Copy Number

NASA CR 182125 _

NASA

LARGE SCALE ADVANCED PROP-FAN

(LAP)HIGH SPEED WIND TUNNEL TEST REPORT

By: William A. Campbell

Peter J. Arseneaux

Harry S. Wainauski

(NAqA-CR-I_212_) LARG_-SCAL_ AnVANCED

PROP-FAN (LAP) NIGH SPEFO WIND TUNNEL TFST

REPORT (k_amilton Standard1) 199 p CSCL 2IF

NQO-iOO_

UnclJs

G3/O? 0237147

" ;-_% --HA MiLTON_DARD DWiSlON --

:= - _ - - _--:', :_:_= Prepared for

......... National Aeronautics: a==n_d_Splce Adminis-trat_fon --_--" _NA _SA_gWis Resea__rch centei .......

- __ .... -=_ __--__Contract NAS3--23051

!

J

I

NASA CR 182125

whole or in part. Date for general release July, 1989.

NASA

LARGE SCALE ADVANCED PROP-FAN

(LAP)

HIGH SPEED WIND TUNNEL TEST REPORT

By: William A. CampbellPeter J. Arseneaux

Harry S, Wainauski

HAMILTON STANDARD DIVISIONUNITED TECHNOLOGIES CORPORATION

Prepared for

National Aeronautics and Space AdministrationNASA-Lewis Research Center

Contract NAS3-23051

AiJ

SECTION

l.O

2.0

3.0

4.0

5.0

6.0

7.0

CONTENTS

ABSTRACT

SUMMARY

INTRODUCTION

PROP-FAN DESCRIPTION

4.l General Description4.2 SR-7L Blade

4.3 Pitch Change Actuator4.4 Pltch Control

4.5 Hub and Blade Retention

4.6 Spinner

INSTRUMENTATION DESCRIPTION

5.1 General Description

5.1.1

5.1.2

Electronlc Data Acqulsltlon System

Prop-Fan Diagnostic MonitoringInstrumentation

5.25.3

Steady Pressure Measurement System

Unsteady Pressure Measurement System

TEST FACILITY DESCRIPTION

6.1 Wind Tunnel Description

6.2 Test Rig Description6.3 Test Rig Modifications

SPINNER AND CENTERBODY DRAG MEASUREMENT

7.1 Test Rig Vlbration Survey7.2 Wind Tunnel Corrections

7.3 Test Rig Corrections

7.3.1 Test Objectives?.3.2 Test Procedure7.3.3 D1scusslon and Results

(DAS)

PAGE

1

3

5

11

11

11

1518

1818

23

23

23

27

28

30

31

3131

40

47

47

47

55

5555

63

PRECEDING PAGE BLANK NOT VILMED

iii

SECTION

8.0

9.0

10.0

11.0

12.0

APPENDIX

CONTENTS(Continued)

BLADE STRUCTURAL DYNAMIC EVALUATION

8.1 Test Objectives8.2 Test Procedure8.3 D1scusslon and Results

8.3.1 General Discusslon

8.3.2 Data Reductlon

AERODYNAMIC PERFORMANCE EVALUATION

9.1 Test Objective9.2 Test Procedure

9.3 D1scusslon and Results

9.3.1 Data Reductlon Procedure

9.3.2 Data Presentation

BLADE SURFACE STEADY PRESSURE MEASUREMENT

10.1 Test Objective10.2 Test ProcedureI0.3 D1scusslon and Results

BLADE SURFACE UNSTEADY PRESSURE MEASUREMENT

11.1 Test ObJectlves11.2 Test Procedure

II.3 Discusslon and Results

CONCLUSIONS

12.1 Blade Structural Dynamic Evaluatlon

12.2 Aerodynamic Performance Evaluatlon

12.3 Blade Surface Steady Pressure Measurement

12.4 Blade Surface Unsteady Pressure Measurement

List of Symbols

References

A Chronologlcal History of Test

PAGE

71

71

7177

7784

I05

I05I05106

106

108

125

125125

136

157

157

157

169

175

175176176177

179

183

m_

iv

FIGURE

3-I3-2

3-3

4-I

4-2

4-3

4-4

4-54-6

4-7

4-8

5-I

5-2

5-35-4

6-I

6-2

6-36-4

6-5

6-66-76-86-96-106-116-127-1

7-2

7-3

7-4

7-5

7-6

7-77-87-9

ILLUST TIONS

Prop-Fan Concept

LAP Program Elements

Prop-Fan Test Assessment (PTA) Program Elements

Large-Scale Advanced Prop-FanDesign Requirements and Goals SummaryFeatures of the SR-7L Blade Construction

LAP Pitch Change ActuatorMotor Driven Pitch Control MechanismLAP Control Schematic

SR-7L Hub

SR-7L SpinnerH.S.D. and O.N.E.R.A. Instrumentation List

LAP Instrumentation System Schematic

LAP Blade Pitch Angle Measurement System

Steady Pressure Measurement System

Modane-Avrleux Aerothermodynamlc Test CenterSI-MA Large Atmospheric Wind TunnelSI-MA Wind Tunnel SchematicSI-MA Mind Tunnel Nozzle Inlet to Test Section

Tunnel Driving Power and Reynolds Number as aFunction of Math number

Prop-Fan Installation In Test Sectlon/Chariot No. 3

Prop-Fan Drive System Engines and Gearbox

Prop-Fan Test Rig Power Llmitations

Prop-Fan Drive SystemProp-Fan Test Rig Dr|ve Train "Balance" Schematic

Origlnal Test Rig Configuration

Modlfled Test Rig ConfigurationAxial Mach number Distribution for Freestream

Mach number .354Axial Mach number Distribution for Freestream

Mach number .615

Axial Mach number Dlstributlon for Freestream

Math number .721

Axial Mach number Distribution for FreestreamMach number .754

Axial Mach number Dlstrlbutlon for Freestream

Mach number .788

Ax|al Mach number Dlstrlbutlon for FreestreamMach number .829

Splnner/Body Tare Test

Test Rlg Support Structure for Splnner/Body Tare Test

Prop-Fan Blade Stubs

PAGE

68

1012

1314161719

2021242526

2932333435

3637

383941424344

48

49

50

51

52

5356

5758

V ¸

ILLUSTRATIONS(Continued)

FIGURE

7-10

7-11

7-127-137-147-158-1

8-2

8-38-48-58-68-78-88-9

8-10

8-11

8-12

8-13

8-14

8-15

8-16

8-178-188-199-19-29-39-49-59-69-7

Centerbody Surface Static Pressure Tap Locations(Forward Section)

Centerbody Surface Static Pressure Tap Locations(Mid and Aft Sections)

Spinner Bulkhead Static Pressure Tap Locations

Prop-Fan Test Rig Forces

Spinner Drag Coefficient vs. Corrected Mach number

Centerbody Drag Coefficient vs. Corrected Mach number

Test Rig Support Structure for Structural DynamicEvaluatlon (2 and 4 Blade Configurations)

Test Rig Support Structure for Structural Dynamic

Evaluation (8 Blade Configuration)

Two Blade Test ConfigurationFour Blade Test Configuration

Eight Blade Test Configuration

Strain Gage Arrangement for Two Blade Configuration

Strain Gage Arrangement for Four Blade Configuration

Strain Gage Arrangement for Eight Blade ConfigurationTest Points and Blade Shank Moments (2 Blade

Configuration)Test Points and Blade Shank Moments (4 Blade

Configuration)Test Points and Blade Shank Moments (8 Blade

Configuration)

Comparison of Signal to Noise for Flatwise Shank

Moment Gages (0° Inflow Angle)Trends of IP Response for 3° Angular Inflow

(Shank Moment)

Trends of IP Response for 3° Angular Inflow

(Radial Bending)

Radial Bending vs. Span for 3° Angular InflowComparison of SR-7L Natural Frequency Test

Results to Predictions

Test vs. Analysis (Radial Bending vs. Span)

Test vs. Analysis (Effect of Power)Test vs. Analysls (Effect of Mach No.)

LAP Blade Angle Potentlometer Output

Cp and CTNET VS, 3, 4 Blades, MN = .2, _ = 0=

Cp and CTNET VS, J, 4 Blades, MN = .5, _ = 0 °Cp and CTNeT VS. _, 4 Blades, MN - .8, _ - 0°C, and CTNET VS. J, 2 Blades, MN - .5, _ = 3°

CTNET VS. Cp, Constant O, 4 Blades, MN = .2

CTNE, VS. C,, Constant O, 4 Blades, MN - .5

PAGE

59

60

61646568

72

73

747576787980

81

82

83

85

94

95

96

98

I00

I01102

I07Ii0

Iii

112

113

114

115

L

vl

FIGURE

9-8

9-99-I0

9-ll

9-12

9-13

9-14

lO-I

I0-2

I0-3]0-4

10-5

10-6

10-7

]0-8

I0-9

lO-lO

lO-ll

lO-12

lO-13

lO-14

lO-15

lO-16

lO-17

lO-IB

]0-19

I0-20

ILLUSTRATIONS(Continued)

CTNET VS. Cp, Constant J, 4 Blades, MN = .8

CTNET VS. Cp, Constant J, 2 Blades, MN - .5, 9 = 3°

CTNET VS. Cp, Constant J, 8 Blades, _ = 0°CTNET VS. J, Constant Cp, 4 Blades, 9 = 0°

CTNET VS. J, Constant Cp, 8 Blades, _ = 0°

Comparison of Measured and Predicted Performance,4 Blades

Compar|son of Measured and Predicted Performance,8 Blades

Steady Pressure Test Rig Support StructureProp-Fan Steady Pressure Test Set-Up

LAP Blade Installation, Steady Pressure Test (2 Blade)

LAP Steady Pressure Blade (Camber Side)

LAP Steady Pressure Blade (Face Side)Steady Pressure Blade (Camber Side)

Steady Pressure Blade (Face Side)

Arrangement of the Pressure Taps on the Blade

Surface-Projected View

LAP SR-TL Steady Pressure Distribution of On-LlneData (Operating Condition No. 7)

LAP SR-TL Steady Pressure Distribution of On-Llne

Data (Operating Condition No. 2)

LAP SR-7L Steady Pressure Distribution of On-LlneData (Operating Condition No. 3)

LAP SR-7L Steady Pressure Distribution of On-Llne

Data (Operating Condition No. 4)LAP SR-TL Steady Pressure Distribution of On-Line

Data (Operating Condition No. 5)

LAP SR-7L Steady Pressure Distribution of On-LineData (Operating Condition No. 6)

LAP SR-7L Steady Pressure Distribution of On-Line

Data (Operating Condition No. 9)

LAP SR-?L Steady Pressure Distribution of On-Line

Data (Operatlng Condition No. 8)LAP SR-7L Steady Pressure Distribution of On-Line

Data (Operating Condition No. 7)

LAP SR-7L Steady Pressure Distrlbutlon of On-LineData (Operating Condition No. lO)

LAP SR-TL Steady Pressure Distribution of On-Line

Data (Operating Condltlon No. 11)

LAP SR-7L Steady Pressure Distribution of On-Line

Data (Operating Condition No. 13)

PAGE

116

117

118

120121

122

123

128127

128130131

132133

134

140

141

142

143

144

145

146

147

148

149

150

151

vii

FIGURE

10-21

I0-2210-23

10-24

II-I

!I-2

11-311-4

II-5

11-6

I!-7

11-811-911-1011-11

ILLUSTRATIONS(Continued)

LAP SR-7L Steady Pressure Distribution of On-LineData (Operating Condition No. 12)

Repeatability Problem for Static Low Power Point

Repeatability Problem for Static High Power Point

Illustration of Repeatability at My = .20

Unsteady Pressure Test Rig Support StructureProp-Fan Unsteady Pressure Test Set-Up

LAP Blade Installation Unsteady Pressure Test (2 Blade)

LAP Unsteady Pressure Blade (Camber Side)

LAP Unsteady Pressure Blade (Face Side)Unsteady Pressure Blade, Transducer Locations

(Face Side)

Unsteady Pressure Blade, Transducer Locations(Camber Side)

LAPISR-TL Unsteady Pressure Blade Transducer Mounting

Unsteady Pressure Test Set-Up With Wake Generator

Examples of Measured Unsteady Pressures

Illustration of the Terms "Advancing" and "Retreating"

for Angular Inflow

PAGE

152

153

154

155

158

159

161

163164

165

166

167

168171

173

TABLES

TABLE

7-I

8-I

8-II

8-III

8-IV

8-V

8-Vl

8-VllI0-I

II-I

ll-ll

Splnner/Centerbody Drag Test Points vs. Corrected Mach No.

Operating Conditions for Mn >.l and SampleIRP Tabulations (2 Blade Configuration)

Operating Conditions for Mn >.l and SampleIRP Tabulations (4 Blade Configuration)

Operating Conditions for Mn >.l and SampleIRP Tabulations (8 Blade Configuration)

Measured Vibratory Peaks vs. LimitsP-Order Content of Strain Gage Output at 3° Inflow Angle

(2 Blade Configuration)

Comparison of IP Content of Strain Gage Output

Obtained by Two Methods

Comparison of Test vs. Analysis (IP)

Operating Conditions for Blade Steady Pressure Testing(2 Blade LAP Propeller)

Operating Conditions for Blade Unsteady Pressure Testing

(2 Blade LAP Propeller)Chart of Transducers from wh|ch Data was Acquired vs.

Test Condition

PAGE

62

86

87

8990

92

93

103

135

160

170

ix/x

l.O ABSTRACT

High Speed Wind Tunnel Testing of the SR-7L Large Scale Advanced Prop-Fan(LAP) is hereln reported. The LAP |s a 2.74 meter (9.0 FT) diameter,

8-bladed tractor type rated for 4475 KW (6,000 SHP) at 1698 RPM. It was

designed and built by Hamllton Standard under contract to the NASA Lewis

Research Center. The LAP employs thin swept blades to provide efficlent

propulsion at flight speeds up to Mach .85.

Testing was conducted in the ONERA SI-MA Atmospheric Wind Tunnel in Modane,France. The test objectives were to confirm the LAP was free from high speedclassical flutter, determlne the structural and aerodynamic response toangular Inf|ow, measure blade surface pressures (static and dynamic) andevaluate the aerodynamic performance at various blade angles, rotationalspeeds and flight Mach numbers.

The measured structural and aerodynamic performance of the LAP correlated

well with analytical predictions thereby providing confidence in the computer

codes used for design. There were no signs of classlcal flutter throughout

all phases of the test up to and including the 0.84 maximum Mach number

achieved. Steady and unsteady blade surface pressures were successfully

measured for a wide range of Mach numbers, Inflow angles, rotatlonal speeds

and blade angles.

No barriers were discovered that would prevent proceedlng with the PTA

(Prop-Fan Test Assessment) Flight Test Program scheduled for early 1987.

1/'2

Z

2.0 SUMMARY

This report describes the procedures followed and results obtained during

High Speed Wind Tunnel Testing of the SR-7L Large Scale Advanced Prop-Fan(LAP). The LAP Is a 2.74 meter (9 foot) diameter, 8 bladed advanced

propeller designed to attain high propulsive efficiency at flight speeds up

to Mach .85. The Prop-Fan achieves this superior speed and efficiency byemploying thin swept blades and high disc loading. The High Speed Wlnd

Tunnel Test was conducted in the ONERA SI-MA Large Atmospheric Wind Tunnel

facility In Modane, France.

Testing was accomplished during two separate tunnel entries. This wasnecessitated by a test rlg failure during the first entry. Prior to

Interruption of the test in early 1986,.a11 structural dynamic, aerodynamic

performance and a limlted amount of blade steady pressure testing was

completed. Testing resumed in early 1987, and culminated in the completionof the steady and unsteady blade pressure tests. A complete chronological

history of both tunnel entries Is provided in Appendix A.

The purpose of the Wind Tunnel Test was to confirm the LAP was free from highspeed classical flutter, determine the structural and aerodynamic response to

angular inflow, measure blade surface pressures (both steady and unsteady)

and evaluate the aerodynamic performance at various blade angles, rotational

speeds and Mach numbers. Results from these tests would assist in determiningthe readiness of the LAP and Its Instrumentation systems for follow-on

Prop-Fan Test Assessment (PTA) flight testing. The Wind Tunnel Test was

accomplished in two phases. In the first phase, structural dynamic and

aerodynamic performance data were collected concurrently, for the 2, 4 and 8

blade configurations, over a wide range of blade angles, Mach numbers and

rotational speeds. Due to rig constraints, structural dynamic evaluatlon ofthe LAP operating at a fixed angle of attack was limited to the 2 bladed

Prop-Fan configuration at a 3° inflow angle. During the second phase of

testing, a specially fabricated static pressure tapped blade was employed to

map the blade surface steady pressure distributlon for a range of operatingconditions. Due to drive system power limitations, testing was accomplished

utilizing a two bladed Prop-Fan configuration to provide blade loadingssimulating the take-off, cutback and design cruise conditions. The final

phase of testing, also utilizing the two blade configuration for reasons

mentioned above, employed another specially instrumented blade incorporating

high frequency response pressure transducers. Data from these transducers wasused to define and evaluate the blade surface unsteady pressure distribution

for the same operating conditions run during the steady pressure test.

Unsteady pressure testing included evaluating the effects of a wake In the

propeller inflow at a 3° angle of attack.

2.0 (Continued)

Results from the High Speed Wind Tunnel Test demonstrated that the SR-7L

Prop-Fan was free of high speed blade flutter over the entire operating

envelope tested. Additionally, all measured blade surface and blade shank

strains were well below allowables set prior to testing. Good correlationwas found between measured and analytlcally predicted IP strain sensitivities

for the SR-7L blade. Results confirmed that IP strain peaks Inboard on the

blade and lessens near the tip, and that, in general, blade strains were

found to increase with power and Mach number. Measured aerodynamic

performance for the four blade configuration corresponded well withanalytlcal predictions over the entire range of points tested. Though good

agreement for the eight blade configuration was found at Mach numbers of .70

and .73, performance was slightly underpredlcted at .5 Mach number. Steadyand unsteady blade surface pressure measurements were successfully collected

for a wlde range of Mach numbers, rotational speeds, blade angles and inflow

angles. In addition to confirming the presence of tip edge and leadlng edgevortex flows at Mach numbers of zero and .2, shock waves were evident at the

traiIlng edge at .7 and .78 Mach number. Unsteady pressure responses wereclearly present as evidenced by a dominant l-P response in the angular inflow

data and significant 2-P response in the wake inflow data. Sinusoldal

response was evident on the pressure (face) side of the blade for all casesexamined for angular inflow conditions and on the suction (camber) side under

low loading conditions. Under high loading condit|ons, the suction (camber)side exhibited non-slnusoldal response resulting from the presence of tip and

leading edge vortices.

3.0 INTRODUCTION

The Prop-Fan, a high speed, high efficiency alrcraft propulsion concept waslaunched during the "O|l Crunch" days of the mid 19?O's. In response to the

national need to reduce fuel consumption, Congress directed NASA to address a

series of aircraft related technologies aimed at increasing the fuel

efficiency of airline operation. In response, NASA created the Aircraft

Energy Efficiency (ACEE) program which addressed fuel savings through

advancements in both airframe and engine technology. The element of the ACEEprogram offering the greatest potential fuel savings was the Advanced

Turboprop Program (ATP) as described In Reference I. The NASA Lewls Research

Center had total responsibility for the ATP project which is summarized in

Reference 2. The objective of the ATP was to demonstrate technology readinessfor efficient, reliable and acceptable operation of advanced turboprop-powered

commercial transports at cruise speeds up to Mach 0.8 and at altitudes above

9,BOO meters (30,000 ft.) while maintaining cabin comfort levels (noise and

vibration) comparable to those of modern turbofan-powered aircraft. The

technology would also apply to possible new military aircraft for a variety of

missions. Out of this project evolved the Prop-Fan concept.

Although high propulsive efficiency from turboprops was nothing new, the

standards of high cruise speed and cabin comfort set by the contemporary

turbofan powered alrcraft were beyond the capability of any turboprop poweredaircraft. The Prop-Fan concept evolved to satisfy the requirements of high

speed and altitude with improved efficiency while maintaining a high degree of

cabin comfort. It is characterized by the large number of blades (8 or lO),

thin airfoil sections, and swept blade planforms (Figure 3-I).

Once the concept and its benefits were defined, NASA conducted a systematic

program to verify that the predicted benefits could be achieved and that

there were no unsolvable problems in Implementing the concept. The potentialbenefits of the Prop-Fan propulsion concept have been investigated in numerous

propulslon and aircraft systems studies conducted by the airframe and engine

manufacturers under NASA sponsorship (References 3 thru 9). These studies

have shown that the inherent efficiency advantage that turboprop propulsionsystems have achieved at lower cruise speeds may now be exte,ded to the

higher speeds of todays turbofan and turbojet powered alrc_aft. By applying

swept wlng/reduced dlameter technology to the design of prol=eller blades, and

by achieving higher disc 1oadlngs through the use of a greater number ofblades, It was found that the inherent fuel efflclencyof t_e propeller can

be extended to speeds up to .85 Mn. The efficiency of the l_rop-Fan should

a11ow aircraft to be designed that are 15% to 25% more eff$cIent than today's

most technologlcally advanced turbofan powered airlines (l_mference I0).

Since 1975, Hamllton Standard has been deeply Involved v_lfl_,the NASA Lewis

Research Center In the development of the Prop-Fan. I/_I_ re{ently, thiseffort utillzed a series of .622 meter (2 ft.) dlametermc_e_s which

Incorporated differing numbers of blades as well as chamg_ _n blade shape.These models underwent exhaust|ve testing in several wlm_ _mels at NASA and

OF POOR Q'J_kb! _ '_

-L

ORICINAL PAOE_.A,.,r_ A_'4D WHITE- PHOTOGRAPH

FIGURE 3-1. I_OP-FAN CONCEPT

6

3.0 (Contlnued)

United Technologies as wel] as on a NASA Jetstar acoustic research vehicle.

In these tests the targeted efficiencles were demonstrated, the source noisewas characterized, and structural dynamics verlf|ed. Detailed descriptionsof these tests and results have been the subject of numerous technical

papers; a summary of which can be found in References II and 12.

Although the results of the aerodynamic performance and source nolse tests can

be confidently scaled from model to product size, the structure of the solid

homogeneous model blades Is so different from that envisioned for a productthat extrapolation of model structural behavlor Is unacceptable. The

verification of the structural integrity of a large-scale Prop-Fan then

becomes the final major technical hurdle to be crossed before industry

acceptance of the Prop-Fan as a viable aircraft propulsion scheme. Thisverification was initiated in 1983, when Hamilton Standard, under the

sponsorship of NASA Lewis Research Center, embarked on a program to build andtest a 2.74 meter (9 foot) diameter, 8-bladed Prop-Fan. Thls Prop-Fan was

designated the SR-7L or the Large Scale Advanced Prop-Fan (LAP). The majorelements of the LAP program are depicted in the summary schedule shown in

Figure 3-2. Detail design and fabrication of the Prop-Fan components(blades, hub and blade retention, spinner, pitch change mechanism, pitchcontrol and instrumentation system) was initiated early In 1983 building on a

preliminary design conductedunder an earlier contract. Various bench tests of each component then

followed to verify key design characteristics.

Design, fabrication and test of an aeroelastlcally scaled .622 meter (2 foot)

Prop-Fan model was included in the LAP program to obtaln an early assessmentof the Prop-Fan's aeroelastic characteristics. This model has been

designated SR-7A. Other objectives of the SR-7A testing Included the

measurement of aerodynamlc performance and noise.

