I': ." I

:•, CfB

Iq,

Fig. A.174

161

-~ ~~ aa~ tn~ o tank

i -1.59degrees

Re 643,000

1.0 jf

1 . c f

Fi.A.7

16

0,10 tan tfl~ + tano 2 \

1 -1.5 degreesRe =525,%000

.0

L

'I-

df

.1 f-.

Fig A..0

16

=tan

i =2.27 degcesRe 676,000)

-1.00

Fig A17

16

40 1.

. .S~ =. .. .. . .. .-

7tana1 + tn

i 2.23degreCsRe 604,000

iezj/

MI.16

n- tan + tanL

L; 5.99degrevsRe = 771,000

.0 --

/

CfB

1

r

Fig. A. 179.'--

166

It s. Sd9O2~

taLn81 + tanO~3__ II I I I I I I

i = 5.54degreesRe = 518,000

I.O

167

I

r.4

p.

I Fig. A. 180

167

'---~~~~~~.. ...--; .-.:..-,:..-.-.,........... .... ....... ... ;..........: .-.. :._...

tan tan. I +ta~i2

2I

8.79degree±s

=768,000

I..o

I -lfB9.0m

Fi.A 8

I6%

t 1ano + tan 0

8.76degreesRe 493,000

1.0

II / -1.0

-II

---.- C f B

l , -1.0

*y

•I

Fig. A.182

169

-' . ,'..',I : .' '.,N .- , ,''.-.'''.,''' . •" " ""'- -'-. .-" -. . ..--".'" '." ". ..""'. " '""""," "-"|

APPENDIX B

A. STORAGE OF EXPERIMENTAL DATA

1. Storage Devices

a. Cassette Tape

All raw and reduced data acquired during the

current study via the HP-3052A acsQiSition-system were stored

on cassette tape. Table B.1 summarizes the raw data file-

names according to inlet air angle and cassette tape number.

Tapes were filed with the Director, Turbopropulsion Laboratory,

Naval Postgraduate School, Monterey, Ca.

2 b. File Naming Scheme

All files were named using a six character alpha-

numeric code that allowed the investigator to immediately

identify stored data in any file and on what date it was

recorded. Here is how the code works.

1st character: Indicates the type of traverse conducted

("B"lade-to blade, "S"panwise or "I"nstrumented blade data).

2nd character: Indicates the "state" of the data set

(ra"W", re"D"uced or "F"inal).

3rd character: Indicates which probe was involved in

taking the data set, either "U"pper or "L"ower. No letter

here means that the probes acquired data simultaneously.

4th, 5th and 6th characters: These numbers indicate

the julian date in 1983 that the test run was conducted.

' " 1704 -" .. , - + .: + -. -. .. . -. .-..... .. . . . . . . . . . . . . . . . . . . . . .

* For example, a file containing data from a raw blade-to-

blade survey using the upper probe conducted on January 15,

1983 would be named BWU015. When upper and lower probes

were traversed together the "U"pper or "L"ower character

was dropped and the entire julian date filled the remaining

four spaces. For example, if the upper and lower probes

were both traversed in the blade-to-blade direction on Jan-

uary 15, 1983, the resultant raw data filename would be named

BW3015.

Sometimes, in order to completely describe a

reduced data file, parts of the julian date had to be sacri-

ficed. Raw data from a spanwise survey of the upper probe

one inch from the suction side of the centermost instrumented

blade conducted on February 1, 1983 would be named SWSU32

("S"panwise, ra"W", 1 inch from "S"uction side, "U"pper probe,

julian date 30"32"). It should be noted that when spanwise

surveys were conducted, both probes moved together. To pre-

clude interference of the lower probe wake with the upper

probe measurements, the upper probe was place 1 inch from

the suction side, and the lower probe was placed 1 inch from

the pressure side of the center blade. The composite data

file was named with the upper probe in mind, however, even

though the lower probe was stationed at a different blade-

to-blade position. For example, a raw spanwise survey named

SWS077 would show that the upper probe was 1 inch from the

171

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

suction side of the center blade, while the lower probe was

1 inch from the pressure side.

Some specialized filenames reflect the type of

data stored. Referencing constants calculated for a test

conducted on February 15, 1983 would be named RC046A or

RC046B where the "A" would mean that the referencing constants

were calculated using the plenum total pressure and the lower

wall static pressure as reference total and static pressures.

"B" would indicate that the reference total and static pres-

sures were supplied by the pitot-static probe.