Testing of the SR-7L rotor under the LAP program included in-house whirl,static and high speed wind tunnel tests. Whirl Rig Testing was successfully

conducted on the G-5 rlg at Hamilton Standard. The objectives of the testwere to measure the stiffness of the blade retentlon system, evaluate the

control dynamic characteristics of the blade pitch change system and determinethe wear rate on blade actuation and retention hardware. Whirl Rig Testlng

was conducted using stub weights to simulate the Prop-Fan blades. The stubs

provided appropriate centrifugal loading but generated essentially no thrustor drag. The Static Rotor Testing, with SR-7L blades installed, was success-

fully conducted at the Wright Aeronautical Laboratory at Wrlght Patterson AirForce Base in Dayton, Ohio (Reference 13). The goals of the Static Rotor

Test were to measure the static aerodynamic performance of the LAP, assessstall flutter characteristics and investigate the structural behavior and

integrity of the SR-TL blades for static operating conditions. The final

component of the LAP test program, High Speed Wind Tunnel Testing, is the

subject of this report.

,-,I

I,-.

"ll

a.

,,.J

I-,-

<

c

O0

o.I=

C0D.E _0 _

ki

0

o

W

U.@

ORIGINAL

BLACK AND WHITE

PACE

FHOIOGRAPN"

{/1I-ZLI.I

1.1.1J1.1.1

<

0

_L

rL<.-1

Z<1.,r,

orl,.

WUZ<

t_<uJ

<U

_J

<-i

_'1'

w

r:

E

r

3.0 (Continued)

The LAP program is comPlemented by another NASA sponsored program, theProp-Fan Test Assessment (PTA) program, which takes the large scale Prop-Fan

(developed under the LAP program) and mates it with an Allison Gas Turbinesupplied gas generator and gearbox to form a Prop-Fan propulsion system. The

ma_or elements of the PTA program are depicted In Figure 3-3. Following

completion of the Static Englne Test at Rohr's Brown Field Facl]ity, the

quick engine change (QEC) nacelle will be mated to the wing of Gulfstream II

aircraft which will ultimately serve as the f]|ght test vehicle for the

Prop-Fan.

This report addresses the procedures and results of the High Speed Hind Tunnel

Test conducted at the ONERA SI-MA Atmospheric _ind Tunnel in Modane, France.

This facility was selected for three reasons. First, it is capable of

reaching high cruise Mach numbers. Second, it is sufficiently large (8 meteror 26.25 ft. diameter test section) to avoid excessive wall interference

effects; and third, it has an existing model drive system. Although the power

capability of the drive Is only about one fourth of what the Prop-Fan is

designed to absorb, proper blade loading can be reached by running with apartial set of blades (eight, four, and two blade configurations). The

specific objectives of the test are listed below.

To conduct a careful and controlled search for any evidence of

classical flutter. Because of the greater air density of the :nd

tunnel, it is possible to more closely approach the flutterthreshold than at the 10,668 meter (35,000 ft) flight altitude. At

design Mach number the wind tunnel operates at an effective altitude

of about 4,267 meters (14,000 ft). Analytic predictions and tests

of the SR-7A model strongly suggested that classical flutter wouldnot be encountered.

To measure steady and unsteady surface pressure on the blade as wellas overall Prop-Fan performance. One blade was Instrumented with

465 static taps (20 chordal at 13 radial stations on the camber sideand |6 chordal at 13 radial stations on the face side of the blade)

to obtain a complete pressure map. Another blade had 26 dynamicpressure sensors (7 chordal at the 35 inch station and 6 chordal at

the 49 inch station, for each side of the blade) to assess unsteady

effects. These measurements will provide bench_rk data forunderstanding the physlcs of transonic flow over the blades and for

verlficatlon of analytic computer codes.

To determine the structural and aerodynamic response of the Prop-Fan

to angular Inflow. Analysis of data from this simple, known angularinflow condition will slgnlfIcantly contribute to the understanding

of Prop-Fan behavior in the more complex, airplane installed flowfield.

LARGE - SCALE PROPFAN

FROM LAP

UNIQUE TILTNACELLE

MOOIirlED TURBOSHAFT IIb / \ II

ENGINE GEARBOX /_

" _SYSTEM STATIC TEST

AT ROHR BROWN FIELD

AIRCRAFT.OOELTESTS\ / /

k; (.:._.'_ " /'_ _._ CAB,,,ENVIROMENT

1 T- 1.2f,_;_lN tillS" _ "-;-'_',_ EN-ROUTE NOISEL I _'I._JL_ - FLIGHT TESTING _"_-a'mlJ_a_-,'_Imb _ INSTALLED OPERATING

CHARACTERISTICS

FIGURE 3-3. PROP - FAN TEST ASSESSMENT (PTA} PROGRAM ELEMENTS

10

4.0 PROP-FAN DESCRIPTION

4.1 General Description

The Large Scale Advanced Prop-Fan shown _n Figure 4-1, is a 2.74 meter(9 ft.) diameter 8 bladed tractor type Prop-Fan rated for 4474 KN (6,000 HP)

at 1698 RPM. To achieve the program objective of verifying large scale

Prop-Fan structural integrity, a number of design requirements and goals were

established as summarized in Figure 4-2. The requirements Includecharacteristics Judged essentlal to meeting the program objective as well as

design features established from prior work. The goals, on the other hand,

represent design targets and were judged less Important to the program

objective. The LAP is designed to be mounted on a standard 60A spIined

propeller shaft for an existing turboprop gearbox. It has a hydrau11callyactuated blade pitch change system and a hydromechanlcal pitch control that

allows the Prop-Fan to operate In a speed governing mode or, with minormodifications, in a Beta Control mode, as was the case in this test. Beta

Control mode operation was chosen to provide the operator the capability toselect desired blade pitch angles while running. A pitchlock feature Is also

incorporated in the actuator. This feature malntalns the propeller blade

angle In the event of a loss of system operating oil pressure. The design of

the actuator and control is based on proven technology used in Hamilton

Standard's military and commercial propellers. A brief description of each of

the major elements of the LAP as depicted in Figure 4-I, is presented below.

4.2 SR-7L Blade

Features of the structural configuration of the SR-7L blade are shown inFigure 4-3. These Include a central aluminum spar which forms the structural

"backbone" of the blade, a multi-layered glass-cloth-reinforced shell

overhanging the leading and trailing edge of the spar, a nickel sheath which

covers the leading edge of the outer two-thirds of the blade, and anon-operational integral heater In the Inboard leadlng edge area. Though the

scope of the LAP testing never Included utillzation of the blade heaters, Itwas decided to Install the heaters to evaluate the structural response of a

blade closely resembling that of a typical blade configuration. The remaining

internal cavitles are filled wlth Iow-denslty rigid foam. The outboard

portion of the spar is Intentionally moved toward the blade leading edge to

increase stabillty by reducing overhung mass in the tip trailing edge, whileat the same time increasing the Integrity of the leadlng edge from the stand-

point of resistance to foreign object damage.

The blade design makes use of a NACA Series 16 airfoil outboard and a Serles

65 clrcular arc airfoil Inboard. Each blade has an activity factor of 227.3

wlth 45" of blade sweep at the tip. The blades were designed with

predeflectlon so that the blades will assume the desired aerodynamic shape at

the cruise operating condition (Reference 14).

11

BLADE

=:

BLADE RETENTION

CONTROL

ACTUATOR

SPI

FIGURE 4-I. LARGE-SCALE ADVANCED PROP-FAN

12O_!GINAV p_v

_Lf-,CK At, iD WHi-rE PHOIOGRAPH

_ UNITEDTECHNOLOGIES

DESIGN REQUIREMENTS

Configuration

DiameterNumber of blades

2.74 meter (9 ft)6

• Design point

Cruise Mach numberAltitudeTip speedPower loading (SHPID2)

0.810,668 meters (35,000 ft)244 meterslsec (800 ft/sec)32

• Structural Integrity

Flutter free over normal flight envelope (M <0.8)Stresses within allowable limitsOverspeed toleranceCritical speed margins

Safety featuresLeading edge projectionLighting protectionIcing protection (installed but not operational)Overspeod protection

• Reverse thrust capability

DESIGN GOALS

• Net efficiency (isolated nacelle)

• NoiseNear field (designpoint cruise, max.free field, 0.81))Far field

-78.6% AT M = 0.8, 10,668 meters (35,000 ftXCruise)-52.0% AT M = 0.2, SL (TO)

.144 db overall sound pressure level

•FAR 36 minus 10 db

• Stall flutter -None st 100% TO power and rpm; M = 0- 0. 2

• High speed (classical) flutter -None over extended flight envelope,(M __<0.85) 105% max operating speed

• Over speed limit (hub, blades,blade retention)

.120% max operating speed - no yield-141% max operating speed - no failure

Foreign object damageMinor-Birds up to 4 ozModerate -2" Hail; Birds to 2 Ib

Major-Birds up to 4 Ib

-No damage to pdmary blade structure-Some loss of matedal or airfoil distortion;

operate at 76% power for 5 minutes-Some loss of matedal or airfoil distortion;

maintain ability to feather

• Blade life .35,000 hr - replacement with scheduled malnt.-50,000 hr - meantime between unscheduled

removal

FIGURE 4-2. DESIGN REQUIREMENTS AND GOALS SUMMARY

13

z

FIGURE 4-3. FEATURES OF THE SR-7L BLADE CONSTRUCTION

14

4.2 (Contlnued)

Although some improvements in sweep/stress/stability trade-offs were

predicted through the use of advanced composites, it was decided not toinclude these In the final blade design. Their use would requlre the

development of new manufacturing technology, both in terms of suitable

construction methods and processes, and lengthy development of design

allowables to reflect the manufacturing process.

It was felt that the scope of the program would be best served by utilizlng

the service-proven combination of an aluminum spar enveloped with a

fiberglass shell for which processes and stress allowables are well known.

4.3 Pitch Change Actuator

The pitch change system Is comprised of two components, a pitch change

actuator and a control. The pitch change actuator Is the prime mover for

blade angle change and Is located within the Prop-Fan hub as shown inFigure 4-4. Its primary components are an internal stationary piston, a

translating outer cylinder with an integral yoke to engage each of the blade

trunnlons and a pltchlock and servo assembly which contains a four-way

metering beta valve assembly, a pitchlock screw, a ground adjustable lowpitch stop, a servo piston and a ball screw to drive the pitchlock screw andbeta valve (Reference 15).

As stated earlier, the LAP was modified to operate In the Beta Control mode

throughout all phases of the wind tunnel test. This was accomplished throughthe use of an electromechanlcal controller (D.C. motor and gearhead) mounted

on the front of the dome as shown in Figure 4-5. The controller provides the

rotary input to the pitchlock servo directly, replacing the rotary input from

the half area servo. In order to give the D.C. motor full control of the

blade angle, the servo and ballscrew must be disconnected from the pitchlockscrew by removing the quill shaft. This disables the control signal and

allows the motor to work directly on the pitchlock screw without having to

fight the servo output. This permits the operator to remotely position blade

angle as deslred while running.

The actuator was designed to present state-of-the-art technology and low

development risk technique that has been used on a number of existing

propeller systems. The design uses mostly steel for the load carryingmembers and all surfaces subject to sliding seal wear are chrome plated to

increase durability. The actuator was designed to conservative stress and

deflection levels to minimize development effort while maintaining a

reasonable but not minimum weight.

The pltch change mechanism was designed such that all malfunctions will

either cause the system to pltchIock or go to Feather. An additional safety

Feature on the LAP is a ground adjustable low pitch stop. This 11mlts the

minimum blade angle under all circumstances.

15

BLACK

FIGURE 4-4 LAP PITCH CHANGE ACTUATOR

ORIGINAL PAGE

AND WHITE pI_OTOGRAPIq

16

± _.=

uo

a-(n

2m

Z<-rUILl:E..J0n,

Z0(J

u!-

;[IJJ>m

Q

0I-o

W

17

4.4 Pitch Control

To minimize cost and development time, It was decided to utilize a modified

version of an existing turboprop propeller control. Based on the type of

engine and gearbox planned for use with the LAP, the control selected was a

modified 54460 control (Figure 4-6). The 54460 hydromechanlcal integral oiicontrol, which Is currently In use on the Grumman E-2 and C-2 aircraft, was

modeled after the Lockheed C-130 and P-3 controls. Since the first production

unit in 1956, there have been over 12,000 built and they have logged over

80,000,000 operating hours.

The primary func_ion of the LAP control, as modified for wind tunnel testing,was to generate the hydraulic pressure necessary to assure proper actuator

operation. Two gear type pumps located in the stationary control and driven

by the Prop-Fan shaft provided the system hydraulic pressure.

4.5 Hub and Blade Retention

The LAP hub assembly forms a seml-rlgid llnk between the blades, which

provide the thrust, and the engine shaft, which provides the torque(Reference 16). The hub and tailshaft is a one piece forged component which

is carburlzed, heat treated and machined (Figure 4-7). A single row ball

bearing retains each of the eight blades In the hub, while the tailshaft

secures the Prop-Fan to the engine shaft through two cone seats that are

preloaded against each other by the Prop-Fan retaining nut. The hub alsoforms the support for the pitch change actuator system, the control and the

spinner.

The retention transmits the loads from the blades to the hub while

accommodating changes in blade pitch. The single row ball bearing retentionprovides ease of maintenance by allowing individual blade replacement without

disassembly of the hub. It has a through hardened Inner race which seatsagainst the aluminum blade shank and an outer race which Is integral with the

barrel. The outer race Is carburlzed to achieve the hardness necessary to

support the ball loads. The balls are kept apart from each other by an

elastomerIc separator. The rotational speed of the Prop-Fan keeps the

retention submerged In oll which is contained in the hub by eight blade seals(Reference 17).

4.6 Spinner

The LAP spinner and rear bulkhead assembly Is essentially a reinforcedflberglass/epoxy shell, supported by the hub and actuator, and Incorporating

an aerodynamic shape to facilitate proper alr flow around the blade roots

(Figure 4-8). Its primary function is to insure proper Prop-Fan aerodynamicperformance. The rear bulkhead, which mounts on the rear of the hub arms, Is

the main structural support for the spinner and provides a mounting surfacefor much of the instrumentation hardware in the rotating field.

18

!

lg

!

n

-r

_I0D_I-Z0UD.

..I

I_CK Abed WHITE Pi-_OTOGRAP_},:I

FIGURE 4-7. SR-7L HUB

20

OP..IGIF-'#.L P/_.OE'

BLACK A _'_n,-,,., '.,f.,'r4t"t-E -.I ,,- -._.--,D,,Du,_,...,,,j,...,,;..,

FIGURE 4-8 SR-7L SPINNER

21/22

r_

5.0 INSTRUMENTATION DESCRIPTION

5.1 General Description

In order to accomplish the objectives set forth in the Plan of Test, three

separate Instrumentatlon arrangements were employed to gather and record the

desired data. For the structural dynamic and aerodynamic performance tests,

the system described in sections 5.1.I and 5.1.2 was used. For the blade

surface steady pressure test, a system description is provided In section 5.2

and for the unsteady pressure test, the system used is presented insection 5.3. Figure 5-I provides a summary listing of all the

instrumentation, rotating and stationary, and indicates whether It was

provided by H.S. or ONERA. Additional detailed descriptions of the key data

gathering devlced required for specific phases of the test are presented in

the appropriate Test Procedure section of this report.

Common to all three instrumentation arrangements was the frequency modulated

multlplex electronic data acquisition system (DAS), as shown in Figure 5-2,

and several Prop-Fan system diagnostic monitoring devices.

5.1.1 Electronic Data Acquisition Systems (DAS)

The electronlc data acquisition system for the LAP provided the capacity totransmit 33 channels of information from the electronic measurement devices

on the rotating slde of the Prop-Fan to the data collection and monitoring

equipment in the stationary field. This was accomplished by employing an

eight rlng platter-type sllp ring assembly which provided the electrlcal

Interface for the DAS between the rotating Prop-Fan assembly and the

non-rotatlng control.

As illustrated in Figure 5-2, three rings were utilized to transmit data to

the stationary field. One of the three carried the output of a potentiometerwhich was mounted as shown in Figure 5-3 and used for monitoring blade angle

position. The other two rings transm|tted the remaining 32 channels of

information which consisted of signals from two miniature pressure transducers

for monitoring actuator high and low pitch pressures and combinations of upto 30 blade strain or unsteady pressure signals, as determined by the

particular test being conducted. Transmittal of these 32 signals on only two

sllp rings necessitated the use of FM multiplexing. The DC signals from the

strain gages and pressure transducers in the rotating field were divided intotwo groups of sixteen. The signals were then converted to frequency modulated

signals by two groups of voltage controlled oscillators. Each group was then

multlplexed by a mixer, allowing thirty two channels to be transmittedthrough two sllp rings. The two groups of sixteen channels were each

detranslated in the stationary field to four groups of four multiplexedchannels (IRIG Standard IA thru 4A). Each set of four channels was recorded

on one track of a standard Honeywell IOl tape recorder. Simultaneously,

eight discriminators were used to demodulate any two groups of four channels

for real tlme monitoring of data. One discriminator was tuned to the centerfrequency of each channel.

_G PAGE

23

H.S. INSTRUMENTATION

ROTATING:

HIGH AND LOW PITCH OIL PRESSURE MEASUREMENTBLADE ANGLE MEASUREMENT

BLADE VIBRATORY STRAIN MEASUREMENTS

BLADE SURFACE STEADY AND UNSTEADY PRESSURE MEASUREMENTS

STATIONARY:

CONTROL SUPPLY PRESSURE MEASUREMENT

CONTROL SUMP OIL TEMPERATURE MEASUREMENT

HEAT EXCHANGER _P MEASUREMENT

CONTROL AND SCANIVALVE ACCELEROMETER MEASUREMENTS

INSTRUMENTATION PROVIDED BY ONERA (ALL STATIONARY)

TEST RIG INSTRUMENTATION:

RIG ACCELEROMETER MEASUREMENTSPROP-FAN RPM PICK-UP

RIG BEARING TEMPERATURE MEASUREMENTS

BALANCE AND TORQUEMETER MEASUREMENTS (FIGURE 6-I0)

CENTERBODY PRESSURE MEASUREMENTS (FIGURES 7-I0, 7-11)SPINNER BULKHEAD PRESSURE MEASUREMENTS (FIGURE 7-]2)

TUNNEL INSTRUMENTATION:

STATIC PRESSURE MEASUREMENTS:

- 4 METERS UPSTREAM OF THE PROP-FAN PLANE OF ROTATION

- ADJACENT TO THE PROP-FAN PLANE OF ROTATIONSTAGNATION PRESSURE MEASUREMENTS

AIR DENSITY MEASUREMENT

STAGNATION TEMPERATURE MEASUREMENT OF FLOW

STATIC TEMPERATURE MEASUREMENT OF FLOW

FIGURE 5-]. H.S. AND ONERA INSTRUMENTATION LIST

24

!_Jl

!

uI

b-<

_J

2Wl-

>-

z0b-<I-zb.I

n,1-(/)z

<-i

W

i-l-l-l-l-l-l-i-

25

FIGURE5-3. LAPBLADEPITCHANGLEMEASUREMENTSYSTEM

26

5.1.1 (Continued)

The FM electronic instrumentation system provided inherent noise immunity for

data transmission. The frequency response for the system was DC to lO00 HZ.Overall accuracy of the system was ±3% RSS. Time correlation betweenchannels was +13.8 mlcroseconds.

The electronic Instrumentation system allowed for up to ten blade shell or

shank gages to be installed on any blade, though a maximum of 32 gages could

be active at any one time. Straln gaged blades were closely monitored during

flutter and critical speed testing as well as measuring strains at various

operating conditions. The blade shank strain gages were employed to measureblade bending moments.

A total of sixteen gages could be selected from blades one through four and

an additional sixteen gages could be chosen from blades flve through eight.

Selection of the desired combination of gages was accomplished using eightprogrammable connectors mounted on the Prop-Fan hub. Programming of the

connectors required using Jumper wires to connect the sockets of patch boards

in the connectors. The bridge completion circuits for the strain gages werelocated on circuit boards in the blade cuff.

Monitoring of instrumentation during the test was accomplished with an

oscilloscope, a spectrum analyzer and a vlsicorder. The oscilloscope

permitted a time domain display of four channels simultaneously. The

spectrum analyzer provided the capability to dlsplay any one channel in thetime or frequency domain. The analyzer also had transient capture and

playback features. The vlslcorder provided a hard copy plot ofinstrumentation signals versus time.

5.1.2 Prop-Fan Diagnostic Monitoring Instrumentation

In addition to monltorlng and collecting data from the rotating fleld, a

number of parameters intended to provlde protection for the Prop-Fan system

were measured and digitally dlspIayed by the stationary field instrumentation.These included control sump oil temperature, control supply oil pressure,

differential pressure across the heat exchanger and control/Scanlvalve

accelerometer measurements. The control sump temperature was measured by athermocouple installed inside the oll draln port of the control. Control

supply oil pressure was measured by a transducer Inslde the control. This

transducer slgnal was transmitted via an existing connector on the control.

The heat exchanger differential pressure was measured by a AP transducerconnected across the o11 exlt and return ports on the aft face of the control.

The control vibration was measured by two accelerometers mounted on thepropeller control houslng. One accelerometer was oriented to sense motlon inthe horlzontal dlrectlon and the other to sense motlon in the vertlcal

27

5.1.2 (Continued)

directlon. Additionally, during conduct of the steady pressure test, an

accelerometer sensing vertical motion, was installed on the Scanlvalve

fairing assembly. The "once per revolution" signal was orlglnally planned to

be provided by a pickup mounted on the control and triggered by a target on

the rotating propeller bulkhead. However, the system eventually used was

provided by an eddy current proximity probe targeted on a 59 tooth wheel(on a 60 t(_oth basis) mounted on the propeller shaft. The rotational speed

was averaged over ten revolutions.

5.2 Steady Pressure Measurement System

In addition to employing the instrumentation systems described In section

5.I.I and 5.1.2, a specially designed pneumatic instrumentation system was

used to collect and measure blade airfoil surface steady pressures. Thls

system consisted of a speclally fabricated blade with rows of pressure tapsInstalled at thirteen radial stations and a scanlvalve mounted on the nose of

the Prop-Fan. (See Figure 5-4.) The pressure taps were connected to the

scanlvalve by 36 capillary tubes run along the acutator dome.

The scanivalve Instrumentation system provided 36 channels for transmitting

steady pressure data from the surface of the blade to the stationary field..

The scanIvalve itself consists of a rotating and a stationary portion. The

rotatlng portion was attached to the front of the actuator dome. The radialtubes from the steady pressure measurement blade were connected to the

rotatlng portion of the scanlvalve. Each tube was connected to one channel

of the scanlvalve. The statlonary portion of the scanlvalve contained apressure transducer which monitored one channel of the scanIvalve at a time.

The stationary portion of the valve protruded through the leading edge of thepropeller spinner and was restrained against rotation. Switching of the

scanlvalve channels to be monitored by the transducer was controlled by a

pneumatic signal requiring a clean alr source of 150 psi. The scanning ratewas adjustable from 0.l to lO seconds per channel. The scanivaIve was

enclosed In an aerodynamically shaped fiberglass fairing to maintain a well

behaved inflow to the Prop-Fan. The umbilical, which connected the

scanlvalve wlth the control and monitoring equipment outside the tunnel, was

enclosed In a conduit with an airfoil shaped cross section. This alsominimized disturbance of the inflow to the Prop-Fan.

On 11ne monitoring of the scanlvalve pressure transducer sIgnals was

accompllshed uslng the digital readout from the scanlvalve controller. Inaddltlon to recording the measured pressures on magnetic tape, on-site data

manlpulatlon and prlnt-out generation was provlded by a Fluke DA computer

coupled wlth a plotter and dot matrix printer.

28

OF., I_0 OR QdAL_I ,Y

n

O_(/1

IX

\

29

5.3 Unsteady Pressure Measurement System

Collection of blade surface unsteady pressure data was accomplished by

utilizlng a specially instrumented blade, coupled with the instrumentation

arrangement described earlier in sections 5.l.l and 5,1.2. Twenty-six highfrequency response pressure transducers were |nstalled in two rows on each

side of the blade. The unsteady pressure signals were transmitted from the

rotating to the stationary field through the FM multiplex electronic data

acquisition system. The signals were monitored on the four-channe]

oscilloscope and recorded on the IRIG tape recorder.

L

30

6.0 TEST FACILITY DESCRIPTION

6.1 Wind Tunnel Description

The LAP High Speed Rotor Test was performed in Chariot No. 3 In the SI-MA

Large Wind Tunnel at the Modane-Avrleux Aerothermodynamlc Test Center

operated by the Office National D'Etudes et des Recherches Aerospatlales(ONERA), France (Figures 6-I and 6-2). The tunnel is a continuous, closed

loop, single return, atmospheric facility incorporating bleed slots In the

tunnel walls to a11ow air exchange capability and tunnel circuit screens tomlnlmlze turbulence, as shown In Figures 6-3 and 6-4. Tunnel cooling is

accomplished by alr exchange wlth outside air of up to 20% of the testsection mass flow rate. The tunnel Is driven by two coaxial counter-rotating

Pelton water turbines from a water supply wlth an 865 meter head, sufficient

to generate a maximum of 88MW (I00,000 HP). The test section velocity can be

continuously varied from 8 m/sec to Mach 1.0, well encompassing the range forthis test. The curves of Figure 6-5 glve a representative sample of the

tunnel driving power and Reynolds number as a function of test section Mach

number. Tunnel pressure altitude varies from approximately 1,100 meters

(3,600 ft.) at low speed, up to 6,000 meters (20,000 ft.) at Mach l.O. At

Mach 0.8 the tunnel operates at an effective pressure altitude of

approximately 4,260 meters (14,000 ft.). Stagnation temperature ranges from

-20°C to +60°C, depending on the external ambient temperature and the Machnumber in the test sectlon.

The tunnel has three Interchangeable test sections or chariots which are

positioned in the aerodynamic clrcult. Each of the chariots contain adifferent test section shape and size and are speclallzed for a wide variety

of test capabilities. The use of Individual chariots a11ows a test to beconducted in the tunnel while the next test Is being set-up In another

chariot located In l of 2 adjacent mounting stations. The LAP was installed

In Chariot No. 3 which has a propeller test rlg permanently mounted In its

test section (Figure 6-6). The test section Is 14 meters long, has a circularcross section with a diameter of about 8 meters and non-permeable walls.Several lateral fillers or Inserts were added to the test section specifically

for this test to assure the desired range of Mach numbers could be attained.

They serve the addltional purpose of area compensating for the propeller test

rlg blockage generated by the 970 mm diameter center body. A more detaileddescription of the tunnel facilities is contained in References 18 and 19.

6.2 Test Rig Description

The Prop-Fan was Installed on the test rig as shown In Flgure 6-6. The power

source for the Prop-Fan drive system Is twln Turbomeca gas turbines driving a

common gearbox, enclosed In a streamlined pod located approximately ? metersdownstream of the Prop-Fan plane, as shown In Figure 6-7. The engines were

rated for a combined maximum power of 1000 KW (1341) HP at standard

conditions (59°F, 29.92 In. Hg.) as depicted In Figure 6-8. Velocity

measurements made upstream of the engine Inlets Indicated no disturbance at

the Prop-Fan plane due to the operation of the engines. The gearbox was a

31

o 50 lOOM

\

FIGURE 6-1. MODANE - AVRIEUX AEROTHERMODYNAMIC TEST CENTER

32

ORtGfNAL PAGE

BLACK AND WHITE PHOTOC_RAPH

FIGURE 6-2 S1-MA LARGE ATMOSPHERIC WIND TUNNEL

33

_ 15_.M "-I

/I _. T.oT,_IR L,- - iil_l

,0M E×..osT I ! WORKING CO-AXI'L_II I -_".... '-I _Ec_ION _--N-il!, _,,._',,! i//I I__-_-J_,.._t_-I_, w , il__r

---t-_.-l-.-I_-l-l-.:l_tfl-.-I- , -_i--._ IIl-'b,_JI// I I IL I1.,"1 'i'T'_-FI i , L I". _! !','_',!

. .. ,_" ! ,. IT !r l lltIH i I!l]-._.:_l

I s ,,,ov.o,.E =,-,.R,OTS _ I _!i |i WATER

I "_-_.-_ -- I AIR INLET

F__l_-I i!_

L.__'-° ..... _ :li WATEr' PIPE

. FOR ENGINES !I;AH = 865M

FIGURE 6-3. S1-MA WIND TUNNEL SCHEMATIC

_4

BLACKORIGINAL

ANO WH ITE

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

PI-.iOTOGRAPH

FIGURE 6-4. S1-MA WIND TUNNEL NOZZLE INLET "I1B_SE'¢TION

3.5

10

6

e (MWJ

0.5 MACH 1

lOO

8o

60

4o

2o

NOTE: THE CHART ABOVE IS FOR A REFERENCE CHORD

LENGTH OF 0.7 M

FIGURE 6-5. TUNNEL DRIVING POWER AND REYNOLDS NUMBER AS A

FUNCTION OF MACH NUMBER

36

0

o)

f

!

<_

w

tg

z

z0

zIil

I-0z

Z

t-,0r_<-I-i.i

z0F-

ILl

-I--

I--

Zi

Z0I

l-<-I-i<I-

Z

Z<

0O_

bJm"

C_

37

0R!0!NAL PA__

BLACK AND WHITE PHOTOGRAPH

= .

- O_--POOR QUALITY

FIGURE 6-7. PROP-FAN DRIVE SYSTEM ENGINIES_I_II_IF_.ARBOX

38

POWER

Kwi1200

1oo0

800

600

OPERATION LIMITED

(MAX, CONTINUOUS)

._'____'_"_'_

0 500 I000

MINIMUM PERMISSIBLE

PROP-FAN SPEED

TO 5 MINUTES

..

_NORMAL OPERATING REGION !:._j ....',_;..._,_._'i b

',-_'7;._:_";'_',7:,._,,. -_,-"'_':",_";:_.:.' _':'__.;_;_ t_-_• _,;"_'Z,; ":. :":".'4_';...........,,_....._._..;1";$>,,."_............ ,.7, .- .'-,_; ," ,',-'_ ,..":;'_" "'_'"""%% _;:'_!".,_'_

".::_!,.,::%_._:'_._._,_._'Z'._<_s;"z......_.._..........,._.,..

1500 2000 22.00 2460

PRPM

FIGURE 6-8. PROP-FAN TEST RIG POWER LIMITATIONS

39

6.2 (Continued)

speed reducer providing a reduction in RPM of approximately 3:1 from the

engine to the Prop-Fan. Power was transmitted to the Prop-Fan from the

gearbox through a long drive shaft and a balance (Figure 6-9). The drive

shaft housing was supported by rods and damping cables attached to the tunnelwalls. The particular rod and cable arrangement used In each test was

dependent upon the type of test being run and the number of blades

installed. These arrangements were established during a pre-test vibration

survey conducted In the tunnel by ONERA. Universal joints at each end of thedrive shaft allowed the Prop-Fan to be operated at various inflow angles

relative to the flow through the test section.

A balance and torquemeter, Installed in the drive system, allowed the thrust,

torque, sides forces and bending moments acting on the Prop-Fan to be

measured. A diagram of the balance is shown Is Figure 6-10. Forces andmoments are transmitted through the balance by six strain gaged elements

whlch are referred to as dynamometers. Thrust, yawing moment and pitching

moment are transmitted and measured by the three axial dynamometers (TI, T2

and T3). Balance frlctlon torque and the lateral and vertical forces aretransmitted and measured by the three tangential dynamometers (Zl, Z2 and Y).

The torque supplied to the Prop-Fan is computed by subtracting the measured

friction losses in the balance from the torque measured and transmitted by

the torquemeter. A11 six dynamometers have a capacity of 2000 daN. Two

flexible couplings decouple the transmission of the torque from the thrustmeasurement with one of the couplings being fitted with a gage bridge

(torquemeter).

A stationary aerodynamic fairing or centerbody was installed around the drive

system. This was also commonly referred to as the minimum body. The minimumbody provided a downstream extension of the aerodynamic contour of the

Prop-Fan spinner and was designed to reduce the air veloclty passing through

the root portion of the Prop-Fan rotor. At the inner blade radii thecombination of thick airfoil sections and the large number of blades presents

significant blockage to the flow, which could result in choking if the

velocity was not moderated. The minimum body Is approximatly 35% of the

diameter of the Prop-Fan rotor. A description of the minimum body pressuretap arrangement is presented In Section 7.0.

6.3 Test Riq Modifications

The LAP Prop-Fan High Speed Wind Tunnel Test was completed following the

incorporation of several structural improvements to the propeller test rig

drive system. These modifications, as shown In the comparison of Figures

6-II and 6-12, were added to preclude recurrence of a cone seat fretting

corrosion problem encountered early In the testing, which led to an

unanticipated interruption of the test. In addition, the modifications in

genera! provided a more structurally sound rig design, better able to handle

the loads Imparted by the Prop-Fan.

40

w!--

>

w>

n,r_

Z<I,

0n,

ID

w

It.

41

U

l-

IJJ"l-u171

uz<-I

In

Z

<

I.-1.1.1>

k9I

II

I-Z<

_k

"7_D

IJ.l

l.l..

r

42

i

Z0

<

m

U.Z0

I-

wI,-

Z

m

0

,2"rID

W

w

43

\

\\\

<

)" k-b. Z< bJ-r _.u) Id_. r_

0 l"

r_ _j

U bIbl ZL 0

0Z

0

OU

_z<

b_.j><

(2b.

c: 0b_wOu

z0m

<

0

U.z0U

0

n,

1-01

I-

0ILl

0

e,i

_D

W

44

6.3 (Continued)

The most significant modification to the drive system was the elimination of

the cone seat retention configuration on the aft end of the prop shaft. It

was this area which experienced the severe fretting corrosion problem early|n the test. The new design provides for a one piece prop shaft mounted in a

substantially enlarged forward test rig housing incorporating increased

capacity cylindrical roller bearings. Since a safe structural operating

envelope for the Prop-Fan had been defined earlier in the testing, thebalance was removed to provide room for the test rig structural improvements.

45/46

r

7.0 SPINNER AND CENTERBODY DRAG MEASUREMENT

7.1 Test Rig Vibration Survey

Before testing was conducted In the SI-MA wind tunnel, the whirl flutter

stabillty of the Prop-Fan/test rlg system was evaluated. Stability of the

Prop-Fan/test rlg was a major concern because the LAP was the largestassembly ever Installed on the test rlg.

The Prop-Fan was assembled and installed on the propeller drive system in the

tunnel test section in preparation for the vibration survey. To e11minate

the possibility of damaging any of the actual LAP blades during this survey,

blade stubs were installed in place of the blades. Additlonal weights werehung from selected blade stubs to account for the weight difference between

the stubs and blades. The stubs were shimmed to minimize any movement in theblade retention during the shake test.

The survey was conducted using two electromechanlcal shakers positioned at

various locations along the axis of the mlnlmum body. Data was obtained

using two accelerometers mounted on the test rlg force balance Just aft of

the Prop-Fan. The data was used to determine the elgenfrequencles of the

various rlg vibration modes in the horizontal and vertical planes.

The results of the survey confirmed the need for an unsymmetrical rlg supportstructure which pre-test analytical whirl flutter stability studies hadsuggested. Though the vibration test demonstrated that the 4 bladed SR-7L

would operate whirl flutter free, it was agreed that rig damping

characteristics would be closely monitored during all test envelope expansion.In order to incorporate the maximum test rlg resistance to whirl flutter

onset, several different rlg support structure conflguratlons were employed,based on the number of blades Installed, as depicted In Figures 7-8, 8-I,8-2, lO-I and ll-l.

7.2 Wlnd Tunnel Corrections

Immediately following completion of the test rig vibration survey, ONERA

conducted the wlnd tunnel calibration test. The purpose of this calibration

was to remove the effects of the presence of the test rig on the measured

Mach number. The ONERA approach was as follows. First, a compressible flowcalculation was performed to establish the axial velocity distribution at theworking radius of the blade (.6R) through the plane of rotation of the

Prop-Fan. Next, two independent axial velocity surveys were made at the sameradial location, utllizlng a speclal pressure tapped plpe and scanlvalve

system. The results of these surveys are shown in Figures 7-) thru 7-6. The

uncorrected Mach number data represents actual measurements _alken, while thecorrected Mach number data is the difference between the ¢a_¢ulatlons and the

measurements of Flgures 7-I thru 7-6 added to the wall leach m_:mber measured

2 meters (6.56 ft.) upstream of the Prop-Fan plane of rotat_om. Thus, thecorrected Mach number data represents

47

r,,@

¢IO¢o:EDZ

3:U<X.I<

8..,I

Moo= .354

;RRECTED MACH NUMBER

I

UNCORRECTED MACH NUMBER

c

|o00 MM 500 MM

I

500 MM 1000 MM

BODY LENGTH, MILLIMETERS

FIGURE 7-1. AXIAL MACH NUMBER DISTRIBUTION FOR FREESTREAM

MACH NUMBER .354

48

n,

n.,

en

Z

"!"u

J

u0.I

MACH NUMBER

CORRECTED MACH NUMBER

I

I

1!000 MM 500 MM 0 500 MM

BODY LENGTH, MILLIMETERS

FIGURE 7-2. AXIAL MACH NUMBER DISTRIBUTION FOR FREESTREAMMACH NUMBER .615

1000 MM

49

a,tO

v

a,

m_r

Z

U<:[

.J<

,1.73

.72

I

.71

MACH NUMBER

MACFI NUMBER

!

,!000 MM 500 MM 0 _ MM

I BODY LENGTI'L MILLIMETERS

FIGURE 7-3. AXIAL MACH NUMBER DISTRIBUTION FOR FREESTREAMMACH NUMBER .72I

! 000 MM

5O

Eio

¢Id

:i

z=u<2.J

u0,J

,8O

.79

.78

.77

.76 _,

.75

.74

.7_

Moo,, .754 I

I

CORRECTED MACH NUMBER I

""*__UUNCORRECTED MACH NUMBER

I

1000 MM 500 MM 0 500 MM

t"t'*"

I ooo MM

BODY LENGTH, MILLIMETERS

FIGURE 7-4. AXIAL MACH NUMBER DISTRIBUTION FOR FREESTREAMMACH NUMBER .754

51

U}

m:E

Z

'I"u<:E.J<_J0..I

.SS

.84

.83

.82

.81

.80

.79

.78

,77

.76

MN .788

Iq.

, //,

1000 MM 500 MM

I

0 500 MM

I BODY LENGTH, MILLIMETERS

1000 MM

FIGURE 7-5. AXIAL MACH NUMBER DISTRIBUTION FOR FREESTREAMMACH NUMBER .788

57

r,*

m_E

Z

Zo<X

,J<u0.J

,gZ

,90 ' Moo" . I

.89 829

.88

.87

.86 I

CORRECTED MACH NUMBER.85

.84

"" I -?---JFo.oo-_o,_o..o,.o.m

,81 _ _"_'_ '"

.80

1000 MM

FIGURE 7-6. AXIAL MACH NUMBER DISTRIBUTION FOR FREESTREAM

MACH NUMBER .829

53

7.2 (Continued)

the Influence of the tunnel walls on the axial velocity distribution. Note

there Is very little difference at the lower Math numbers between thecorrected Mach number data and the uncorrected Mach number data as measured

at the 2 meter wall location. In addition to the velocity calibration

described above, ONERA used the Prandlt-Young correction as discussed in

Reference 20 to further adjust the velocity data before computing an advanceratio, 3. This correction, as defined below, is a compressible correction to

the velocity to account for tunnel wall interference.

l

II l w

V

_4 _1

2(1 + 2 _4)I/2<1 - MN2) I/2

or

Vcoe m 1 - Vmeas

2(I + 2 _4)I/2(I - MN_) I/2

where

and

'1:4

THRUST

p x DISC AREA x V2

PROPELLER DISC AREA

TUNNEL CROSS-SECTION AREA

Note that this correction Is largest where the thrust Is largest and thevelocity the smallest, _.e. low Mach number testing.

54

7.3 Test Rig Corrections

7.3.1 Test Objectives

7.3.1.I To determine the aerodynamic drag on the Prop-Fan spinner as a

function of Mach number for Mach numbers ranglng from .2 to .85.

7.3.1.2 To determine the aerodynamic pressure drag on the test rigcenterbody as a Function of Mach number for Mach numbers ranging From .2 to.85.

7.3.2 Test Procedure

In preparation for the splnner/body tare test, the test rlg and associatedsupport structure were set up as depicted In Figures 7-7 and 7-8. In order

to isolate and evaluate the affects of the presence of the spinner and center-

body In the flow fleld, the LAP blades were replaced with blade stubs, whose

external contours were machined to match that of the spinner (Figure 7-9).

As defined in Figures 7-I0 and 7-II, the Forward end of the centerbody serves

as an extension of the external contour of the spinner, and was designed to

alleviate compressibility losses in the blade root section by reducing the

air velocity passing through the central portion of the Prop-Fan rotor. Atthe inner blade radii, the combination of thick airfoil sections and the

large number of blades presents significant blockage which could lead tochoked flow. The moderation of velocity caused by the centerbody reduces the

possibility of choked flow occurlng at the high subsonic Mach numbers at

which the LAP operates. Figures 7-I0 and 7-II also define the location of

the centerbody surface static pressure taps. There are four rows of

twenty-nlne taps per row positioned a]ong the length of the centerbodysurface. The rows of taps were spaced 90° apart, clrcumferentially.

Figure 7-12 defines the location of an additlonal set of static pressure taps

In the space between the spinner rear bulkhead and centerbody. These tapsconsisted of four rows, ten taps per row, extending radially outward, located

at the centers of equal areas. Each row was oriented 90° apart from another.

Collection of the static pressure measurements from the locations describedabove, while varylng tunnel Mach number, was accomplished u_ing a scanivalve

system. Splnner/centerbody drag data was collected at a total of 42 test

points. All of the data was collected at zero degree Inflow angle.Table 7-I lists the Mach numbers at which data was collected.

55

_g

FIGURE 7-7. SPINNER/BODY TARE TEST

56 .... v'_Pt_i _- PHO'i O&RAPI_I

_tALANCE

GEOMETRICAL

CENTER

BLADE PLAN _

AXIS

WIND TUNN___ ._vO -__513.4"-_

ROD

(_:35 ROD

L:47Z0 _ _1 ._:26 RODSiL._-,o,o _ _::':,o

_o tTi_oB cNOTE: ALL DIMENSIONS IN MILLIMETERS

_) = DIAMETER

L = LENGTH

FIGURE 7-8. TEST RIG SUPPORT STRUCTURE FOR SPINNER/BODY TARE TEST

5?

<Id

_->_0

Z<

r

v

m_

E_

=_XXXX

_ .zzzzz

<w a

_>_PO_

w_ZZZ

m

_J

<

m

Z<b.

(1.

r,,

h.

58

A

Z

oI-UId

Qm<

0Ii

0mI1:ta

ZIdu

i tN

EIn

;/o_.

NN

IL0

W

It

/©C

tIEIEN

W

Uw

v

N

_N__

Z0w

I'-L)i.iU3

6 ........... ¢z <

= ! 0<

Z9I-<LJO.J

Q.<!-

1.1.1n,'

(/'jldn,,Q.W

U<I.I.n,'

>.C]Omn,,b.II-Zb.lU

bJn,,

(3m

k.

_9

Z0

i-.U_J(4

l-b.<

Q0mr_

I-

u

Z0

Uwu)

a

X

).

0m

wh

zwu

F_

F_oN

N

T_F_

o

F_

F_@

F_

F_

F_

F_

r_

F_

m_mm_mmm_mv

== _oo_oooo_o:::::::

I

: m v" _" =', m _ _= _" N N N N N N N N N N

Z

oI-U

0

<

Ob.

0

Z0

b,

0

,¢

-r<U 0

gl Ui.-r <

z0m

H-ULd

]-.1.6<

z

r_:E(/)Z0

H-

U0-I

_L<

hi

E

U

<1-.

I,Ll

<b.

m

1:1

I-ZLd

i

r_

0

60

w

utZ

I-u<0I-.l

I-

(_ NNNNNNNNNIq

i: ..........

0 0

\

m

_m0 N

N m ___m. _. _. m. m ?.--me, iN N N _l (q (q eq _ m |

o o 0 ooo0 00_ I

Z

oI-<u0.J

Q.<I-

W

fflUl

I1.

Um

I-<

<U,I

_1

ILlZZE(I)

r_

I,dIZ

(3b.

61

Table 7-I. Splnner/Centerbody Drag Test Points

SPINNER/CENTERBODYDRAG CORRECTED

TEST MACH

PT. NO.

24224424624825O2552572602622642662682702"/22742762872882892902912932942952962972993OO3023033053O830931031l312314315316317318

499

789

836

834

833

8OO848

.789

.739

.686

.638

•590

•494

•447

•348.244

.201

.201

.201

.201201298298298298298494494590590494

.298

.298

.298

.299

.298

.201.201.201.201.201

62

7.3.3 Discussion and Results

7.3.3.1 Spinner Draq

As depicted in Figure 7-13, the spinner drag (Ds) was measured directly from

the axial force measured by the balance (Fs) with corrections for the backpressure force (FBp) and the losses due to thermal effects in the flexible

coupling (FTH).

Ds = FB - FBp - FTH

The axial force applied to the propeller shaft (F,) was measured by three

dynamometers T,, Tz and T3 In the test rig balance as depicted In

Figure 6-I0. The losses (FTH) due to the thermal effects of the flexible

coupling were measured by strain gages bonded directly to the flex coupling.The back pressure force (FBp) iS the result of the difference between the

free stream static pressure (Po> and the integrated pressures fn the space

between the rear of the spinner bulkhead and the face of the centerbody

(PN), where aN IS an area weighting factor and A, Is the spinner basearea.

4O

FBp = ( _ (aNPN) -- Po) AsN=I

As - .507 mz

Figure 7-14 presents the spinner drag coefficlent (Cso) as a function of

corrected Mach number (McoR). The spinner drag coefficient Is given below

and Is expressed In terms of the spinner drag force (Ds), dynamic pressure(qo) and projected spinner base area (As).

Ds

CsD

qoAs2

As = .519 m

Since the spinner drag data was corrected for the back pressure effects, itrepresents the axlal components of the forces appl|ed by the static pressure

acting normal to the spinner surface and the spinner boundary layer wall

stress acting on the wetted area of the spinner. A quadratic equation wasfitted to the spinner drag data, as shown below and on Figure 7-14, and was

used to compute spinner drag corrections to measured thrust during Prop-Fan

performance testing.

Cso - 2.7699M" - 5.3225M 3 + 3.7727M z - 1.1215M + .2398co. co. co. co.

63

DCBT

•",.'OE.E.U..--T_,/ OS'_,_°P_T.(FTH) DCBT " f(Ps -- PO)dA

FLEXIBLE COUPLING LOSSES

A. SPINNER/CENTERBODY TARE TEST (WITHOUT BLADES)

TApP / / DCB

BALANCE j

MEASUREMENT (FB)

B. PROP FAN TEST PT. (WITH BLADES)

FIGURE 7-13, PROP-FAN TEST RIG FORCES

64

m

o _Ju

1,1,1m

Z

L_<

r_

r_0U

>

Z

u

<

r_

65

7.3.3.1 (Continued)

The spinner drag coefficient, as expected, increased with Mach number forMach numbers in the range from .3 to .85. However, an unexpected decrease indrag coefficient was observed between Mach .2 and Mach .3. The decrease indrag coefficient cannot be readily explained from the data that wascollected. Theoretically, the drag should vary as the square of thevelocity. The large number of data points taken at Mach .2 and Mach .3 doesconfirm that the decrease in spinner drag coefflcient is a real phenomenon.