2. Thesis Logbook

A chronological log of the present study with notes

and printouts of all data files was deposited with the Director,

Turbopropulsion Laboratory.

B. PROBE CALIBRATION DATA

A folder documenting the calibration of United Sensor

probes serial number 847-1 and 981-2 was filed with the

Laboratory Manager, Turbopropulsion Laboratory. Non-dimensional

velocity (X) and pitch angle (PHI) calibration files for

both probes appear on each data storage cassette tape. Their

names reflect the file type and the applicable probe. For

example, the pitch angle calibration file for the 981 probe

is named "PH1981", the X velocity calibration file for the

847 probe is named "X847", and so on.

172

TABLE B.I. RAW DATA STORAGE

Raw Raw Raw TapeBTOB Spanwise Instrumented NumberData Data Blade Data

24.63 BW3144 SWS144 IW3144 11

24.74 BW3143 SWS143 IW3143 11

27.55 BW3123 SWS123 IW3123 7SWP123

27.90 BW3125 SWS125 IW3125 7SWP125

31.32 BW3119 SWS119 IW3119 6SWP19

31.56 BW3121 SWS121 IW3121 6SWP123

35.50 BW3138 SWS138 IW3138 10

35.60 BW3139 SWS139 IW139B 10

39.02 BWU077 SWU077 IW077 2BWL077 SWL077

38.68 BW3082 SW3082 IW082B 2

38.34 BW3097 SW3097 IW3097 3

38.89 BW3103 SW3103 IW3103 3

39.28 BW3127 SWS127 IW127B 8SWP127

39.40 BW3128 SWS128 IW3128 8SWP128

39.16 BW3129 SWS129 IW3129 9SWP129

39.36 BW3133 SWS133 IW3133 9

42.66 bw3109 swsi09 iw3109 4. swp109

43.13 BW3110 SI110 IWIIOB 4SWP 110

45.92 BW3114 SWS114 IW3114 5SWP114

45.89 bW3114 swsll4 iw3114 5swp114

173

APPENDIX C

EFFECT OF AVDR ON THE AXIAL FORCE COEFFICIENT DERIVED

FROM PROBE SURVEY DATA.

The notation used in this section follows [Ref. 14] and

[Ref. 15]. The application of the principle of momentum

-, conservation to the control volume shown in Fig. C.1 results

in the following expression for the resultant force on the

control volume;

R% A

.7.

Y ,---,vw -4

Figure C.I. Blade Space Control Volume

174

9, ~Jd!

In the enlarged view of the edge of curved surface a, the

unit outward normal vector, na-, is given by the relation

: CO 01 Ao = - sly% a( (c-2(C-2)

A

and similarly, on surface b, nb is given by

A AA SWO AnYl (C-3)

The elemental area "s" on the curved sidewall of the control

volume can be expressed as the projection of an element of

the base area dS, where

CI%"I dy" c (C-4)

so that,

d4 dxdy (C-5)

Resolving the elemental force, dF , acting on the curved

sidewalls of the control volume into normal and sheer compo-

nents gives_bA . A

J F = -PV d.4 - t d.A(CI (C-6)

Neglecting wall friction, T , the tangential force becomes:

Rvt Vu ,(cA?-V.VU)A

an ~ : i :~~z zSS?2~ VU A (C-7)Sand the axial force becomes:

SR,,,. S V. A - S f.\ JA,

CAI ~ A (C-8)

SL :. 0.

175

* :~ ~ ~ *.**.*.%,* *-.-.-.*.; .-.

Using Eq. (C-2), Eq. (C-3), Eq. (C-5) and Eq. (C-6),

dFL ly 4. CI 4 xS Cos oc)S f= -p cy (c-9)

and similarly,

which represent additional axial force components due to

pressure forces 9n the control volume side walls.

Equation (C-7) is identical to Eq. (B-3) of [Ref. 1] which

was derived for a strictly two-dimensional flow. Substitu-

tion of Eq. (C-9) and Eq. (C-10) into Eq. (C-8) yields,

. d(C-i1)

which contains a fifth axial force term not included by Cina,

and a modification of the first and third terms resulting

from the effects of AVDR.

Since the precise function p(y) is not known from

station 1 to station 2, a linear change of pressure with

y was assumed as depicted in Fig. C.2.

(D

Figure C.2. Assumed Pressure Function

The fifth term of Eq. (C-il) can then be approximated as:

-~~~ 2S (j(V~O f~ ][v. (C-12)176

" ,; ; . . . . .. . . . . . . . . .