7.3.3.2 Centerbody Drag

As mentioned earlier, the Prop-Fan was tested in the presence of a centerbody

which was designed to alleviate compressibility losses in the blade rootsections. With the force measurement as shown in Figure 7-13, it has been

shown that the Prop-Fan net thrust (TNET) cannot be directly measured onthe force balance (Reference 21). This Is true because, as discussed in

References 21 and 22, the thrust of the Prop-Fan blades changes the pressure

acting on the centerbody, thereby changing the pressure drag. The presence ofthe body also causes an increase in thrust on the rotor equal in magnitude to

the change In the centerbody drag. The change In centerbody drag is commonlyreferred to as the buoyancy force (BF). The measured Prop-Fan thrust has

been classically referred to as apparent thrust (TApp) and is the largestforce component sensed by the balance. Since the increase in centerbody drag

negates the increase In Prop-Fan thrust, there Is no net increase In thrust

produced by the Prop-Fan system. However, the thrust measurement (Fs), In

Figure 7-13 (Views A and B), does not sense centerbody drag, therefore,

measured or apparent trust (TApp) iS corrected by subtracting the buoyancyforce, which Is the difference between the centerbody drag measured at the

Prop-Fan operating point of interest (DcB) and the centerbody drag measuredat the same Mach number without blades (DcBT),

The centerbody pressure drag (DcBT), was determined by pressure integration

of the longitudinal rows of area-weighted static pressure taps. The

integratlon requires a simple summation because the pressures are measured in

the centers of equal annular areas.

44

DCBT " X (PN - Po)AN + E(PM - Po)AMN=I

WHERE: Po = FREE STREAM STATIC PRESSURE

PN, PN " STATIC PRESSURE AT TAP N, MAN : INCREMENTAL FORWARD CENTERBODY AREA AT TAP N

AM - INCREMENTAL AFT CENTERBODY AREA AT TAP M

66

7.3.3.2 (Continued)

As a result of several last minute profile modifications, by ONERA, to theaft end of the Prop-Fan test rig centerbody, the local static pressure tapspreviously requested by Hamilton Standard were omitted. Therefore, thebuoyancy force correction applied to the apparent thrust is In error.However, the change In the buoyancy force that would have occurred is smallenough to result in no significant effect to the data. Therefore, theequation above is further simplified to the following:

44

DCBT = _ (PN - Po)ANN:I

Figure 7-13 presents the variation of centerbody drag coefficient withoutblades (CcBoT) with corrected Mach number (McoR)- The centerbody dragcoefficient without blades Is given below and is expressed in terms ofcenterbody drag force without blades (DcsT), dynamic pressure (qo) andprojected centerbody area (Acs).

CCBDT

DCBT

qo AcBAcB = .22 m2

A quadratic equation was fitted to the centerbody drag, as shown below and on

Figure 7-15 and was used to compute the buoyancy force correctlon to measured

or apparent thrust during Prop-Fan performance testing.

CcsoT = 7.6393M" - 16.786M 3 + 12.124M 2 - 3.7165M - .2320COR COR COR COR

It is observed from the data that the centerbody drag coefficient decreases

with increasing Mach number. This indicated that the centerbody surface Mach

numbers are Increaslng at a faster rate than the free stream velocity.

7,3.3.3 Performance Corrections

With the Prop-Fan blades installed and thrusting, as depicted In Figure 7-13,

the balance measures the algebraic sum of the apparent thrust, the spinner

drag, and the back pressure force. Therefore, the apparent thrust of the

Prop-Fan Is obtalned as shown In the foIlowlng equation:

T, pp = Fa - F.p + Ds + FT,

67

./

/

ONMN

n,,0U

IE

I

./

,,¢,

0_u

IE

N

÷

n,.o

_u

tO

In,,0

-IE

u_

H

I--QmUU

?m.?

I-amU

U

?

o

c_

r_0u

o

0

r_Wr_

Z

U<

w

UI,I

F,,OU

i-Z[di

O

_D<n-

>OOmn,I,di-ZLdU

_K

ILl

_D

z

68

7.3.3.3 (Continued)

As defined earlier, the centerbody drag is obtalned from centerbody surfacepressure integrations"

44

DCB = E (PN - Po)ANN:!

Also, as mentioned previously, the buoyancy force was obtalned from thedifference between these and the tare run pressure Integratlons"

BF = DcB - DCBT

F|nally, the net thrust (TNET), which is defined as the propulsive force ofthe blades operating In the presence of the spinner and centerbody flow fieldwlthout the increase in thrust due to the mutual interaction, is obtained bysubtracting the buoyancy force from the apparent thrust:

TNET = TApp - BF

69/70

8.0 BLADE STRUCTURAL DYNAMIC EVALUATION

8.1 Test Objectives

8.1.1 Confirm that the SR-TL Prop-Fan was free of high speed blade flutter

over the portion of its operating envelope that could be run in the ONERASI-MA wind tunnel.

8.1.2 Evaluate the IP blade strain sensitivity of the SR-7L Prop-Fan for a

range of blade angles, rotational speeds, Mach Numbers and inflow angles.

8.1.3 Compare the measured and analytically predicted IP blade strain

responses of the SR-TL Prop-Fan for selected operating conditions.

8.2 Test Procedure

Testing of the SR-7L Large Scale Advanced Prop-Fan was conducted In the ONERASl atmospheric wlnd tunnel in Modane, France. The Prop-Fan was mounted so

that the rotor plane was located in the throat of the wind tunnel. The

tunnel throat was eight meters in diameter.

The drive system as described earlier in section 6.2, was supported by rodsand cables attached to the tunnel walls as illustrated In Figures 8-I and 8-2.

A balance and torquemeter, also described in section 6.2, provided the

capability to measure forces and moments acting on the Prop-Fan.

Test rig vibration was monitored by two sets of horizontal and verticalaccelerometers. The accelerometers were located on the drive train housing

in two planes aft of the Prop-Fan.

A stationary aerodynamlc falrlng was located downstream of the Prop-Fan. The

fairlng served as an extension of the external aerodynamic contour of thesplnner. The fairing resulted in an approximate 35% blockage of the flow

through the Prop-Fan rotor.

Power available from the test drive system was significantly ]ower than the

rated power of the Prop-Fan. Therefore, In order to simulate operation at

high power loading conditions, the Prop-Fan was run in two and four bladeconfiguratlons as well as with eight blades. Thls allowed power loadings perblade to be achieved, which correspond to intermediate and high power

operating points with elght blades. The disadvantage of operation with twoand four blades was the negation of the Inter-blade cascade effects which are

present in the eight blade design. These effects tend to be destabilizing inthat they lower the Mach Number at which the onset of classical flutter

occurs.

The mlsslng blades were replaced with stubs in the two and four blade

conflguratlon as depicted in Figure 7-9. The ends of the stubs were machlnedto match the external contour of the spinner. The Prop-Fan is shown in the

two, four and elght blade conflguratlons in Figures 8-3, 8-4 and 8-5.

71

?2

\

tnr_

==.J

Z

u_Z0

oZ

_3

FIGURE 8-3. TWO BLADE TEST CONFIGURAI"IOIM

74

ORIGINAL PAGE

BLACK AND WHITE PHOTOGRAPH

FIGURE 8-4. FOUR BLADE TEST CONFIGURATION

Bi_ACK #,_,;u WHITE PHOTOGRAPII75

FIGURE 8-5. EIGHT BLADE TEST CONFIGURATION

?6

B._'_._ /_,--_u li_,. L PHOT_O_RAPI-f

8.2 (Continued)

The strain gage arrangements for the two, four and eight blade configurations

are shown in Figures 8-6, 8-7 and 8-8. The blade surface gage locations werechosen either to correspond with the points of maximum strain for the blade

normal modes or to provide the distribution of strain along the entire span

of the blade. Blade shank gages were also used to measure vibratory bending

moments.

The Prop-Fan was operated in a Beta control mode during the high speed windtunnel testing. In this mode the blade pltch angle was selected by the

operator uslng a manual control. The test procedure consisted of starting

the Prop-Fan at a blade angle (B3/4) of 20° to 25° and increaslng power toobtain 1200 RPM to 1500 RPM rotor speed with the tunnel flow drive system not

operating. Mach number was then increased in increments. Foliowlng eachincrease In Mach number the blade angle was Increased to maintaln the rotor

speed In the 1200 RPM to 1500 RPM range. At the Mach numbers of interest,

test points were run at two or three different power settings and over a

range of rotor speeds. Blade straln gage data was recorded for thirty seconds

at each test point. Aerodynamic performance and test rig vibration data were

also logged concurrently with the strain gage data.

8.3 Discussion and Results

8.3.1 General Discussion

During initial balance runs of the Prop-Fan, which was conducted at zero Math

number, test rlg critical speeds were discovered at 360 RPM and 540 RPM.Both crlticals were highly undamped, allowing rig vibration to grow rapidly

if operation was attempted at these speeds. The critical speed at 360 RPM

corresponded to the predicted rig first critical. Since both of thesecrlticals were well below the planned test operating speeds, they did not

pose a problem. No other critical speeds were apparent within the testrotational speed range. For the low rotational speeds at which dynamic

balancing was accomplished, dynamometer vibratory stresses were the limitlngfactor rather than rlg vibrations measured by the accelerometers. This may

have resulted from relatively low accelerations causing large displacements

due to the low frequency of the response.

Figures 8-9, 8-10 and 8-II present a mapping of the test points run duringstructural dynamic testlng. The test points acquired for the four blade

configuration spanned the entire planned operating range. Power supplied to

the Prop-Fan by the turbines was limited to 800 KW due to the elevated

ambient temperature in the tunnel.

Operation between lO00 RPM and 1500 RPM resulted in high rig vibration forthe two blade and the eight blade configurations. Therefore, this operating

region had to be avoided. The vibration frequency was not IP and was not

believed to be caused by the Prop-Fan. With the exception of the 2 bladed

configuration at 3° inflow angle (_), operation at inflow angles of 3° orlO: also resulted In a high level, low frequency vibration which precluded

testing at those conditions.

77

FORWARD TO AFT VIEW

®

BLADE CAMBER SIDE VIEW

3F

®

GAGE r/R TIP

! .44

2 .77

3 .64

4 .69

5 .71

6 .84

7 .71

8 .84

9 .34

I0 .44

I 1 .44

12 .57

13 .78

GAGE NUMBERING CONVENTION - X XX

FIGURE 8-6. STRAIN GAGE ARRANGEMENT FOR TWO BLADE CONFIGURATION

78

®I

FORWARD TO AFT VIEW

®

BLADE CAMBER SIDE VIEW

5F

GAGE r/R TIP

! .44

Z ,77

3 .84

4 .69

5 .71

6 .84

7 .71

8 .84

9 .34

10 .44

I I .44

12 .57

I 3 .78

FIGURE 8-7.

GAGE NUMBERING CONVENTION -X XX

STRAIN GAGE ARRANGEMENT FOR FOUR BLADE CONFIGURATION

79

®!

FORWARD TO AFT VIEW

® /

®

o,,;BLADE CAMBER SIDE VIEW Z .77

3 .84

4 .69

5 .7!

6 .84

7 .71

8 .84

g ,34

10 .44

1 1 .44

12 .57

! 3 .78

GAGE NUMBERING CONVENTION -X XX

B-A°E.o!.GAGE NO/

FIGURE 8-8. STRAIN GAGE ARRANGEMENT FOR EIGHT BLADE CONFIGURATION

8O

1000

8OO

SO0

qoo

20O

8OO L06O

SR-7L 2 NRT TEST POINTS

+

×

+

Z_

12O0 I qO0 t600

RPH

1800 2000

3000

25OO

ici

z 2000

o

t 500

i

f 1000

SR-'7L 2-HI::I'I" TEST DFITR(HODFINE]

; cp

t_

O (

poO

!

i

0800 1000 t 200 I qO0 t 600 1800 2000

RPH

FIGURE 8-9, TEST POINTS AND BLADE SHANK MOMENTS (2 BLADE CONFIGURATION)

81

==w

.=

Q.

I000

BOO

80O

qO0

20O

8OO

SR-'/L q WRY TEST POINTS

t I 1 IP"l MN= .2-.35 (_ DEG. INFLOW

C_ MN=.4-.65 {_ D£G. INFLOW)

z_ MN= .7-.836 (_ DE:G. INFLOW}

I

Om

B p-

12001000

If

l_iO0

I

,t, . t

1$00

.-..v,.j

1800 200'

RPN

r

l

_=

Ii.

3000

25OO

2000

1500

1000

500

SR-7L q-_Rr TEST DRTR {NOORhrE_

I I i / l I

1 _ IRP-GAG£ IF (_ D£G. INFLOW)

2 (3 ]RP-GAG£ 3F (_ DEG. INFLOW)3 A IRP..GAGE SF (S_DEG. INFLOW)

thI)

mA

° !i

........ J_ .

1000

1],1

rl (9

)

i il........... i ............

0 ..... | .... i ....

000 1200 lqO0 1600 1800 2000

RPM

FIGURE 8-10, TEST POINTS AND BLADE SHANK MOMENTS (4 BLADE CONFIGURATION)

82

I000

8OO

v

m.

6O0

2OO

SR-TL 8 NR¥ TEST POINTS

I I I I I

[] MN: .19-.35 (_ DEG. INFLOW)

(_ MN= .4-.65 (_ DEG. INFLOW)

_,N= .7i .. (_,=)_G.jINFLOW)

I

I ITI

O I) I

ot1

800 1000 1200 Iq00 tS00

RPH

1800 aaoo

3000

2500

2000

i 15oo

m

iooo5VA.

50O

i

I

2 O

3 A

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

800 1000

SR-'7L 8-NRY TEST DATA (HODRNE)

J

_eF._GE: 3F'(_ ocG. INFLOW)lAP-GAGE: 4F (_ DI[G. INFLOW)

[I RP.-GAGE:I SFi(_ DE:?. INFLOW)I

1200 lqO0

RPIq

o

!']

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

600 1800 2000

FIGURE 8"1 1. TEST POINTS AND BLADE SHANK MOMENTS (8 BLADE CONFIGURATION)

83

8.3.1 (Contlnued)

Mach numbers above .73 resulted in negative thrust for the eight blade

configuration due to the low power available from the turbines. Since the

balance was not designed for negative thrust, the maximum Mach number for the

eight blade case was limited.

8.3.2 Data Reduction

The strain gage signals were recorded on magnetic tape for each of the gages

shown In Figures 8-6, 8-7 and 8-8. These signals were then fed through a

peak stress (strain) converter, a device which determines the peak vibratory

strain occurring in a unlt of time (.l seconds for thls data). The output ofthe peak stress (strain) converter is then statistically analyzed to obtain

the average amplitude (mean) and standard deviation of each signal. The IRP

(infrequently repeating peak) Is defined, for the purposes of this report, as

the mean plus twice the standard deviation. It Is a conservative measure of

the strain amplitude normally used to estimate blade fatigue life.

Figures 8-9, 8-10 and 8-II show summaries of the conditions analyzed along

with plots of the IRP shank moments for the 2, 4, and 8 bladed configurations.It can be seen that the highest values recorded were for the two bladed

configuration wlth angular inflow (3 degrees). It Is also noted that there

are significant blade-to-blade dlfferences. However, it was found that the

data contained significant high frequency noise (several thousand HZ), whichtended to artificially increase the IRP values. In general, no significant

differences In strain gage signal were revealed between the two, four and

eight blade configurations at uniform inflow angle, for the same gage (3F).

Figure 8-12 shows an example of the signal to noise problem and itsvariability between different gages. Because of the low response levels, the

data reduction was not repeated with the noise filtered out.

Tables 8-I, 8-11, and 8-111 show the conditions of the runs made along with

some sample IRP tabulations, showing blade-to-blade variations.

Table 8-1V shows a tabulation of peak IRP values for each gage versus thelimits determined based on the threshold at which blade fatigue damage begins

to accumulate. Because of the high noise content, all of the IRP values

(Figures 8-9, 8-10, 8-II and Tables 8-I, B-If, 8-IIl and B-IV should beinterpreted as conservative upper bounds on the strains actually felt by the

blades. As such, the gages shown to be exceeding the Iimits should actually

be Interpreted as below the limits with the noise filtered out.

84

Z0F-<

I

z0ULdQ<,..I

-i

Z<

,J

o

0'11_ZoZ o

_:e4<I1_ IIuJ ZzlEO_

XW

<

-I,

0ww5z

<z_0

:E<Iv

01-1J

0WU)

z

W.J(.9Z<

0..J14=Z

o

C3

Inl,,IJLO<LOl--ZILl

0

vZ<"r

I.I.I(/I

P<-I14.n_0I.I.LI.I(/I

0z0I-,..I<Z(3(/II,i.0Z0

n_<

0U

e,i|

co14Jn_

(3I.i.

85

ZOb-

o=

m

u.ZOo

t_

-Iw

XOF-

-/

I-

n_

W-I

t_X

A

n_O

(/1XOm

!-

Oo

Z

l-

t_

0.O

0

W-J

86

ORIGINAL PAGE I_.

OF POOR QUALIT_

Z0

<n_

z0L_wr_<..]m

Z0

_I

nn_

ILl

B.

r_Z<

Az:En_0

Z0

i-

Z0u

zI--

ulI!.0

W-JII1

!,-

87

A

ZV.ZOo

w

88

ORIGINAL PAGE rS

O]_ P.OOR QUALITY

ZOY-<rr

Om

B.

ZO_9ILl

<-]Ill

ZOI-.<-]

m<F-11.n,

Ill..l

<

QZ<

AZ

O

ZO

m

ZO(3(.9Z1-<n,

I1.O

m!

0D

1.11--1m<i--

ORICENAI/ PACE IS

OF, POOP, QUALITy

89

Gage Description

TABLE 8-]_Z'. MEASURED VIBRATORY PEAKS VS. LIMITS

Peak IRPStrain observed

for 8-way(0' inflow)

fraction o! limit

Peak IRPStrain observed

for 4-way(O' inflow)

fraction of limit

Peak IRPStrain observed

for 2- way(O' inflow)

fraction of limit

Peak IRPStain observed

for 2- way(3' inflow)

fraction of limit

Radial Bending (44%r/R) Gage 1 .43 .33 .79

Radial Bending (71%r/R) Gage2 1.12" .48 1.05"

Radial Bending (84%r/R) Gage3 ,22 .45 .46 .88

Trailing Edge Bending Gage 4 .55 .6 1.01 *T. ,..

Shear (71%r/R) Gage5 .04 .07 .05 .11, i

Shear (84%r/R) Gage 6 .05 .07 .08 .12

Chordwlae (71%r/R) Gage7 .22 ,21 .16 .26

Chordwise (84%r/R) Gage8 .36 .23 .20 .33

Radial Bending (34%r/R) Gage 9 .41 .30 .72

Shear (44%r/R) Gage 10

Chordwise (44%r/R) Gage 11 .24

Radial Bending(57%r/R) Gage 12 .26 .45 ,49 1.11 *

Radial Bending(78%r/R) Gage 13 .29 .29

Flatwise Shank Moment - Gage F .29 .47

,22EdgewiseShank Moment - Gage E

.52 1.03"

.76.28

.17.19 .27

* It is fell that because of the noise present, these values can be Interpretedas within limits.

Notes: 1. The gage limits were set based on the strains reachinga Ikmslmld at whichblade fatigue damage begins to accumulate (vibratory strain _lmrimposedupon steady strain).

2. Becauseof the amount of noise in the strain signals, the Blipvalues listed here should he interpreted as conservativeu_ Immds onthestrains actually felt by the blades.

9O

ORIGINAI_ PAGE IS

OE POOR QUALIT_

8.3.2.1 Angular Inflow

As previously noted in section 2.0 and paragraph 8.3.2.1, the only angularinflow data obtained was for the two bladed configuration at 3° inflow

angle. A spectral analysis of each strain gage signal was performed using an

FFT (Fast Fourier Transform) algorithm. This analysis employed a Hanning

window to help overcome leakage effects. Correlatlons were then made with

the predictions of IP response.

Table 8-V shows the P-order content obtained for the 3 degree conditions

tested (shank moment and radial bending). It can be seen that the 2Pcomponent averaged about 25% of the IP component (with 2P/IP approximately

.54 near 1500 RPM) and that the 3P component was generally much smaller. It

Is noted that at least a portion of the excitation force driving the 2P

response may have come from the test rlg drive system. Twice per revolutionvibration Is characteristic of a shaft with universal joints. It is also

likely that nonlinearities In the aerodynamic excitation, possibly due to

observed vortex loading phenomena, or tunnel flow irregularities in the test

section, are causes of higher order excitation. The aerodynamic analyses

used for the predictions do not consider this effect.

The data digitized for the spectral analysis was sampled at constant time

steps. Even small variations in RPM are known to result in substantially

lowered P-order magnitudes. By synchronously sampling the data (digitizing

at constant fractions of a cycle instead of constant time step) this problemcan be overcome. This is known as a 'speed corrected' spectral analysis.

However, since the RPM trace was not available for this test, this was not

possible here. An alternative procedure was used for several strain gage

results to double check the amplitudes obtained from the non 'speed

corrected' spectral analysis. A calculation was made of the averageamplitude of a band pass filtered (IP frequency mid band) strain record.

This procedure is cumbersome but should produce conservative strains because

the amplitudes are those generated by a peak stress converter (with a finitereset rate). Table 8-VI shows that the band pass filtered values are

typically I0% higher than the spectral values. Even with no RPM variation,

the spectral magnitudes are known to be up to 16% low, due to the effect

known as leakage (with a Hannlng window as employed here) In digital signal

processing terminology. It is concluded that the actual magnitudes arewithin lot (higher) of the spectral values.

Figures B-13 and B-14 show the trends of IP response (shank moment and blade

bending) with respect to power, RPM and Math number for the 2 bladedconfiguration at 3° inflow angle. As expected, the response increased with

power and Mach number. Variations of RPM are influenced by system

resonances. Figure B-15 shows the variation of bending strain with spanwlselocation. Again, the trends are believable and consistent with the data

collected during the four and eight blade structural testing.

91

ORIGINAl5 PAGE Ig

OE POOR QUALITY

i+ + + + ÷ + +

i0 o,0 0 0 0

mLK_ 010 0:0 0 0 0 0 0 0 0 ¢1, 0 O 0 0 0 0 O. 0 _ ¢_.

''_ o o o o o o o o olo o oloI

q

_!_ _ _,a _,a o .., o _ .... o ........

• . ......... _ " ¢/ _.

J

Zi

0

!-<I-

0

<

Zu

<

l-

u.0I-

Z/,iI-

z0U

r_tdoe_

9I!.

Ill.J

<I-

• . , ........ -- .

• _,_.:_i_,_.: ....... _,_,_:,s._ .... _!_ _ -,_ ....

92

z0_,,_ _.

0_ :_._--

o o o. o o.

o.

o

o

o

o o

_.0 _.- ._ _ ._ _ .:

¢_ _ "_- o. . .

_im,_ _ : o. o. o. o o.

IP III _,'

_o .g.° |

C

£

0

C

0

v

C

E

93

A

II

:Ev

>-l-

p

zwini-

_X.,0

0¥Lz

O=1. tn

td

iiiiiILl

00 O

O O_ m

- _

\

0 O 0 o

0 0 0 0

(..-13/i,M-N) AJ.IAI.LISN3S .I.N_II_IOIAI• INVHS 3SIMJ.Y"] 4 d!

oo

M) rv

O

O

O

z

z

OIEIE

_,<0"1"l. ln

ul

bl

A •i!

v /

_= i :,

_ .. ",, __

/

/ /

Ill

_0O O O O OO O O O O

(_13/l_l-N) XJ.IAI.LISN_IS J.N31_Oi_I_INVHS 3SIMJ.V'I_i d i

O

IU

0

n,

Q[d

_.J

<z

)!0

oo

o o 00o

I-0n,

oO

O

.I-I-

_0_. m

Z>_P

<_u

O_-

h,

_Zm

iiiii

IIIII

lllIl

l_!ilIiiiI -

!i!!!

0 0 0 0 0 00 0 0 0 0 0

(_3/_°N) X.LIAI.LISN3S .Z.N3WOW_NYHS ]S|MJ.V'IJ d|

O

Oo

O

O

r_

o <

OIn

O

o

I.=Z14J_E0

Z<-r(n

0-I!.1.Z

n,<-I

L9Z<

ILlQ<.J0IIN

6'1Z

oIg)QZ

I-

_4

II0W

l,

94

!

i

z

r

iw

I!!!!

"1

\\ ",..,.\.!._

(3]/N|BMLSO_31N! IL]AiLI_N]$ N[UVLS dl

!t-

J.

z

r,.

Y

1t[

i

||111

I

\r :,

/it// .." /

II|#IIItAI||OM_IW LLIAILIINH N1¥¥[| dl

..'. iiiiiI1|11

_ -ttt|i

z 111111

N:

!

_LIAILISN3S NIUHIS d!

Zw

ZLU

--I<B<n,

-II,Z

n,<-I

(.9Z<

C]<..Im

(nZ

Ii.

l-

,i"r

U,

9G

,'7

zw

g

zw

160

lqO

lEO

100

OO

BO

_0

20

SRTL=IP BENOING STRQIN SENSITIVITY VS. SPRN(MN=.S.5OOKW}

L

o

i _ LSOO

IBOOT" ;'II

,I

1

!

l

f

i ....

I

rI

i

)._ooo

FRRCTION

! ]i i

i !

ii

t " t

I J!Ii

I

OF SPRN

l,O000

,T

z

g

LI

160 ,.

lqO

120

I00

> BOl-,-

:z

ouJ

o,.

SRTL=IP BENDING

..... J 700- '- IOOO 'M

STRRIN SENSITIVITY VS. SPRNtMN=.TS.570KN}

I

"%% x

\ ,!°_.°. - •

".,[-A - • _ :

it

o 6 " " o._o6- - o;.oGo " oZeo_o - =.ooooFRRCTION OF SPRN

FIGURE 8-15. RADIAL BENDING VS SPAN, 2 BLADES, 3° ANGULAR INFLOW

96

8.3.2.1 (Continued)

The plotted values were normalized wlth respect to E.F. (excitation factor)which Is defined as:

E.F. = _ (VT1644.8)2(p/po)

Where _ is the inflow angle In degrees, VT the true airspeed In Km/HR,

p the alr density, and po the alr density at standard sea levelconditions.

Only the IP component of the response has been plotted to investigate trends.

The IRP values include the higher order content as well as unfiltered noise.

8.3.2.2 Analytical Method

A finite element model of the SR-7L blade was developed during the design

phase. The same model was used here to predict IP response for correlation

with test. The steady state aerodynamic loads were calculated using the HS

computer code H444. The steady alr loads along wlth centrifugal loads, wereapplied to the model which was analyzed using the in-house finite element

code, BESTRAN (Reference 23). From this solution the differentlal stiffnessmatrix was obtained and added to the structural stiffness matrix. The

dlfferential stiffness matrix represents the addltlonal stiffness of theblade under centrlfugal loading. The unsteady (IP) alrloads were then

calculated uslng the HS computer code H337. The loads were applied to the

finite element model and displacements and surface strains predicted.

Mohr's circle relationships were employed to calculate strains In thedirection of the gages for correlation with test. Shank moments were

calculated us|ng the reaction loads at the root of the blade and assuming a

linear variation of moment up to the shank location.

The H444 and H337 codes are both aerodynamic 'strip' analyses that have been

calibrated to predict the steady and unsteady alrloads on swept Prop-Fan

blades. The H444 code employs a Goldsteln formulatlon to calculate inducedvelocltles, whereas the H337 code uses a skewed wake theory. Two-dlmenslonal,

compressible alrfoll data was used by both codes to predict the aerodynamic

loads. Trlangular plate elements were used In the BESTRAN code to do the

finite element analysls. Each component of the composite blade (i.e., shells,spar, foam, sheath) were represented by separate plate elements. They were

a11 tled together using constraints based on plane-sectlons remaining plane.Further detail on the codes and their use can be found In the LAP Blade

Design Report (Reference 14).

Figure 8-16 Illustrates a Campbell diagram of the SR-7L blade. Shown are

BESTRAN frequencies compared wlth test values. It is assumed that the RPM

range tested was far enough away from IP resonance that the Influence ofdamping on strain magnitudes can be neglected.

97

N'r

).uZb4

0bJ

ml<¢3I"<21

300

ZS0

ZOO

I_0 --

100

O-FREQUENCIES FROM WPAFB TEST

I-PRE-TEST PREDICTION (BESTRAN)

oz f. vo- I I I

o soo _ooo isoo zooo zsoo

ROTATIONAL SPEED"" RPM

FIGURE 8-16 COMPARISON OF SR-7L NATURAL FREQUENCYTEST RESULTS TO PREDICTIONS

9B

@.3.2.3 Test vs. Analysis

Three cases were chosen to analyze and compare to test. The condltions werepicked to study the influence of power and Mach number variations on the

predicted strains. Case l (run #I176) was at 1673 RPM, 794 KH, .5 MN.

Case 2 (run #1190) was at 1698 RPM, 244 KN, .5 MN. Case 3 (run #1214) wasat 1707 RPM, 829 KW, .724 MN.

The IP gage strains were calculated using the methods outllned in the

prevlous section. Comparisons were then made to the values obtained from a

spectral analysls of the test data. See Figures 8-6, 8-7 and 8-8 for gage

locations. Figure 8-17 shows the variation of radial bending straln as afunction of spanwise 1ocatlon. As can be seen the correlation is quite

reasonable wlth the In-board predicted strains being about 30% too high.

Nowever, as previously noted in paragraph 8.3.2.1, the 'measured' strain

levels are felt to be up to I0% low.

Figure 8-18 shows the correlatlon of shank moments and root radial bending

(gages F and l) as a function of power, (holding RPM and Mach number). Both

the test and analysls show the expected trend of increased IP response wlth

Increased power. The test values are consistently below those calculated.

Figure 8-19 shows the same quantities plotted against Mach number (holdingRPM and power). The trends are consistent with expectations (higher response

with increased Mach number). Table 8-VII shows a comparison of all the

strain gage results for the three conditions. As noted from the plots, thecorrelation with radial bending strains and shank moments Is quite

reasonable. However, the correlation Is rather poor with the gage placed

near the trailing edge, and also with the chordwlse and shear gages. This Is

consistent wlth previous experience. It |s felt that these values are

affected more by local distributions of aerodynamic loads, whereas the radial

gages are strained by an integration of loads outboard of a given gage.Further study Is needed to better define the local load distributions In

order to more closely correlate wlth the measured strains.

8.3.2.4 Blade Flutter

Prior to the high speed wlnd tunnel test, an unstalled flutter analysis wasconducted for the two, four and eight bladed Prop-Fan configurations. The

analysis indicated that the Prop-Fan would be free from unstalled flutter

throughout the wlnd tunnel test operating envelope. The tests confirmed the

predictions In that no flutter or tendency to flutter was measured orobserved. The wlnd tunnel test results were also a good Indlcator that no

unstalled flutter would occur durlng the subsequent flight test program.

Though the maximum Mach number achieved for the elght bladed configuration

was 11mlted to .73, the resultlng dynamic pressure at the tunnel's effectivepressure altltude of 4,360 meters (14,OOO ft.) Is greater than the maximum

dynamic pressure expected for the f11ght test.

99

J

.=.

z:

(3

i L°

-~

E

z

ru

| J

0

oJ/

/

//

0_

Jl |NI_LS;OV3lll '1¥10H¥ 4!

!

|

c

_oe.

z

u)

_ -,,,

/,,-- /

0

| l |

o

o

a

Z<n

>

Z

/mZ111III

-I<i

Q<n,

m

ul>-..I<Z<(n>

p.

bJ!--

"7cobJ

m

lOO

:E

i-,Z&d3E0X

z

Z

L

T_'STVS. ANALYSIS-- EFFECT OF POWER (AT 1700 RPM;MACH NO. =.5)

2500

1 _ TIST _FLATWI ;E(GAGE 3F}

2 _) _IALYSISiFLRTWI'iE(GAGE F I3 ""_'" ST TEDGEWI _E {GAGE 3E}

Z000 U _- AIAI Y_TSIFn_FWT ;F (C,AC, F F

1500 !

RUN #C 1176

lOOO !RUN # 0 , _.. I

I '

L. i0 ......... _:':'"':"':",""L": "''°'" -'_':2"L ................... _ . . . { ....

o zoo 400 600 8oo iooo

POWER (KW)

600

5OO

TEST VS. ANALYSIS-- EFFECT OF POWER (AT 1700 RPM; MACH NO, : .5)

1 _ TEST -AVG. F 31A_O 7] I

2 C ANAL'SIS- G GE 1

Z

<n,J-m0

Um

a.m

400

300

ZOO

IO0

0

C RUN #

1176

RUN #1190 _ _'Z

0 200 400 600 800

POWER (KW)

t ....

1000

FIGURE 8-18. TEST VS. ANALYSIS (EFFECT OF POWER)

lOl

:S

l-Zld:E0

Z<Z

a.

2500

2O00

1500

10o0

500

TEST VS. ANALYSIS-- EFFECT OF MACH NO. (AT 800 KW; 1700 RPM)

2 O ALTSI5 FLATWI IE(GAGE F | o RUN1214#

3 - -,_,.-. ST "DGEHI _E (GAGE

__L _-J- AI Y_IS ENr,FWT _Ff_,Ar, F F

J .,.

RUN # J!176

b .......... ! ...........

0.4000 0.5000 0.6000 0.7000 0.8000

MACH NO.

0.9000

Zm

<r,,I-u)O

u:ia.

600

500

4O0

300

200

TEST VS. ANALYSIS-- EFFECT OF MACH NO. (AT 800 KW; 1700 RPM)

I _ TIIST-RVG OF 31 RND 71

2 O QIIRLYSIS GRGE RUN #

O 1214

RI_IN # /!176

f

IOO !0.4000 0.5000 0.6000 0.7000 0.8000 0.9000

MACH NO.

FIGURE 8-19. TEST VS. ANALYSIS (EFFECT OF MACH NO.)

102

ii ,iE

,: ..: _ .: - ,.: _ 0, :

.'_

i|_ '",jg e-

l

_i ° ""

_1_-,_ o o o "-,!

i ,.:e_

i_; " li #fiiii; "_ _ _:ill l_, lit lit.i lil

i

li.,z,=,,

illm(il>-,.,I<Z<

>I-,

I.LlI-'1.1..0

0

,<O,.

0

I,Ll

-1lid

<I-

103

8.3.2.4 (Continued)

The maximum Mach numbers achieved for the two and four bladed Prop-Fan

configurations were .80 and .84, respectively. Though these configurationsdid not permit the effects of blade cascade to be evaluated, they did

demonstrate the Prop-Fan's resistance to unstalled flutter onset for

subsequent blade surface pressure testing.

8.3.2.5 Whirl Flutter

Due to the flexibility of the ONERA test rig, whirl flutter was a concern forthis test. These concerns were reinforced when initial calculations of the

installations' stability showed whlrI flutter to occur within the operating

envelope. For the Inltlai computations, the forward support structure, shownIn vlew A of Figure 8-I and views A and B in Figure 8-2, had a three rod

configuration that resulted In undesireable symetrical pitch and yaw

stiffness. To stabilize the two and four bladed configurations, a two rod

support structure, as depicted In vlew A of Figure 8-I, was utilized. Thetwo rod configuration, which provided asymetrlcal pitch and yaw stiffness,

resulted In a more stabilized system. The computations also demonstrated the

need to add cables, as shown In view A of Figure 8-2, for the eight bladed

configuration to increase structural damping while improving the distributionof the pitch and yaw stiffness. The end result of these rig modifications

was successful completion of the wind tunnel test with no whirl flutterinstabilities encountered.

104

9.0 AERODYNAMIC PERFORMANCE EVALUATION

9.1 Test Objective

To measure the aerodynamic performance of the SR-7L Prop-Fan for a range of

blade ang]es, rotational speeds and Math numbers.

9.2 Test Procedure

Aerodynamic performance data (thrust and power) was measured concurrentlywith the structural dynamic data. As a result, the data Includes losses In

performance resulting from the presence of the strain gages and strain gage

wires on the aerodynamic surfaces of the blade. Installation of the gages

was accomplished so as to minimize these losses.

Power avallable from the two turbine engines driving the Prop-Fan in the wlnd

tunnel was slgnlficantly lower than the rated power of the Prop-Fan.

Therefore, testing was conducted using two and' four blade configurations as

well as eight blades.. The two and four blade configurations permitted

operation at power loadlngs per blade that correspond to high and Intermediate

power operating points respectlvely for the eight blade Prop-Fan deslgn. Thedisadvantage to thls approach is that the eight blade performance cannot be

easily extrapolated from the two or four blade test results. This is due to

the aerodynamic Interact|on between the blades. This interaction is more

significant for the eight blade design than for the two or four bladeconflguratlons, due to the reduced spacing between blades. Unfortunately the

aerodynamic performance data for the 2 blade configuration was llmlted, and

rather than attempt to project the full power elght blade performance from thethe two and four blade test results, the available results are compared to

analytical predictions. This serves to verify the analytical techniques andprovides confidence that the predictions for the eight blade high power

performance may also be correct.

The missing blades were replaced with stubs (see Figure 7-9) in the two and

four blade configurations. The ends of the stubs were machlned to match theexternal contour of the spinner.

The Prop-Fan was operated in the Beta Control mode during the aerodynamlc

performance testing. In this mode, Hamilton Standard personnel were able tochange the blade pitch angle during testing by means of an Increase/decrease

pltch switch located in the control room. For a fixed Mach number and a

constant power supplied by the turbines, the Prop-Fan rotational speed was

varied by increasing or decreasing blade pitch angle. At the Mach numbers of

interest, aerodynamic performance data was collected for two or threedlfferent power settings and over a range of rotational speeds.

Approx|mately 140 performance data points were collected.

105

9.3 Discussion and Results

9.3.] Data Reduction Procedure

The blade pitch angle (B314) was measured and recorded for each test point.

An electrical signal proportional to blade angle was provided by means of a

potentiometer mounted on the rotating side of the Prop-Fan. The potentiometershaft was positioned by the rotation of the No. 7 blade retalning ring

through a cable and pulley arrangement (Figure 5-3). Flgure 9-I demonstrates

the linearlty of the output of the blade angle measurement potentiometer as

monitored during a pre-test calibration check.

The Prop-Fan rotational speed was measured by use of a IP pickup. The sensor

was triggered by a gear _K)unted on the test rig drive shaft. The rotational

speed was averaged over ten revolutions.

The power absorbed by the Prop-Fan (BHP) was determined by multiplying thetorque supplied to the Prop-Fan (QcoR) by the rotatlonal speed (N). Torque

supplied to the Prop-Fan (QcoR) was computed by subtracting the measuredfrictional losses in the balance (Q_L) from the torque measured by the

torquemeter (QMEAS).

BHP = N x QcoR

Where: OcoR - OM_s - Q_L

The net trust (QNET) determlned during testing is the unlnstalled thrust of

the Prop-Fan rotor, operatlng in the presence of a spinner and centerbody andIs computed from the following equation"

QNET = F8 + Ft. - Fs, + Ds - BF

Where"

F, = axial force measured by the balanceFTH = losses due to thermal effects

F,p = back pressure force

Ds = spinner dragBF - buoyancy force

The temperature correction term (FTH) compensates for the effects of

changes In temperature on the balance strain gages.

The back pressure term (FBp) corrects for the increase In measured thrust

due to the differential between the pressure behind the spinner bulkhead and

the free stream pressure. The back pressure force Is calculated by

multlplylng the difference between the average pressure measured by the tapsshown In Figure 7-I0 and the free steam pressure by the pro_ected area of the

spinner bulkhead.

106

100.

-ZOO0.

I

-1500. -I000.

7_

BLAD£ ANGL£ (_3/4)

-500. O. 500.

PO'rENTIOMETER OUTPUT (mV)

!000. 1500. 2000.

FIGURE 9-1. LAP BLADE ANGLE POTENTIOMETER OUTPUT

107

9.3.1 (Continued)

The splnner drag force (Ds) is computed by multiplying the spinner drag

coefficient by the free stream dynamic pressure and the reference area of the

spinner, as defined in paragraph 7.3.3.1. The spinner drag coefficient was

determined as a function of Mach number during the spinner drag test

(Figure 7-14).

The buoyancy force term (BF) eliminates the apparent Increase in thrust

caused by the interaction between the Prop-Fan rotor and the centerbody. The

buoyancy force is determined by measuring the centerbody drag at theperformance test point of interest and subtracting the centerbody drag without

blades for the same Mach number. The centerbody drag was determined during

performance testing by integrating the difference between the pressures

measured by the taps shown in Figure 7-I0 and the free stream static pressure

over the surface of the centerbody. (See paragraph 7.3.3.2). The centerbodydrag without blades was computed by multlplylng the centerbody drag coeffi-

c|ent by the free stream dynamic pressure and the centerbody reference area.

The centerbody drag coefficlent was determined as a function of Mach numberduring the splnner drag test as depicted in Figure 7-15.

Mach number was established from the ratio of static pressure (measured four

meters upstream of the Prop-Fan rotor) to stagnation pressure. Staticpressure was also measured In the plane of rotation as a backup. The ratio

of static-to-stagnation pressure was correlated with data taken during a

pre-test ca]ibratlon in order to compute the Mach number (see Paragraph 7.2).

The Prandlt-Young correction was applied to the computed Mach number to

compensate for the effects of the tunnel walls and the thrust produced by theProp-Fan.

9.3.2 Data Presentation

The most complete aerodynamic performance data was acquired for the four bladeProp-Fan configuration. Operational problems encountered with the test rig,

while running the two and eight blade configuration, limited the operating

envelope for these configurations. The r|g operational problems wereaddressed earlier In section 8.0. The structural failure of the centerbody

during the test also significantly delayed the test program. Therefore, In

order to expedite the program, the test points were limited to the boundariesand a few Interior points of the operating envelopes for the two and eight

blade configurations.

108

9.3.2 (Continued)

The aerodynamic performance data was nondimenslonalized for analysis

according to the following set of equations.

(POWER COEFFICIENT)

Cp=

BHP (po/p)

5.674 (ND/IOOO)3D 2

(NET THRUST COEFFICIENT)

CTNET =

TNET(Po/p)

340.42 (ND/IOOO)2D 2

(ADVANCE RATIO) V

J = 60 --ND

Where:

BHP = power, KW

TNET = net thrust, N

D = Prop-Fan diameter, m

po/p = density ratio, sea level to ambientN = rotational speed, RPM

V = free stream velocity, m/sec

It should be emphaslzed that the performance data was acquired only during

structural testing where blade angle was continually varied to maintain a

constant power level. Accordingly, the data was plotted as curves of powercoeffIclent and net thrust coefficient versus advance ratio to eIimlnate

blade angle (63/4) as a variable. If desired, thls data can be converted to

efficiency (n) by the relationship:

(EFFICIENCY) CTNeT

q= X,,1Cp

The power and thrust coefficlent data as a function of advance ratio are

presented for the four blade conflguratlon in Figures 9-2, 9-3 and 9-4 andfor the two blade conflguratlon In Figure 9-5. The data was then

cross-plotted to derive the more conventional maps of CTNET versus Cp.Plots of CTNET versus Cp for the two, four and eight blade cases are

presented In Figures 9-6, 9-7, 9-8, 9-9 and 9-I0. The llmlted scope of the

two blade and eight blade data Is apparent. Data taken at 3° inflow angle

rather than at 0° Is presented for the two blade conf!guratlon, because abetter dlstrlbutlon of test points was run at that angle. Examlnatlon of

these plots shows that the net thrust coefficient exhlblts smooth consistent

varlatlons wlth power coefficient and advance ratlo.

109

SR-7L WIND TUNNELPERFORMANCE DATA

• O9

O8

.O7

.o6

.05

.04

CTNET

.03

.O2

.01 | •

/If

CTNET

Cp "" "--"

_Z_ --800KW

O =500KW

C] =2!;0KW

//

//

//

//

//

I P/

//

of/

f

1o

/

4J

0 im

0.6 0.8 1,0 !.2 1.4 1.6

ADVANCE RATIO, J

FIGURE 9-2. Cp AND CTNET VS J, 4 BLADES, MN = .2, _J= 0°

110

SR-7L WIND TUNNELPERFORMANCE DATA

.24_

.ZZ

.Z0

.1

.16

.14

CTNET

.IZ

.10

.O8

.06

.04

.02

1,0

_ .8

.7

Cp

o6

.5

.4

m ,3

.Z

m .I

1.8 2.0 Z.Z 2.4 Z.6 Z.8 3.0 3.2 3.4 3.6

ADVANCE RATIO, J

FIGURE 9-3. CpANDCTNETVS. J, 4BLADES, M N=.5,¢=0 °

111

SR-7L Wi ND TUNN ELPERFORMANCE DATA

.!o

.09

.o8

.o7

CT NET

.06

.05

.O4

.03

.02

.01

.10

.9

.8

.7

Cp

.6

.5

.4

.3

.2

.I

P = 800KW

O P = 500KW

[] P = 300KW

,,I l

f

3.2 3.4 3.6 3.8 4.0 4.2 _4 4.6 4.B

ADVANCE RATIO, J

FIGURE9-4. CpANDCTNETVS. J, 4BLADES, M N=.8,_=0 °

112

SR-7L WIND TUNNELPERFORMANCE DATA

CTNET

1.0 _ .50

.09 _ .45

.08 _ .40

.07 _ .35

.06 _ .30

Cp

.05 _ .25

,04 -- .20

.03 _ .15

.02 _ .IO

.01 _ .05

I I f'_

CTNET / .

Cp m_

/A P'- 800KW

O P'50OKW

[] P=250KW _ jJ

,. .t /

J

_i 1 i ,_

_m" /

1.9 2.0 2. I 2.2 2.3 2.4 Z.5 2.6

ADVANCE RATIO, J

FIGURE 9-5. CP AND CTNET VS J, 2 BLADES, MN -- .5, _=3 °

113

m

SR-7L WIND TUNNELPERFORMANCE DATA

.32

.30

.28

.26

.24

.22

.20

.18

CT NET .16

,14

.12

,I0

.08

.06

.04

,02

/

L_

,10 ,20 .30 .40 .50 .60 .70 .80 .!)0 1.0

Cp

FIGURE 9-6. CTNET VS Cp, CONSTANT J, 4 BLADES, MN = .2.

114

SR-7L WIND TUNNELPERFORMANCE DATA

CTNET

.20

.18

.16

.14

.12

.10

.08

.06

.04

.02

J

/j///,//,'

___lr J = 3.53.0

.J = 2.5

/ IJ=2,0

/,//

• 10 .20 .30 .40 .50 .60 .70 .80 .90 1.0

Cp

FIGURE 9-7. CTNETVS Cp, CONSTANT J, 4 BLADES, MN " .5, _=0 O

115

SR-7L WIND TUNNELPERFORMANCE DATA

,16

.14

CTNET

.12

.1o •

.08

.06 ....