Assuming that

.. S. AVDR - j ,AVDR d,,(C-13)

substitution of Eq. (C-13) and Eq. (C-12) into Eq. (C-I1)

yields, & a +3? '0 (,AV D )dx Jrva

S-f dx-(C- 14)

Equation (C-14) was reduced to force coefficient form

using procedures explained in [Ref. 1] and then incorporated

into existing software documented in (Ref. 81.

17

177

LIST OF REFERENCES

1. Cina, Frank S., Subsonic Cascade Wind Tunnel TestsUsing a Compressor Configuration of DCA Blades, M. S.Thesis: Naval Postgraduate School, Monterey, Ca., 1981.

2. Molloy, William D., Preliminary Measurements and CodeCalculations of Flow Through a Cascade of DCA Bladingat a Solidity of 1.67, M. S. Thesis, Naval PostgraduateSchool, Monterey, Ca., 1982.

3. Rose, Charles C., ard Guttormson, Darold L., Installationand Test of a Rectilinear Cascade, M. S. Thesis, NavalPostgraduate School, Monterey, Ca., 1964.

4. Bartocci, J. E., An investigation of the Flow Conditionsat the Lower Measurement Plane, and in the Plenum Chamberof the Rectilinear Cascade Test Facility, M. S. Thesis,Naval Postgraduate School, Monterey, Ca., 1966.

5. Moebius, Richard C., Analysis and Testing to Improvethe Flow from the Plenum of a Subsonic Cascade WindTunnel, M. S. Thesis, Naval Postgraduate School, Monterey,Ca., 1980.

6. McGuire, Alan G., Determination of Boundary Layer Transi-tion and Separation on Compressor Blades in a LargeSubsonic Cascade, M. S. Thesis, Naval Postgraduate School,Monterey, Ca., 1983.

7. 3052A System Library (9845A), Hewlett Packard Company,, 1978.

8. Turbopropulsion Laboratory, Naval Postgraduate School,Technical Note 80-03 (revised), Data Aquisition Programsfor the Subsonic Cascade Wind Tunnel, by D. A. Duval,1980 (revised by S. J. Himes, 1983).

9. NASA Report SP-36, Aerodynamic Design of Axial FlowCompressors, edited by Irving A. Johnson and RobertA. Bullock, 1965.

10. NASA Report 1016, Effect of Tunnel Configuration andTesting Technique on Cascade Performance, by John R.Erwin and James C. Emery, 1951.

178

, - .

11. Duval, David A., Evaluation of a Subsonic Cascade WindTunnel for Compressor Blade Testing, M. S. Thesis, NavalPostgraduate School, Monterey, Ca., 1980.

12. Contractor Report, Procedure and Computer Program forApproximation of Data (with Application to MultipleSensor Probes), by H. Zebner, August 1980.

13. Turbopropulsion Laboratory, Naval Postgraduate School,Technical Note 82-03, Computer Software for the Cali-bration of Pneumatic and Termperature Probes, by F.Neuhoff, September 1982.

14. Vavra, Michael H., Aero-Thermodynamics and Flow inTurbomachines, Krieger, 1974.

15. Class Notes, Aerotharmodynamics and Design of Turbo-machines, R. P. Shreeve, Professor, October 1982.

1

179

INITIAL DISTRIBUTION LIST

No. Copies

1. Defense Technical Information Center 2CameronStationAlexandria, VA 22314

2. Library, Code 0142 2Naval Postgraduate SchoolMonterey, CA 93940

3. Department Chairman, Code 67Department of AeronauticsNaval Postgraduate SchoolMonterey, CA 93940

4. Director, Turbopropulsion Laboratory, Code 67Sf 1Department of AeronauticsNaval Postgraduate SchoolMonterey, CA 93940

5. Mr. George DerderianNaval Air Systems CommandCode AIR-310EDepartment of the NavyWashington, D. C. 20360

6. Dr. A. D. WoodOffice of Naval ResearchEastern/Central Regional Office666 Summer StreetBoston, MA 02210

7. Chief, Fan and Compressor Branch(Attn: Nelson Sanger)NASA Lewis Research CenterMail Stop 5-921000 Brookpark RoadCleveland, OH 44135

8. LCDR S. J. Himes, USN4681 Revere RoadVirginia Beach, VA 23456

9. Turbopropulsion Laboratory Code 67 10Naval Postgraduate SchoolMonterey, CA 93940

180