04.02 4.4

J=3. =4.0

.10 .20 .30 .40 .50 .60 .70 .80

Cp

FIGURE 9-8. CTNET VS Cp, CONSTANT J, 4 BLADES, MN = .8, ¢ =0

116

SR-7L WIND TUNNELPERFORMANCE DATA

.IO

.o9

.o8

.07

,o6

CTNET

.05

,04

.O3

.02

.01

J=2.4

/

J/"

/

SJ,,

/

.I .2 ,3 ,4

Cp

FIGURE 9-9. CTNET VS Cp, CONSTANT J, 2 BLADES, M N : .5, _ : 3 °

117

SR-7L WIND TUNNELPERFORMANCE DATA

,26--

.:'4-

.22-

,20-

.18-

.16--

14--

CTNET

.12-

°10_

.08r

.06-

.04.,

.02-

/

J= .91r

J=.8

mMN=.t9

mmm,,m, MN =.50

J=2,64

//

/

J

/7 /

//"1

J

J=2.31

//

/

//

/

.10 .20 ,30 .40

Cp

//

.5o .60

FIGURE 9-10. CTNET VS Cp, CONSTANT J, 8 BLADES, _ = 0°

118

9.3.2 (Continued)

The four blade and eight blade data was cross-plotted again to obtain curvesof net thrust coefficient versus advance ratio for a constant power

coefficient. Comparison of Figures 9-11 and 9-12 shows that for the same

advance ratio and power coefficient, a higher net thrust coefficient isobtained with the four blade configuration than with eight blades.

Therefore, as expected, the four blade configuration, at lower power

loadings, is more efficient than the eight blade Prop-Fan.

Comparlsons of the calculated and experimentally determ|ned performance of

the four and elght blade Prop-Fan configurations are shown In Figures 9-13and 9-14. These calculations were made using a refined lifting l|ne method.

The predlcted and measured performance agree very well for the four blade

conf_guratlon over the entire range of test poInts. Although the performance

of the eight blade Prop-Fan design was underpredicted at Mach 0.5, good

agreement between measurement and prediction was obtalned at Mach numbersof .70 and .73. Moreover, the trends of thrust w_th power are predicted

accurately.

119

SR-7L WIND TUNNELPERFORMANCE DATA

0.28

Cp = 0.10

0.24

0.20

0.16

CT NET

0.12

A 0 Cp = 0.20m

[I_ _Cp = 0.30

Cp = 0.40

ADVANCE RATIO, J

FIGURE 9-! I. CTNET VS J, CONSTANT Cp, 4 BLADES, _ = 0°

120

SR-7L WIND TUNNELPERFORMANCE DATA

0.28

CTNET

0.24

020

o16 _\_0.12

0.08

0.04

0

OCp = 0.2El Cp = 0.24

Cp = 0.29

0 1.0 2.0 3.0 4.0I ADVANCE RATIO, J

5.0

FIGURE 9-12. CTNET VS. J, CONSTANT Cp, 8 BLADES. _J = 0°

121

SR-7L WIND TUNNELPERFORMANCE DATA

CT NET

.12

.11

.IO

.09

.o8

.07

.06

.05

.o4

.O3

.o2

//

/

/ //

/ I

c l /i /

I/ 1

/ //

/ /

I

///

Z_

//

/

//

,//v

f

CALCULATED

PERFORMANCE:

M= 0.2, J= 1.0

M: 0.5, J= 2.0 "" " "" --

M---- 0.5, J= 3.0 _=,,m

TEST DATA:

(_ M= 0.2, J= 0.95

r"IM= 0.5, J= 2.0

/_M= 0.5, J= 3.!

.01

0 .....

0 .05 .10 .15 .20 .25 .30 .35 .40 .45 .50

Cp

FIGURE 9-13. COMPARISON OF MEASURED AND PREDICTED PERFORMANCE, 4 BLADES

122

SR-7L WIND TUNNELPERFORMANCE DATA

CTNET

c iiIII / ,/CALCULATED PERFORMANCE:

M =.5, J = 2.17 i f

M =.5, J = 2.64 t

.11 ' .n M =.70, J = 3.0_

.= 73,J:36 / if" // /

I I I / oif ,/,

.10 /

TEST DATA : ifr

O M =.5, J = 2. I7 ii

.09 ° M =.5, J=2.64 C I /" /A M =.7, J= 3.04 it

I_ M =.73, J= 3.59 / / f.oe / if / i

/ ,/ /

.o, I ,"x' _tI,il

it// I

// I.o5 it

/ /.o4 / , I

I /.03

/

.oz

.oi

.05 ,10 .15 .20 .25 .30 ,35 .40 .45 .50

Cp

FIGURE 9-14. COMPARISON OF MEASURED AND PREDICTED PERFORMANCE, 8 BLADES

123/124

lO.O BLADE SURFACE STEADY PRESSURE MEASUREMENT

I0.I Test Objective

To measure the steady pressure distribution on the surface of the SR-7L

Prop-Fan blade for a range of blade angles, rotational speeds, and simulated

flight Mach numbers.

10.2 Test Procedure

In preparation for blade surface steady pressure testing, the Prop-Fan was

installed on the drive system, as described In section 6.2, and supported as

shown In Figures lO-I and I0-2.

As noted earller in sectlon 2.0, collection of blade surface steady_pressEre_

test data was interrupted during the first tunnel entry In early 1986 and hadto be rescheduled for a second tunnel entry In early 1987. Though the steady

pressure test data collected was quite limited as a result of theinterruptlon, enough Informatlon was obtained to indicate that a revised

pressure tap layout was desirable. A new steady pressure blade incorporating

an improved pressure tap layout was fabricated for use during the second

tunnel entry. It provided a much higher density of surface pressure taps inareas on the blade where the local steady pressures were more sensitive to

changing operating conditions.

The steady pressure measurement blade (S/N 009) was Installed in blade

posltlon 7. For balancing purposes, a counterweight blade (SIN 058) wasInstalled in position number 3, as shown in Figure I0-3. As a precaution,

signals from 2 shank mounted straln gages were monitored throughout the

test. Special contour matching blade stubs were Installed in the remalning 6hub arm bores. The test rig power capabllitles necessitated conducting all

testing using a 2 bladed Prop-Fan conflguratlon, thereby permittlng operation

at power loadlngs per blade corresponding approximately to the take-off andcruise condltlons of the eight blade Prop-Fan design.

The Prop-Fan was operated In the beta control mode during the entire test.In thls mode, Hamilton Standard personnel were able to change the blade pitch

angle during testing by means of an Increase/decrease pitch switch located inthe control room. For a fixed Mach number and a constant power supplied by

the turbines, the Prop-Fan rotatlonal speed was varled by increasing or

decreasing blade pitch angle.

The blade pitch angle (6314) was measured and recorded for each test point.

An electrlcal slgnal proportional to blade angle was provided by means of a

potentlometer mounted on the rotatlng side of the Prop-Fan as Illustrated in

Figure 5-3.

•.,r,_ /.L-._'._--_--_;"_ _-'-----N_ m_

125

BALANCE

GEOMETRICAL

CENTER

CABLEB \ A "lzo R

c

"\11"_'_ " I"-I I I _h'--'-'--"--_

CABLES

(_: 14, 3L:3700

NOTE:

ROD ROD ROD

...... +r-- t'

L::3550 : :: 00

A B L:_,,,o C,

AL-DIME",,ON"INM,,.'I'ETERS(_ = DIAMETERL = LENGTH

126--e'_,:l _kOV3.t-S +-01, 3}

FIGURE 10-2. PROP-FAN STEADY PRESSURE TEST SET-UP

127

ORIGINAL PAGE

BLACK AND WHITE PHOTOGRApHr

#5

/--- COUNTERWEIGHT BLADE

FORWARD TO / SIN 058

AFT VIEW #3 /

I

8

_ STEADY PRESSURE BLADE

S/N OOS

# 7 ,_1 I(BLADE ANGLE MEASUREMENT

SYSTEM LOCATION)

FIGURE 10-3. LAP BLADE INSTALLATIONSTEADY PRESSURE TEST (2 BLADE}

128

10.2 (Continued)

The Prop-Fan rotatlonal speed was measured by use of a IP pickup. The sensor

was triggered by a gear mounted on the test rlg drive shaft. The rotational

speed was averaged over ten revolutions.

The power absorbed by the Prop-Fan was determined by multiplylng the torque

supplied to the Prop-Fan by the rotational speed. Torque supplied to the

Prop-Fan was computed by accounting for the measured frlctional losses in the

test rlg relative to the torque measured by the torquemeter.

Collection of blade surface steady pressure data was accomplished by

utilizing a specially Instrumented SR-7L blade as i11ustrated in Figures 10-4

and 10-5, coupled wlth a Scanlvalve TM pressure measurement system.

Thirteen rows of pressure taps were located on both the face and camber sides

of the steady pressure measurement blade. The pressure tap location andnumbering scheme used for data acquisition and reduction are deplcted In

Figures 10-6, 10-7 and I0-8. This numbering scheme differs from that In the

referenced test plan. The pressure tap channels which span the blade were

fabricated by bonding a thln plastic skin to a channellzed adhesive layer.Each channel was connected to a tube embedded in the root of the blade which

led out to the blade shank. The tubes were connected to the Scanivalve

mounted on the dome cap of the Prop-Fan, protruding through the nose of the

spinner. One Scanlvalve channel was provided for each tube. The stationary

portion of the Scanlvalve contalned a pressure transducer that could be

scanned by remote command to monitor one channel at a time. This arrangementallowed pressure measurements to be made at only one radial station per run.

Pressure measurements were made by masking off all the rows of pressure taps

except at the section of interest. The Scanlvalve was then cycled through

all channels to record the pressures at one radial station. Thirteen runs

were required at each Prop-Fan operating point to obtain a complete pressuremap for the blade surface at the operating point.

The Scanlvalve was enclosed in an aerodynamic falrlng to ma}ntafn a wellbehaved inflow to the Prop-Fan. The umbilical, which connected the

Scanlvalve to the control and monitoring equipment outside the tunnel, was

enclosed in a conduit with an airfoil shaped cross-section. This alsominimized disturbance of the flow.

Table 10-I lists the operating conditions that were run _r_m_} the test along

wlth tolerances showing the maximum variation In the parameters a11owed whentestlng at different radial stations. In general, the p_=ce<Bure for setting

a speclflc test condition was to set Mach number and them a{IL1ustthe rotor

speed and blade angle, to obtain the desired power coeff_¢_emt am<l advanceratio.

129

ORIGINAL PAGE

BL&CK AND WHITE PHOTOGRAPH

FIGURE 10-4. LAP STEADY PRESSURE BLADE {CAMBER SIDE}

130 " AG_

OF __UA_rIp_-_

FIGURE 10-5. LAP STEADY PRESSURE BLADE (FACE SIDE)

131

ORIGtNALBL.._CK .At,JD WHITE

PACE

Pl-i !.]h__ _,_ p _,

20'2O

20

20

20

Z0

Z0

ROW ! 3

)W IZ

I]

OW 10

R o_w p.._ _I

ROW 8

oPRESSURE PORT LOCATIONS

.NUMBERING SEQUENCE

ROW 7

ROW 6I

20

2O

ROW 4

ROW 3

20

ROW Z

CAMBER SIDE

FIGURE 10-6. STEADY PRESSURE BLAJ[_

132

• PRESSURE PORT LOCATIONS

eNUMBERING SEQUENCE ROW I 3

ROW 12,

ROW 1 I

ROW 1

ROW

ZI

p3636

36

ROW 7

ROW 6

ROW 5 36

ROW 4

ROW 3

36

ROW 2

ROW !

FACE SIDE

FIGURE 10-7. STEADY PRESSURE BLADE

133

l,J- !,20

I,J= I, I

CAMBER SIDE

(BLADE OUTLINE MIRRORIMAGE OF FACE SIDE)

56.0 t 142.24

52.0 132.08

48.0 "121.92

44.0 1 1.76

40.0 01.60

36.0

16.0 '40.64

'30.48

20.32

4.0 10.16

0,0' ).00

INCHES CENTIMETERS

FACE SIDE

I, J = !, 36

NOTE: THE VERTICAL SCALE REPRESENTS THE DISTANCEALONG THE BLADE PITCH CHANGE AXIS FROM

THE PROPFAN CENTER LINE.

FIGURE 10-8. ARRANGEMENT OF THE PRESSURE TAPS ON THE BLADE SURFACE

-- PROJECTED VIEW

134

TABLE 10-I. OPERATING CONDITIONS FOR BLADE STEADY PRESSURE TESTING

(2 BLADE LAP PROPELLER)

M J /3 CpCondition Mach Advance Ratio Blade Angle Power Coefficient

No. No. +J2 _+}00 ' __2

NRPM

1 .01 .08 13.80' .0792 .02 .14 15.70 o .0933 .02 .15 18.780 .1524 .03 .18 21.600 .2046 .20 .88 25.650 .0986 .20 .88 30.40 ° .2517 .50 3.065 57.510 .6498 .50 3.055 54.95' .3609 .50 3.063 50.860 .108

10" .60 3.066 54.980 .22611* .70 3.055 55.00 ° .22912" .78 3.07 54.97 o .22313" .78 3.20 54.98' .112

12001200

_ L20o.1200

16651651

1186119011851436168518401782

* RADIAL STATIONS 2,4 AND I0 WERE NOTRUN AT THIS CONDITION.

135

10.2 (Continued)

After establishing the operating condition, the basic wind tunnel and

Prop-Fan parameters (tunnel static temperature and pressure, Math number and

rotor speed) were logged into the microcomputer, a record number was assignedfor filing purposes, and the scanivalve was activated. The scanivalve then

ran through a calibration sequence followed by the pressure data scan. The

data were then plotted in preliminary form and reviewed. If the data

contained questionable features, a second scan was performed or hand scanning

of individual suspicious points was made.

Test points 1 through 4 in Table lO-I approximate static conditions of

increasing power and were selected to provide information on leading edge

vortex flow (there was no applied tunnel flow for these points, althoughthere was some Prop-Fan induced flow). Points 5 and 6 were seledte¢ to - -

Investigate take-off conditions: points 7 through g covered a wide range of

power loading conditions at 0.50 Mach number and bracket through the design

cruise power loading. Points lO through 13 were selected to Investigate

transonic flow characteristics and were all at relatively low power due torig limitations. For these points, power coefficient and advance ratio wereheld constant while Mach number was varied.

I0.3 Discussion and Results

IO.3.l Data Reduction

lO.3.l.l Data Format

The pressure data were reduced to coefficient form and plotted during the

test using a microcomputer system to provide the test personnel with a basis

for Judging the quality of the data following each scan. This preliminarydata reduction included an approximate correction (described below) for the

centrifugal pumping In the tubes and channels that led from the scanivalve to

the pressure taps. The pressure coefficient formula that was used is:

Cp I

Pc - Po

0.sp(v +

Nhere:

Pc - Corrected Blade Surface Pressure

Po - Tunnel Static Pressure

p , Air Density

Vo - Tunnel VelocityVtm Tangential Velocity at Pressure Tap Radius

136

........... 7

10.3.1.1. (Continued)

For each radial station the mid-chord radius (for a prescribed blade angle of

50.0 degrees) was used in determining the approxlmate centrifugal correction

and the tangential velocity. It should be noted, however, that the radii ofthe pressure taps at each station vary a|ong the blade chord, and furthermore,

the radius of any given tap varies with b]ade angle. In addition, the va]ues

describing the chordwise distribution of the pressure taps, which were stored

in the microcomputer, were nominal rather than measured values, and did not

represent the precise distributions. The final data reduction at HamiltonStandard will account rigorously for these effects, and will be presented in

a separate NASA Contractor Report (Reference 24).

10.3.1.2 Discussion of Problems

Before revlew|ng the data, it is necessary to address three basic test

problems, each of which had some effect on the data presented in this

report. The problems are as follows in decreasing order of severity:

l , Particle impacts resulting in failure of the tape seal over rows of

dormant pressure taps or resulting in venting of individual channels.

, Crack formation In the plastic skin which seals the channels on theblade surface.

3. Reference pressure transients which interrupted the scan sequence.

Partlcle impacts on the steady pressure blade occurred on several occasions

durlng the test. In most cases the impacts resulted In little or no damage,

however, four cases of damage to the tape which sealed the dormant rows of

pressure taps occurred and two cases of channel venting occurred due todirect partlc_e Implngement on a specific channel. The data affected by tape

seal fallure were llmlted to radial stations 3, 6, 11 and 13 at Math numbers

generally greater than 0.20. Channel venting due to direct particle Impact

resulted on channels 26 and 29 and was repaired followlng the runs in whichit occurred. The data that were compromised by Impact events will be

e11minated from the final data package.

Prior to the second test run (radial station 13), cracks were found in the

camber side plastic skin layer, between radlaI stations 6 and 8, and 9 and I0,

on the blade. The cracks, which are believed to have been catalyzed by theappllcatlon of a cleaning solvent that "shocked" and embrittIed the plastic,

were not present following the first run (radial station 5); they were found

Immediately after applying the solvent. Continued blade checks showed high

leak rates oo_channels 2 through 8. To correct the leak problem a series oftape strlpswas applied to the cracked areas, sealing them off and

establishing acceptable leak rates. Testing continued with this conflguration

and leak rates were monitored following every other run. The fix was found

_.,_to_e satisfactory on all channels, although channel 2 required further repair_" as the test progressed. In general this problem is not considered to have-, compromised the data.

_- DI::_3G_AI_ PAGE rg

137 DE POOR QUAI/T¥

10.3.1.2 (Continued)

Reference pressure transients occurred during several data scans.

Fortunately, this phenomenon was quite evident in the on-line pressurecoefficient p]ots, so scans that were affected were either rerun entirely or

hand scanned to pick up the affected points. Therefore this problem isconsidered to be of minimal consequence to the test data.

10.3.2 Discussion of Result

The entire spectrum of data Is summarized in Figures 10-9 through 10-21; each

figure shows all of the measured pressure distributions for a specific testcondition, and is in the form of the on-line data reduction. These figures

are presented in order of increasing Mach number. For Mach numbers where twoor more power settings were run, the figures are sub-ordered by Increasing _-

power. Radial stations 2, 4, and lO are left blank for Mach numbers greaterthan 0.50 because no data were collected due to limited test time. Radial

stations with data polnts which may be eliminated or replaced with hand

logged values, for the final data package, are marked with an asterisk.

Figures 10-9 through 10-12 show the pressure distributions for the nominalstatic operating condition. The pressure loading is seen to increase with

increasing applied power, as expected, and the presence of leading edgevortex flow Is suggested in Figures lO-ll and lO-12 by the negative pressure

hump that spans along the camber side leading edge; the vortex then appearsto sweep across the chord In between the 90 and 95 percent radius, resulting

in very high loading in that region.

The data at these static operatlng conditions contain some inconsistencies.

For example, when running radial station 6 the data was found to diverge from

the trends at the neighboring stations. The cause of this inconsistency isnot certain, however, It is noted that the actual Mach number was not zero

(due to the Prop-Fan induced flow) and was variable during the "static" runs(Mach No. varied from 0.02 to 0.04). Though the reason for the Mach number

variation Is not known for sure, it was revealed during a daily tunnel

inspection that a portion of the tunnel flow straightening honeycomb had beenblown out. Subsequent running with the Prop-Fan driving the tunnel resulted

in somewhat higher Mach numbers. To investigate this problem further radialstation 7 was rerun for conditions 2 and 4. Comparisons of the pressure

distributions or these runs are given in Figures I0-22 and 10-23. It is

clear from this comparison that a repeatability problem existed for some of

the static point data.

Figures 10-13 through 10-21 and Figure 10-24 show the pressure distributions

for the remaining operating conditions. The data at these conditions show

good repeatability. For example, Figure I0-24 shows radlal station 8 whichwas scanned twice at 0.20 Mach number, first on the climb up to 0.78 Mach

number (in accordance with the test plan sequence), and then on the decent,

approaching shutdown. The results are seen to be nearly Identical.

138

10.3.2 (Continued)

Some of the aerodynamic effects that are apparent in Figures lO-13 through

10-21 are leading edge vortex loading at 0.20 Mach number for the take-off

case (Figure I0-14), inverted leading edge pressure distributions for the low

power cases at high Mach numbers (Figures 10-15, I0-18, 10-19, I0-20 and

lO-21) and evidence of trailing edge shock waves at the outboard stations at

high Mach numbers (evident by the trailing edge pressure jump in

Figures I0-19, I0-20 and I0-21. The inverted leading edge pressuredistributions, as noted above, are typical for cambered airfoil sections

operating at Incidence below the design angle of attack value.

139

RADIAL STATION I

0.591

2, 0.474

RADIAL STATION 1 .

r/R z 0.341

1,000 0.2 0.4 0.6 0.8 I 0

BLADE CHORD, X/C

FIGURE 10-9, LAP SR-7L STEADY PRESSURE DISTRIBUTION (ON-LINE DATA)

¢

140

.... • i iNOTES: I. TOLERANCES SHOWN DENOTE MAXIMUM 0] !__|2

PARAMETER VARIATION FROM ONE RADIAL / /

S,TATION TO THE NEXT. -05. "_Z. -'PLOTS CONTAIN DATA POINTS WHICH MAY _ I . .

BE REPLACED OR ELIMINATED. 01 ! I r I ,J'

-0.5

:. •o-z o: o _ 9 X CAMBER SIDE"

,,i _-_-_ to rACES'DE i-1.0' 0 L I ,_'

-05;

-,o. o_

-0.1_

-t,O _ 13, 0.9911

5 12, 0.960

-2.0 -0]--Q t I , 0.975

0.964

-1.5 0.938

0.903-1.0 0

0.861

-0.5 0.609

5, 0.747

0.482

i2.0 0_2

-I .5 0.5_

-I .0

-:'.5 -0.5

L_ -Z,O 0

_" -1 .S

¢_ O.S

o.5 RADIAL STATION 1

1,00 0.2 0.4 0.4 0.0 1.0

BLADE CHORD. X/C

_$, 0.591

RADIAL STATION I.

r/R = 0.341

FIGURE 10-10. LAP SR-7L STEADY PRESSURE DISTRIBUTION (ON-LINE DATA)

141

0.59 1

RADIAL STATION I

02 04 04 0,° 1,0I_AD[ CHORD, XIC

RADIAL STATION I .

r/R -- 0.341

FIGURE 10-1 1. LAP SR-TL STEADY PRESSURE DISTRIBUTION (ON-LINE DATA)

142

NOTES:

13

FIGURE 10-12. LAP SR-7L STEADY PRESSURE DISTRIBUTION {ON-LINE DATA)

l_t3

NOTES:

TEST CONDITION NO. 5

M N _ O.ZO= ZS.05 ° + 1.00 °

RPM - 1665 + I 0

J =, 0.88 + 0.0Z

Cp : 0.098 + 0.0Z

TOLERANCES SHOWN DENOTE MAXIMUM

PARAMETER VARIATION FROM ONE RADIAL

STATION TO THE NEXT.

* PLOTS CONTAIN DATA POINT_J WHICH MAY

BE REPLACED OR ELIMINATED,

CAMBER._IDE ....

FACE SI DE

2, 0,474

RADIAL STATION I,

r/R _ 0.341

FIGURE 10-1 3. LAP SR-TL STEADY PRESSURE DISTRIBUTION (ON-LINE DATA)

144

0. 0.2 0.4 0.6 0.11 1.0

BLADE CHORD. X/C

FIGURE 10-14. LAP SR-7L STEADY PRESSURE DISTRIBUTION (ON-LINE DATA)

145

NOTES:

TEST CONDITION NO. 9

X HAMBER SIDE JO FACE SIDE

3, 0.59 I

2, 0.474

RADIAL STATION I,

r/R : 0.341

FIGURE 10-15. LAP SR-7L STEADY PRESSURE DISTRIBUTION (ON-LINE DATA)

146

NOTES:

T£Er CONDITION NO. 8

M H = 0.50

" 54.95" _- 1.00"

RPM "IIg0±10

J = 3.055 + 0.0Z

Cp = 0.360 +_0.0Z

I , TOL£RANCE;S SHOWN DENOTE MAXIMUM

PARAM£TIER VARIATION FROM ON£ RADIAL

STATION TO THE NEXT.

2. • PLOTS CONTAIN DATA POINTS WHICH MAY

BE R£PLACIED OR £LIMINAT£D.

CAMBER SIDE

-I ,0_

-0.$ •

0,$

-0.5_o _2

0._ _

-0.5_ | ,I,

o,i_ |

-o.

10 ixo.sJ O

ol .O

-0.5

0.5

0.

13*

(CAMBER SIDE iFACE SI'D£ " -- -

op==,==:_-.--.--._ 6 __-- I,. o.,_

-05 °'s_ _-'*_-'-----_ _-.._ Io,o.J,.4

,--._ _7, 0,SEt

-0, / /o__-'.. I I __-::--.. 4*/ J__

-05 0'5= 3

-0.s] o,__ _ =,0,474

05_ ' RADIAL STATION _,u -0.'=t " _ ___'_ r/R - 0.34I

_ os___j" STATION I_

• 0 02 0.4 0,8 0,8 1,0

BLADE CHORD, X/C

FIGURE 10-16. LAP SR-7L STEADY PRESSURE DISTRIBUTION (ON-LINE DATA)

147

TEST CONDITION NO. 7 "I'0 l

MN - o.3o " t r_'T- F-T-_- $?,51° + 1,00 ° 1 I I I I J

J - _I,0S_ + 0,0Z -1.01 t _ NV/ "

Cp - 0.649 + 0,02 |

NOTES: I, TOLERANCES SHOWN OI"NOTEMAXIMUM _0,5t _'_,._2"

PARAMETER VARIATION FROM ONE RADIAL ] J l l J J,

STATION TO THE NEXT. 01 _ - _ ;,

2, * PLOTS CONTAIN DATA POINTS WHICH MAY " }

ro:::::o.i

°2.0 .0.5

FIGURE 10-! 7. LAP SR-7L STEADY PRESSURE DISTRIBUTION (ON-LINE DATA)

148

TEST CONDITION NO. 10

M N ', 0.60j_ = S4,98 o +_. 1.00 °

RPM = 14:36 + 10

J = 3.066 + 0.02

Cp = 0.226 + 0.0Z

NOTES: I. TOLERANCES SHOWN DENOTE MAXIMUM

PARAMETER VARIATION PROM ONE RADIAL

STATION TO THE NEXT.

Z. " PLOTS CONTAIN DATA POINTS WHICH MAY

BE REPLACED OR ELIMINATED.

-o.5.oI_______ !*0.5

-o_ 11_

0.5

lO

FACE SIDE

0.082

-- 3.0.59 1

FIGURE 10-1 8.

RADIAL

STATION I

RADIAL STATION I,

r/R = 0.341

LAP SR-7L STEADY PRESSURE DISTRIBUTION (ON-LINE DATA)

149

NOTES: 1.

TEST CONDITION NO. I I

0 _

0.5

M N " 0.70

- 55.00 _ + 1.00 °

RPM - 1685+ IO 1| _

J = 3.055 +_ 0.02

Cp = 0.229 +_ 0,02 o

TOLERANCES SHOWN DENOTE MAXIMUM

PARAMETER VARIATION FROM ONE RADIAL

STATION TO THE NEXT. 10

2. * PLOTS CONTAIN DATA POINTS WHICH MAY

BE REPLACED OR ELIMINATED.

-0.5

o:I Lx

13"

CAMBER SIDE

FACE SiDE

0. S9 !

2.0.474

FIGURE 10-19. LAP SR-TL STEADY PRESSURE DISTRIBUTION {ON-LINE DATA)

150

NOTES:

TEST CONDITION NO, 13

M N = 0.78

]3 = 54.98" + 1.00

RPM : 1782 + 10

J = 3.20 + 0.02

Cp : 0.1 I Z +_ 0.02

I. TOLERANCES SHOWN DENOTE MAXIMUM

PARAMETER VARIATION FROM ONE RADIAL

STATION TO THE NEXT.

2. * PLOTS CONTAIN DATA POINTS WHICH MAY

BE REPLACED OR ELIMINATED.

-1.O-

wh: -0"5_O_Z

o

a. _ 0.5U

1.00 0.2 0.4 0.6 0.8 1,0

BLADE CHORD. X/C

3

0.5

lO

CAMBER SIDE D "

FACE SIDE

0,996

0.986

1 I, 0.975

0,964

0,938

0,903

0.841

6, 0.009

0.747

0.602

2, 0.474

RADIAL STATION I,

r/R -- 0.341

FIGURE 10-20. LAP SR-7L STEADY PRESSURE DISTRIBUTION (ON-LINE DATA)

151

TEST CONDITION NO. 12

N - 0.78- 54.07 ° + 1.00"

RPM : 1040 + 10

J - 3,07 _+ 0.0Z

Cp = 0.223 _+.0.02

TOLERANCES SHOWN DENOTE MAXIMUM

PARAMETER VARIATION FROM ONE RADIAL

STATION TO THE NEXT,

* PLOTS CONTAIN DATA POINTS WHICH MAY

BE REPLACED OR ELIMINATED,

4,]

oi] 13"

-0.5,

o_ I1"

-0.5

0.i 1-_

-0,5-

0.$"

10

9 •

CAMBER 51D£ jFACE SIDE

13, 0.986

12. 0.986

I !. 0.975

0.904

0.038

0.803

O. 801

0.809

0.747

0.082

0.591

2, 0.474

RADIAL STATION 1,

r/R = 0.341

FIGURE 10-21. LAP SR-TL STEADY PRESSURE DISTRIBUTION (ON-LINE DATA)

152

-2.5

-2.0

-1.5

z

-_ -l.o

_m "0. 5

0.0

0.5

1.0

LAP (SR-7) PRESSUREDISTRIBUTION

Radia] Sta.: 7 Record no.: 114

Beta (de9): 15.3 03-10-1987 2h48:01

Hach No. : _02

Speed (rpm): 1209

x comb_ side

io ?ace i_ide

-2.5

-2.0

-I,5

d_

wZ

_-I,0

°0.5

kd

0.0

0.5

Io0

LAP (SR-7) PRESSUREDISTRIBUTION

Radial St_ : 7 Record no. : 204

Beta (deg): 14.3 03-13-ig87 20:55.30

Moch No, = 0.02

Speed (rpm): IIQ3

x combei side

o-Fac_ :ide. _

........ 1 ....

0.0 0.2 0.4 0.6 o.e 1.0 0.0 0.2 0.4 0.6

BLADE CHORD. x/c BLAOE CHORD. x/c

o.e 1.0

FIGURE 10-22. REPEATABILITY PROBLEM FOR STATIC LOW POWER POINT

153

-2.5

-Z.O

-].5

&

-z.o

U

ua -0.5

0.0

0.5

hO

LAP (SR-7) PRESSURE OISTRIBUTION

Radial Sta.: 7 R_cord no.: [03

Beta (de9): 22.4 03-i0-xg87 18:26:06

Mach No. : 0.02

Speed(rpm): |203

i i dx comber sl e

Io Face bide

E

f

0.0 0.2 0.4 0.6 0.8 1.0

BLAOE CHORO. xlc

-2.$

-2.0

-1.5

zW

-hn

N(..I

_-O.g

LAP (SIR-7)PRESSURE OISTRIBUTION

Radial 5ta. : 7 Record no. : 205

Beta (de9): 21.4 03-13-igB7 21:04:30

NOch No. _ O.04

Speed (rpm): 120!

i

Ix combe- sideo ?ace ;ide

0.0

0.5

1.0

0.0 0.2 0.4 0.6 0.8

BI_AOIZC.IO_. xlc

1.0

FIGURE 10-23. REPEATABILITY PROBLEM FOR STATIC HIGH POWER POINT

154

-Z.5

-Z.0

LAP (SR-7) PRESSUREDISTRIBUTION

Rodiol 5to.: B Record no.: 172

Beto (de9): _6. e 03-[2m19_7 20: 2_: 50

Hoch No. : 0.20

Speed (rpm): 1670

x combe side

o Foce _ide

-65

-l.OtJ

-0.5

0`0

0.5

l°O

0`0

-& 5

-Z.0

-1.5

_-].0tJ

-0.5

0`0

0.5

1.0

LAP (SR-7)PRESSURE DISTRIBUTION

Rodio] 5to.: B Record no. : [B|

Beto (deg): 25.0 03-12-1887 21:5g: 4B

Moch No. : 0.20

Speed (rpm): 1701

x combe side

- != r_:e _Q-_

I

0.2 0,4 0,6 0`e hO 0`0 0,2 0.4 O.e 0`8

BLAOE CHORD. x/c 8LAOE CHORD, x/c

FIGURE 10-24. ILLUSTRATION OF REPEATABILITY AT 0.20 MACH NUMBER

155/z56

II.0 BLADE SURFACE UNSTEADY PRESSURE MEASUREMENT

11.1 Test Objectives

11.1.1 Measure the unsteady pressure distribution on the surface of the

SR-7L Prop-Fan blade for a range of blade angles, rotational speeds, and

simulated flight Mach numbers for axial and angular inflow conditions.

11.I.2 Evaluate the effect of a wake in the propeller inflow on the unsteady

pressure distributlon on the surface of a blade.

11.2 Test Procedure

In preparatlon for blade surface unsteady pressure testing, the Prop-Fan wasinstalled on the drive system as described in section 6.2, and supported as._

shown in Figures ll-l and ll-2. Table ll-I shows the conditions that weretested.

The unsteady pressure measurement blade (S/N 054) was Installed in blade

position 3. For balancing purposes, the steady pressure measurement blade

(SIN 009) was Installed In position number 7, as shown In Figure ll-3. As a

precaution, signals from 2 strain gages located on the shank of the unsteadypressure blade were monitored throughout the test. Specla] contour matchingblade stubs were installed in the remaining 6 hub arm bores. The test rlg

power capabilities and tunnel time constraints necessltated conducting all

testing using a 2 bladed Prop-Fan configuration, thereby permitting operation

at power Ioadlngs per blade corresponding approximately to the take-off andcruise conditions of the eight blade Prop-Fan design.

The Prop-Fan was operated in the beta control mode during the entire test.In this mode, Hamilton Standard personnel were able to change the blade pitch

angle during testing by means of an Increase/decrease pitch switch located inthe control room. For a fixed Mach number and a constant power supplied by

the turbines, the Prop-Fan rotational speed was varied by increasing or

decreasing blade pitch angle.

The blade pitch angle (B3/4) was measured and recorded for each test point.An electrical signal proportional to blade angle was provided by means of a

potentlometer mounted on the rotating side of the Prop-Fan as illustrated in

Figure 5-3.

The Prop-Fan rotational speed was measured by use of a IP pickup. The sensor

was triggered by a gear mounted on the test rig drive shaft. The rotational

speed was averaged over ten revolutions.

The power absorbed by the Prop-Fan was determined by muItlplying the torque

supplled to the Prop-Fan by the rotational speed. Torque supplied to the

Prop-Fan was computed by accounting for the measured frlctlonal losses in the

test rig relative to the torque measured by the torquemeter.

157PRECEDING PAGE BI.NNIKNOT FILib_.O

I-0Z

158

OF_ Y

OR,IGINAI. PAGE

BLACI'( AND ',,llIHITE P'r,KTrtY_APH

FIGURE1 !-2. PROP-FAN UNSTEADY PRESSURE _ SET-UP

159

TABLE IT-l.OPERATING CONDITIONS FOR BLADE UNSTEADY PRESSURE TEST

(2 BLADE LAP PROPELLER)

M J _ Cp NCondition Mach Advance Ratio Blade Angle Power Coefficient RPM

No. No. +.02 +1.00 ° +.02 ±10

2 .02 .14 15.700 .093 12003 .02 .15 18.78 o .152 12004 .03 .18 21.60 ° .204 12005 .20 .88 25.65 o .098 16655A .20 .88 27.190 .15 16845B .20 .88 29.50 ° .20 16846 .20 .88 30.40 ° .251 16517 .50 3.065 57.51 ° .649 11868 .50 3.055 54.950 .360 11909 .50 3.063 50.860 .108 1185

10 .60 3.066 54.980 ,226 143611 .70 3.055 55.00 ° .229 tB85

160

#5

F UNSTEADY PRESSURE BLADE

FORWARD TO //S/N 054

AFT VIEW #3 /

8

STEADY PRESSURE BLADE

SIN 009

_7_-" I{BLADE ANGLE MEASUREMENTSYSTEM LOCATION}

FIGURE 11-3. LAP BLADE INSTALLATIONUNSTEADY PRESSURE TEST (2 BLADE)

161

11.2 (Continued)

Mach number was established from the ratlo of static pressure, measured four

meters upstream of the Prop-Fan rotor, to stagnation pressure. Static

pressure was also measured in the plane of rotation as a backup. The ratio

of static to stagnation pressure was correlated wlth data taken during apre-test calibration in order to compute the Mach number. The Prandlt-Young

correction was applied to the computed Mach number to compensate for the

effects of the tunnel walls and the thrust produced by the Prop-Fan.

Collection of blade surface unsteady pressure data was accomplished byutilizing a specially instrumented SR-7L blade as illustrated in Figures II-4

and 11-5, coupled with an FM multiplex data acquisition system. Twenty-six

high frequency response pressure transducers were installed in two rows onthe face and camber sides of the unsteady pressure blade. The p÷esgur_ ---

transducer location and the numbering scheme used for data acquisition and

reductlon are depicted in Figures 11-6 and 11-7.

The pressure transducers were mounted flush with the blade surface as

depicted In Figure II-8. The signal and excitation wires from each

transducer were connected to slgnal conditioning electron|cs located In the

cuff of the blade. The slgnal wires also passed through attenuatingresistors mounted on the blade root. The function of the attenuating

resistors was to establish the gain for the pressure slgnals.

The unsteady pressure signals were transmitted from the rotating to thestationary field through the FM multiplex data acquisition system provided

for the SR-TL Prop-Fan. The signals were monitored on a four-channel

oscilloscope and recorded on a 14-track IRIG tape recorder.

The frequency response of the system was DC to 1000 Hz. Prior to the High

Speed Wind Tunnel Test an evaluation program was conducted to determine the

sensitivity of the transducers to temperature, strain, vibration, and

centrifugal 1oadlng. The results of this test program indicated a maximum 2_

of full scale error due to temperature in the range from 0 to 130°F and amaximum .92% of full scale error due to all other factors.

Generation of the wake In the Prop-Fan inflow was accompllshed by erecting a

vertlcal steel cylinder upstream of the rotor. The cyllnder was lOOmm (3.93inches) In diameter and was located such that its centerllne intersected the

Prop-Fan axis of rotation at a distance of 1.372m (54.02 inches) upstream of

the rotor plane. The wake generated by the cylinder was Intended to create a

twice per revolution (2P) disturbance for the Instrumented blade to pass

through. Figure ll-9 shows the Prop-Fan with the cylinder _n place.

162

• i

ii

ii

ii

!I

Ii

m

I

C20977

FIGURE 1 1-4. LAP UNSTEADY PRESSURE BLADE(CAMBER SIDE)

163

ORICINAL PAGE

8L,_C.K .A,'qD'_-_,ITEPI.K)T'b_GR,kYi,I

ORIGINALPAGEAND WHITE PHOTOGRAPH

~

k _

FIGURE 1 1-5. LAP UNSTEADY PRESSURE BLADE (FACE SIDE)

164

7,

FIGURE ! 1- 6 UNSTEADY PRESSURE BLADE, FACE SIDETRANSDUCER LOCATIONS

165

PT14C_ T18CPT16C

I

' I IPT15C PT19CPT17C

PT23C

PT21C I II

' I ' IPT24CPT20CPT22C PT26C

FIGURE 1 1- 7 UNSTEADY PRESSURE BLADE, CAMBER SIDETRANSDUCER LOCATIONS

166

\\

\\

\

PETG (GLYCOL MODIFIEDPOLYETHYLENE TEREPHTHALATE

SHEET} 0.030 THK-.j,

J_ L_._

SHELL PLIES

\

\

\

]t/

//

/

p TRANSDUCER/ (FLUSH MOUNTED}

,/J

J

FIGURE ! 1- 8 LAP SR-7L UNSTEADY PRESSURE BLADE TRANSDUCER MOUNTING

167

t.V" <_,--_"_,-V;..I.?_.TY

WAKE GENERATOR

FIGURE 1 1-9. UNSTEADY PRESSURE TEST SET-UP WITH WAKE GENERATOR

168

ORIGIN'A/ PA_['

6LACK AND _'HITE PHO'IOGRAPId

=

11.2 (Continued)

The complete range of Prop-Fan operating conditions that were tested is givenin Table ll-I. Because of time limitations, the range of conditions tested

was reduced for the zero degree inflow cases with and without the cylinder.

In addlt_on, transducer wire problems resulted in intermittent and

nonexistent signals on several sensors. The resulting series of testcondltions for which signals were recorded is given for each transducer in

Table ll-II.

11.3 Discussion and Results

To provide an illustration of the data collected during the test, examples

are given in Figure ll-lO of the periodic variations In pressure (withcorresponding frequency spectra) which were measured on the camber side ok -the blade at the 90% radius, 56% chord point (pressure transducer number

PTI6C). Here the three inflow cases are compared for the Prop-Fan operating

condltion defined below.

Mach number, MN - 0.20

Advance Ratio, 3 = 0.883

Power Coefficient, Cp = 0.250

Blade Angle, 8 = 32°

This operating point Is representative of the Prop-Fan take off condition.

The pressure versus time plots at the left in Figure II-I0 were obtained from

a sIgnal enhancing waveform analyzer. Sampling was Inltiated by the recordedonce per revolution plp signal and waveforms from I024 revolutions were

averaged. Thus the repetitive portion of the pressure waveform is enhanced

and the random part Is suppressed. The spectra, shown at the right In

Figure ll-lO, were obtained vla digital Fourier transform analysis.Successive time slices were transformed and averaged for 4.8 seconds. Each

spectrum conta|ns 400 frequency points spaced linearly from 0 to 500 Hertz.

Prel|minary Interpretation of Figure ll-lO is as follows:

In interpreting the waveforms, a transducer can be considered to be scanningthe inlet flow as it rotates. Since the blade position Is known as a function

of time, the time axls can be converted to angular position. The six and

twelve o'clock positions are indicated on the bottom trace. For the trace at

the top of Figure ll-lO, representing clean inflow, the signal level shouldbe low, corresponding to a low distortion level. However, a small slnusoldal

component can be seen In the waveform and spectrum that must be caused by aresidual flow angularity In the tunnel. This can be consldered a backgroundlevel and must be subtracted from the data for the 3° angular inflow and for

the cylinder wakes.

169

F-.

:+0

+,

+<

li:l--

hoI,-,,

<"I-(,,3

+i

li,i,.,.Illi<I--

<7

O__i u

...i Z

:+,-+

_zl-o+

I+,1,11:3.-12:;

I<_

I++!-

a _m-_ z<_ 0

WI-_ I"

hi

<">a

m" o <

l_-"tl._ . _,_ _ a

": _i

- _+_

170 +7<©C+p++++++r+,+,+-

N N N

_, .n. _ r_l I_ _-- r_ ,_':" / U

I-. >" I- u I-_ _" J wZ

__ till. _. ;_ a. J "_. _Ja............... _t_,r......................[ b.

'- zo _ I .--. _. [---.-.-.-.-.-.-.-.-_---ou...._ ................_. _ ................."'"._.--"_--_......!.............J

........ :-.'.::........ _ o .......... ................ o < i_...................._'-_.._ i ,.,Z- ,. - _ _°'"_

%. . o o _.0 ISd _l_l'_GS3_d _" "o _ -

0

o o i ¢

u u E P_ _

>" o >. ..... _I_ .z" _ Jou

° ........._ < uJ"_ ..s- _/_.

zo z0 o> zJo_ L. --_. _/__" _" < i'l' OI _ Olu

.i ...... ......... -I _r -- -

_ _ _ ® oo o, . o/ _. /,., ;

! ..... °o _ I_ ,'o _ _- _o_'_o o n U_ _ _ _; _ <_o

+ , + ' + . hi!-- NU) "Z I"'_ <

Wre

(Jl

n,

>.

<

I--

Z

C_

<

"I"!-

0

.J

<XLd

T

m

ISd "lVNIINON INOM.-I NOI.ILVlElVA 31::ll'tSS3_d eIOV'1fl

171

11.3 (Continued)

In the data for the 3° angular inflow, the angle of attack seen by theinstrumented blade should be nearly a pure sine wave at the once per

revolution (IP) frequency. Simplistic analys_s wou}d _ndlcate that the bladepressure response should also be sinusoidal. The waveform and spectrum showthat this is far from true. Figure 11-11 illustrates the terms "advancing"and "retreating" for angular inflow. Further evaluation of the data at otherpositions on the blade is required to identify the source of thisnon-slnusoidal behavior.

For the data with the cylinder wake, the blade pressure should respond with a

pulse each time the blade passes through a wake at the top and bottom of therevolution. This behavior Is observed in the bottom trace, but the pulse

magnitudes are surprlsingly different at the top and bottom positions_ ....

Another Interesting feature of the data for cylinder wakes is the oscillating

response after the wake pulse.

Slnusoldal response was observed on the pressure (face) side of the blade In

all cases examined for angular inflow conditions.

Sinusoldal response was also observed on the suction (camber) side of the

blade under low 1oadlng conditions. However, under high loading conditions,

non-slnusoldal behavior is present. The non-slnusoldal response appears to

be a result of leading edge and tip vortices which may be distorting theresponse. Another possibility Is the formation and breakdown of the vortices

as the angular Inflow or wake inflow modulates the angle of attack.

An analysis of the blade unsteady surface pressure data will be presented ina separate NASA Contractor Report (Reference 25).

172

ZRETREATING BLADE

INFLOW

i, IL

% _ w

ADVANCING BLADE

FIGURE 11-1f. ILLUSTRATION OF THE TERMS "ADVANCING" AND "RETREATING"

FOR ANGULAR INFLOW

173/174

12.0 CONCLUSIONS

The High Speed Mind Tunnel Test has provided an extensive evaluatlon of the

operating characteristics of the SR-7L Large-Scale Advanced Prop-Fan. Allthe test objectives, set forth in the Plan of Test 267X-135 Rev. D, regarding

acquisition of data were accomplished and I02 hours of operating experience

were attained. No problems were uncovered that would have been considered an

impedance to the planned follow-on PTA Flight Test. ONERA drive system

constraints and testing problems precluded running all desired test points,

however, those polnts successfully run did provide a wealth of aerodynamic

performance, structural dynamic, steady and unsteady blade pressureInformatlon. Further areas of investigatlon are indicated that should

ultimately result in highly accurate aerodynamic design and performance

prediction methods. The conclusions and recommendations derlved from each

phase of the High Speed Wind Tunnel Test are presented in the follow_ng -- -sections. Two separate low number NASA Contractor Reports will be published

providing a more detailed evaluation of the blade surface steady and unsteady

pressure tests.

12.1 Blade Structural Dynamic Evaluation

The SR-TL Prop-Fan was found to be free of high speed blade flutter over the

entire operatlng envelope tested. All the measured blade surface and bladeshank strains were below the a]lowables set prior to testing. These allowable

levels were set to avoid accumulation of fatigue damage to the blades,

therefore, no fatlgue damage to the blades was incurred.

Reasonable correlation was found between the measured and analytically

predicted IP blade bending strains for the SR-7L blade. Results confirmed

that IP strain peaks inboard on the blade and lessens near the tip. Also,

blade strains were found to Increase with power and Mach number (all othervariables held constant).

The Interpretation of IRP (infrequently repeating peak) strain values was

made difficult due to a significant amount of high frequency (>25P) noise in

the signaTs. In generaI, this type noise Is easfTy filtered out, however,this could not be done for the zero inflow angle data due to the low signal

response levels. The angular inflow (3°) strains were determined fromspectral analysis, the amplitudes of which were not affected by "noise"

outside the frequency range of interest.

Signiflcant 2P blade vlbratory response was also measured for the 3° angularInflow case. The amplitude of the 2P response averaged approximately 25% of

the IP component response. It is concluded that at least a portion of the

excitation force driving the 2P response was generated by the test rig drivesystem. Twice per revolution vibration is characteristic of a shaft with a

universal Joint. It Is also noted that nonlinearities in the aerodynamic

excitation, possibly due to observed vortex loading phenomena or tunnel

inflow irregularities, are causes of higher order excitation.

175

12.1 (Continued)

Because RPM traces were not available for the data reduction effort, the

harmonic content of the blade strains was determined from a "non-speed

corrected" spectra] analysis. The spectrum values were compared to valuesobtained from calculations of amplitude of the IP bandpass filtered signals.

It was concluded that the blade strain magnitudes from the spectral analysis

were withln I0% (low) of the actual magnitudes.

For comparison purposes, IP strain predictions were also made for the angularinflow cases. Due to test constraints, angular inflow test data collection

was limited to a 2 bladed Prop-Fan configuration at 3° inflow angle.

Predictions were 16 to 31% higher than test data for the inboard response(flatwlse shank moment and radial bending) and up to 13% lower than test for

the outboard bending response. Th|s is consldered reasonable Correl_tton,

especially when it is considered that the measured values were perhaps up to10% lower than correct. (See prevlous conclusion). As w|th earlier testingof the SR-7A 2 ft. diameter aeroelastlc model, correlation of root response

was marginal for the trailing edge, shear and chordwlse gages (Reference 26).

12.2 Aerodynamic Performance Evaluation

Measured aerodynamic performance of the SR-7L Prop-Fan corresponded well with

analytical predictions for the four blade configuration over the entire range

of points tested. Similarly, good agreement between measured and predictedperformance was found for the eight blade configuration at Mach numbers of .70and .73. However, performance of the elght blade configuration was slightly

underpredlcted at .50 Mach number. The characteristic shape of the LAP

performance curves were slmiiar to those observed for the SR-7A aeroelasticmode] wlnd tunnel tests (Reference 27).

12.3 Blade Surface Steady Pressure Measurement

Steady pressure distributions were successfully measured on the blade surfaceat a11 of the 13 radial stations for Mach numbers of 0.03, 0.20 and 0.50, and

at all radial stations except 2, 4 and lO for Mach numbers of 0.60, 0.70 and

0.78. Subsequently, an uncertainty analysis was performed, which demonstratedthat the measurement errors and uncertainties involved In th|s test, were

acceptable.

Studles of the sensitivity of the correction for centrifugal 1oadlng on the

column of air In the blade's pressure tap channels showed that the

assumptions used In processing the data were valld.

During operation at approximate statlc rotor conditions, the Inflow Machnumber was 0.03 +0.015, where +0.015 Is the maximum statlon-to-station

variation In the-Math number. -The pressure distributions were found to be

very sensitlve to Mach number for these conditions, so some respectivestation-to-station pressure dlstributlon inconsistencies exlst In these data.

176

12.3 (Contlnued)

The chordwlse pressure loading Is found to moveaft with increaslng relatlve

Mach number for the 0.60, 0.70 and 0.78 Mach number cases.

Evidence of tlp edge and leadlng edge vortex flows were found in the statlcrotor data and the 0.20 Mach number data.

The inverted leading edge pressure distributions observed during the low

power high Math number conditions, are typical for cambered airfoil sections

operating at Incidence below the design angle of attack value.

Evidence of trailing edge shock waves Is present at the outboard radialstations In the 0.70 and 0.78 Mach number data.

12.4 Blade Surface Unsteady Pressure Measurement

Unsteady blade surface pressure data were successfully measured over the

followlng range of conditions"

Angular Inflow (3°) 0.02 ! MN _ 0.70Unlform Inflow (0°) 0.02 < MN < 0.50

Inflow with Wake 0.03 < MN < 0.50

The uniform Inflow data shows evidence of distortlon. This appears to be a

result of test section Inlet asymmetry.

The angular Inflow and wake data clearly shows unsteady pressure response.

dominant once-per-revolutlon response is evident in the angular Inflow datawhile the wake data i11ustrates twlce-per-revolution response as the

Instrumented blade passes the wake generating post.

A

Slnusoldal response was observed on the pressure (face) side of the blade in

all cases examined for angular Inflow conditions.

S1nusoldal response was observed on the suction (camber) side of the bladeunder low loading conditions. However, under high loading conditions,

non-slnusoldal behavior Is present. The non-slnusoldal response appears tobe a result of leadlng edge and tip vortices which may be distorting the

response. Another posslbillty Is the formation and breakdown of the vortlces

as the angular Inflow or wake Inflow modulates the angle of attack.

177/178

LIST OF SYMBOLS

_mm,-=__# r']_ ..... ,,

A

AN

AM

AF

&

BHP

BF

CCBDT

CCBDW

CSD

Cp

CT

CTNET

C

D

D

EF

LIST OF SYMBOLS

area

Incremental forward centerbody area at tap N

Incremental aft centerbody area at tap M

blade activity factor = 6250 f"°(b/D) X3dX

Hub/tip

area weighting factor

brake horsepower, KW

buoyancy force, N

centerbody drag coefficient without blades =

centerbody drag coefficlent with blades

sp|nner drag coefficient =

Ds

qo (As)

power coeffic|ent =poN3D s

thrust coeffIcient -

poN2D"

net thrust coefficlent =

TApp - BF

poNZD 4

speed of sound, m/sec

drag, N

dlameter, m

excltatlon factor = 9(VT/644.8)Z(p/po)

DCBT

qo (Ac,)

DcB

qo (AcB)

179

F

IRP

M

N

P

P

PA

Po

PN, PM

PT

PSTAG

Q

qo

r

r/R

SHP

T

TSTAG

Ts

T_

V

q

force, N

Infrequently repeating peak

V

advance ratio = 60 --

ND

Mach number

rotational speed, RPM

power, watt

pressure, N/cm 2

pressure forces in the form (P-Po) Area, N

freestream statlc pressure, N/cm 2

static pressure at tap N, M

total pressure, N/cm z

stagnation pressure, N/cm 2

torque, N'm

dynamic pressure, N/cm z

radius, m

fractional radlus

shaft horsepower

thrust, N

stagnation temperature, °K

static temperature, °K

total temperature, °K

velocity, mlsec

blade angle, deg

effIclency

180

w

Subscrlpts

APP

B

BP

CB

CBT

C, COR -

FL

MEAS

N

NET

0

S

t

T

TH

TNET

mass density, Kg/m 3

inflow angle, deg

thrust

p(d_sc area)V 2

dlsc area

tunnel cross-section area

apparent

balance

back pressure

centerbody

centerbody tare

corrected

funct|onal losses

measured

number

net

free stream

spinner

tangential

true

thermal effects

net thrust

lsl/ls2

REFERENCES

mmmJ,_m ....fl_'_J

REFERENCES

I. J.B. Whitlow and G.K. Sievers, "Fuel Savings Potential of the NASA

Advanced Turboprop Program," NASA TM-83736, 1984.

2. J.F. Dugan, Jr., NASA LeRC, B.S. Gatzen and N.M. Adamson, UnitedTechnologies, "Prop-Fan Propulsion - Its Status and Potential", SAE Paper780995, November 1978.

3. "Cost/Benefit Tradeoffs for Reducing the Energy Consumption of the

Commercial Air Transportation System, Douglas Aircraft Co; NASA AmesContract NAS2-8618, NASA CR-137925 Summary Report, June 1976.

4. "Study of the Cost/Benefit Tradeoffs for Reducing the Energy Consumption

of the Commercial Air Transportation System", Lockheed-CalifOrni_ - - -

Company, NASA Ames Contract NAS2-8612, NASA CR-137927 Summary Report,

August 1976.

5. "Energy Consumption Characteristics of Transports Using the Prop-Fan

Concept", Boeing Commercial Airplane Company, NASA Ames ContractNAS2-9104, NASA CR-137938 Summary Report, October 1976,

6. "Study of Unconvent|onal Aircraft Engines Designed for Low Energy

Consumption", Pratt-Whitney Aircraft, NASA Lewis Contract NAS3-19465,

NASA CR-135065 Final Report, June 1976.

7. "Study of Cost/Benefit Tradeoffs for Reducing the Energy Consumption of

the Commercial Air Transportation System", United Airlines, NASA AmesContract NAS2-8625, NASA CR-137891, June 1976.

8. "Study of Unconventional Aircraft Engines Designed for Low Energy

Consumptlon", General Electric Co., NASA CR-135136, December 1976.

9. "Fuel Conservation Merits of Advanced Turboprop Transport Aircraft",

Lockheed-Californ|a, NAS2-8612, NASA CR-152096 August 1977.

lO. NASA Tech. Mem. 83736 "Fuel Savlng Potential of NASA Advanced TurboProp

Program", September 1984.

11. D.C. Mlkkelson, G.A. Mitchell, and L.J. Bober, "Summary of Recent NASA

Propeller Research", NASA TM-83733, 1984.

12. NASA CR 3505 "Evaluation of Wind Tunnel Performance Testing of an

Advanced 45° Swept Eight Bladed Propeller at Mach numbers from .45 to

.85", Hamilton Standard Dlvlslon, Windsor Locks, CT, March 1982.

183

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

C.L. DeGeorge, J.E. Turnberg, H.S. Wainauski, "Large-Scale AdvancedProp-Fan (LAP) Statlc Rotor Test Report", Hamilton Standard DivisionUTC, Windsor Locks, CT (NASA CR-180848).

W.E. Su111van, J.E. Turnberg, J.A. Vlolette, "Large-Scale Advanced

Prop-Fan (LAP) Blade Design," Hamilton Standard Division, _indsor Locks,CT, (NASA CR-174790).

R.A. Schwartz, P. Carvalho, M.O. Cutler, "Large-Scale Advanced Prop-Fan(LAP) Pitch Change Actuator and Control Design Report", HamiltonStandard Division, Windsor Locks, CT, January 1986 (NASA CR-174788).

M. Soule, "Large-Scale Advanced Prop-Fan (LAP) Hub/Blade Retention

Design Report," Hamilton Standard Division, Windsor Locks, CT _ _(NASA CR-17¢786).

B. Huth, "System Design and Integration of the Large-Scale AdvancedProp-Fan, "Hamilton Standard Division, Windsor Locks, CT, August 1984,(NASA CR-174789).

Pierre, M. "Characterlstlques et Possibillte de la Grand Soufflerie

Sonlque de Modane-Avrleux," ONERA Technical Note 134, 1968.

Masson, A. "Essair D'Hellces dans ia Grande Soufflerle de

Modane-Avrieux," ONERA Technical Note 161, 1970.

Pope, A., "Wind-Tunnel Testing," John Wiley and Sons, 1954.

R.M. Reynolds, R.I. Sammonds and G.C. Kenyon, "An Investigation of aFour Blade Single-Rotatlon Propeller in Combination with an NACA

2-Serles, D-Type Cowling at Mach numbers up to 0.83", NASA RM A53B06,

April 13, 1953.

H.G1auert, "Airplane Propellers, Body and Wing Interference", Volume IV,

Division 2, Chapter VIII, Aerodynamic Theory, W.F. Durand, editor Juluis

Springer (Berlin), 1935 (Dover reprint 1963).

"BESTRAN User's Guide, Auxiliary Program ST570", D.C. 3ennings,

December 14, 1977.

Bushnell, P.R., "Measurement of the Steady Surface Pressure Distribution

on a Single Rotation Large Scale Advanced Prop-Fan Blade at Mach numbersfrom 0.03 to 0.78", (NASA CR-182124).

184

25.

26.

27.

Bushnell, P.R., Gruber, M.E., and Parzych, D.J., "Measurement of the

Unsteady Surface Pressure Distribution on a S|ngle Rotation LargeScale Advanced Prop-Fan Blade at Mach numbers from 0.03 to 0.70",(NASA CR-I82123).

D. Nagle, S. Auyeung, 3. Turnberg, "SR-7A AeroelastIc Model Design

Report", Hamilton Standard Divls|on UTC, Windsor Locks, CT(NASA CR-174791).

G. Stefko, G. Rose, G. Podboy, "Wind Tunnel Performance Results of anAeroelastlcally Scaled 2/9 Model of the PTA Flight Test Prop-Fan",AIAA-87-1893.

lss/ls6

,i

APPENDIX A

CHRONOLOGICAL HISTORY OF TEST

APPENDIX A

CHRONOLOGICAL HISTORY OF TEST

Testing of the LAP Prop-Fan started on February 20, 1986 and continued until

April 9, 1986. The test was terminated at this point following the discoveryof severe fretting corrosion in the ONERA drive shaft retent|on area. During

this time period, 55 hours and 3 minutes of test tlme were accumulated.

Testing resumed on February 27, 1987 and contlnued until March 19, 1987. Theintent of the second phase of testing was to collect blade steady and

unsteady surface pressure data utlIizlng a 2 blade configuration. Testing

was successfully completed after accumulating an additional 47 hours and lOmlnutes of test time.

The following tabulation provides a chronological history of the entire HighSpeed Nind Tunnel Test conducted at the ONERA facil|ty In Modane, France.

I--

I-- ,-

I t I I f I I I

I I I I I I I

I--

i--I

Oi

(J

I

"0

"0

4_

l-¢J

i

I/1

I-

¢1 •"0

U- n_I .,-

,"14=,0",-

¢l.-II-

4-1

0¢J

¢I

O_

0

o,-

0

klJZ0

"0

Ill

t_

4_

I

_0-0

_,- O.

_ _ e'_

_ 0_ e'-

e'_ e--

_ 0 "1

e_

_ e-- Z

e-

CO CO O0 O0 CO

& ,_" & ,_ Jex; o4 eq e_ eq

m3 r-j _'3 _'3

q_ kooo oo

e_ ed

__ L

.,-.

4-)

{.)

°p.-

°.I i I I I

--I

_=.o

v

I--"

l.l..

Oi

>-

2

G#

_:Ii."-

I

i'm

8

I I

° , °

•-I _13 aJ llJ0 r- 4-., _

0 "_- r- _-•"o _ ,- 0 0

•,-- LJ L.I> Q.

0 m

i-- _.) r- e- e-

I I

WZ0

1

0.-_

0

e- ,,.-

I I

"0

0

4.o

e-

_ Z_ o 4.-;

_'-

.,- I _:I

o_ _- • 0

4.o _

•i-J .._ I-. _,1 c

---

0"o _ U_ 0

-1-,, _,_ ._ 0

I I I

0

tel

e-

I_ 0 It I

0 _ 4-_u 0 _-

0 0 _'

•_ ,--- -I-_ e"l, _

_.. ,.-- ,,,I--I _ •

_'m e" _I_ _-- II

l..- _ _ .-m z

I

S-¢.-

o¢-,j

"I_ Ud_4.-) z__ ._--

0_.n 4...)

.I,-._ "I_

c13 (I_

e.-e- ,I-_

(M ('_ 0"1

e- e- e-

,4

e_

_0

I-

i,

3

o1--

L/1

Eu

{.-.p.-=E

v

,c/" _I- O'_

.:: ;4

¢M L#IO

00 C_

.oLr).o

p_

(13 r-

4.m

O(Jv

I--

..l-

_J

i--i

>

I I I

in

"0rt_

_[j ,--

"O Ort_

P- a-_

O

U

O _-

_" "00 C

rtS

O

(fl P-

.fa

t---

Z

IJ

_-O

I r-

::3

0-_

.E'I_

U

>_-- I

_,1 ..E

u': ,--- }l

0 0

_1. (D

•,- I0 0

_.. $-(]j .,.-- _.

._- I

_" "_ "t_,., O g:c

e- > O

_0 _

,J,-: c c i_

0

_0_

.4_

_S-O aO

I _IJ II

C: i: _ z

:_ .._ -_ _- 0

_. f,.. {=

o =

"_ _-- _1_0

_- ..-- _y .4-- • _-

_" _" J:l 0'1 E3 :_ 3 • z O

o.

(-_O0 O

• GI -1-) __-. _L)

II O :_ (-"

z ._,- :3

O O _--'_

P_" _" -I-_ ._

I U'IO(1; -

,-.- E: _=,--

_ O_:_ 4-)

01.--" C__- 0 _._ ,_

-T-

O"_XD

"t_ u

NE

E'Q

oP :3rl_J_C_.(_

%._E

._,(f) L_J

.Z:Z:O

CO 00 00

St._ I- i- $,., _.. I.- _- $.... $-

1.4_ _ _ 1.4. 1.4..

0{3

::3I-

0(3

L_.4=

(-,,J

"--.,I=U _=_

A

E

v

0 0 eJ 0°° °° _ .°CO CO COF-- r-- F-- _'_ ¢vI ¢vI ¢v_

Loo'I

I °" I "°

o

_ _ _o>otlt2 4.._ ,_- 0 0-

:_ ' % _ o = ®_-u_ "oZ

-i=_v

_ _: g _= •

m_- c_ o

_i Ji ""

t_

XD

CO

0q--

t_

-0

.0

c-

tnc-

I I

c-O

tn

°_

c-O°_

c- _j

c-

c-,-- 0

tO

e-t2

X

e- e-- e-- e--

CO O0Cn 0"_ 0"_

e- e- e--

I.- $.. l--

cO

e- e-

X

u

.r-_- u_

,o t,

v

0

_0oOO_

r-_

c_<

6 _i

0,-.I'-- •

e-4-1 ._

_.Jv

c_0.o

1.0ul

o'10oo

LO

CO0ookO1.0

" a_ '--

"_ "Z "Z 'Z0,. 0.. 0.. 0..

< < < <

0 0 0oo .o .o

UO _

t I I

,-- 0.0E • -_

_ 0

= _ o-_

0 _

.:I .,-

r- 0 >0 _ _1_

O0 _0 O0

< ,,_ <

e-.v,.-

I--

=D

I.n

O-

r_

,e-

Q_

0

O0

e,,.'I--O'

c'-

G)E=1

r_

4-1,

l,--v

I--0

co o'D o"10 0 0

L,O _ 1.0I_.'1

! I I

t#1

0

>

'" "_ 0

0 _ "_

I _-

_ _ 0

0 _- c_

o..I ,c..I

r--j _ i--j,

PI---

.r--

0.. o,

v

0 0 0,, ._ ,,

t_

0.out_Lt_

0.o

o_

0,o

0o,

I.n

0 0,o ..

.<","10

=3t--

e..

8v

I'-

Oi

>,

4_

#J4._

c-O

ClJ4.JU

"aO

0

Z e-

04-'

I i i

0 _n _L_

.4-1.1-_ r_

t:e"::3 _ • 0

4-_'_ _ _.

e- _. e-

l 4_•4.;, tn

e- _ o

,.-._ 0

e- 0 _ >

0"_ "_

oe- I_ e-e-

4_

_0

1-- ®

0 0

o_=. ,_-.

r- e-e" e"

0H

_.;_ z

I4_

0

I I I

0 >

0 _ O_

U _ I_1_

e- I

0 "_4_ I 4-} In

,'_ O_-

_l_ oJ • "1_ 4..z"l_ ..i..ze.. _l_ _j_ 0 _.-_ e.-

= o o

0

."s0

0

e- _..

e-l (J

"_ I

0"_

I _

i

t_

Q_e--

0

oo

I1

z

0

r_

4_e-- o'_r_ cL_

r,-,

' c4N

l- I,- $.-

CI_ QJ CI_1.1_ I._ I,_

f,.. I,.. L

$--

r_a0

f,-

Le'J

I,

aO

f,--

I.i--

p_aO

%.

C_I,----

C;

$..

h

u

0,j °,

c-OL_v

i.-i

8

o_ ">'5

r_

0,

I

_=0

=o_ 4-)

_ _=o 0

-.s >

-_,.--

u'_ e" GI 4-_

"-g8 =r._ 0_0 _- _.

o'_ o') o) 0cO _ U =) 0,, ,, ,. °.

I -- r- (_>

> _ _..,-,__ _ c"_,-- _" C_ t_> _ _-_-_ 0 "0

_o _-- ,_

,-- _. _ _:I _- ._

c 0.._ _ _cO

-_,_ O0 _r_

c O0 ,_0 u>_u

10

_q>- ,.-

0

__E '-_v

,o

0

oo co

r,j _.j

f_

¢_j t.-

I I

"1o I--

l-.--

c0

U

I,-- e-

I'- I

0

Q_ •4-P

¢'_ Q.

_.j _

_ 0°f,,- !I_

e--

e-

_ E

1, ReDort No. 2. Government AcceSsion No. 3. Recipient's Catalog No.

NASA CR 1821255. Report Date

JULY 1988

4. Title and Subtitle

LARGE-SCALE ADVANCEDPROP-FAN (LAP)IIIGHSPEED WIND TUNNEL TEST 6. Performing Organization Code

73030

7. Author(s) 8.

William A. Campbell Peter ArseneauxHarold S. l_ainauski

9. P_f_ming Orgini_tion Name end AddressHAMILITON STANDARD DIVISION

UNITED TECHNOLOGIES CORPORATIONONE HAMILTON ROADWINDSOR LOCKS, CT 06096

Sponsoring Agency Name and Addrm

NASA Lewis Research Center21000 Brookpark Rd.Cleveland, Ohio 44135

Performing Or_nizauon Report No.

HSER 11894

12.

10. Work Unit No.

._35-03-01

11. Contract or Grant NO.

NAS3-23051

13. Type of Report and Period Covered

CONTRACTORREPORT

14. Sponsoring Agency Code

15. SuD_emen_ Note=

Project Manager: J. Notardonato, Advanced Turboprop Project OfficeNASA Lewis Research CenterCleveland, OH 44135

16. Abs_a_

High Speed Wind Tunnel testing of the SR-7L Large Scale Advanced Prop-Fan (LAP) is hereinreported. The LAP is a 2.74 meter (9.0 FT) diameter, 8-bladed tractor type rated for 4475 KW(6,000 SHP) at 1698 RPM. It was designed and built by Hamilton Standard under contract tothe NASA Lewis Research Center. The LAP emp_ys thin swept blades to provide efficient propulsiorat flight speeds up to Mach .85.

Testing was conducted in the ONERA SI_ Atmospheric Wind Tunnel in Modane, France. The testobjectives were to confirm the LAP is free from high speed classical flutter, determine thestructural and aerodynamic response to angular inflow, measure blade surface pressures (staticand dynamic) and evaluate the aerodynamic performance at various blade angles, rotational speedsand Mach numbers.

The measured structural and aerodynamic performance of the LAP correlated well with analyticalpredictions thereby providing confidence in the computer prediction codes used for design.There were no signs of classical flutter throughout all phases of the test up to and includingthe 0.84 maximum Mach number achieved. Steady and unsteady blade surface pressures weresuccessfully measured for a wide range of Mach numbers, inflow angles, rotational speeds andblade angles. No barriers were discovered that would prevent proceeding with the PTA (Prop-FanTest Assessment) Flight Test Program scheduled for early 1987.

17. Key Woads(Suggested by Author(s))

Advanced TurbopropProp-FanEnergy Efficient PropellerHigh Speed Wind Tunnel

T_is docu_ wri11 remain under distributi(

limitation uat_I July 1989.

lg. Security Oauif. (of this report) 20. Security Qa_if, (of this I_l t Z_t. No. of Pages 22. _ice"

UNCLASSIFIED UNCLASSIFIED [ 193

_

NASA-C-16@ (Ray. 10-75)