152
NASA CONTRACTOR REPORT 177436 Hover Performance Tests of Baseline Metal and Advanced Technology Blade (ATB) Rotor Systems for the XV- 15 Tilt Rotor Aircraft K. Bartie, H. Alexander, M. McVeigh, S. La Mon, and H. Bishop Prepared by B_EIN_ VESTAL _AHIPANY Philadelphia, PA for Ames Research Center under contract NAS2-11250 NASA National Aeronautics and Space Administration Ames Research Center Moffett Field, California 94035 y ]] https://ntrs.nasa.gov/search.jsp?R=19880016983 2018-06-29T12:26:49+00:00Z

NASA · LIST OF FIGURES LIST OF TABLES LIST OF SYMBOLS ... Check Calibration Results: ... Variation of Cp with CT3/2 for Baseline ATB with No Cuff

Embed Size (px)

Citation preview

NASA CONTRACTOR REPORT 177436

Hover Performance Tests of Baseline Metal and

Advanced Technology Blade (ATB) Rotor Systemsfor the XV- 15 Tilt Rotor Aircraft

K. Bartie, H. Alexander, M. McVeigh, S. La Mon, and H. Bishop

Prepared by

B_EIN_ VESTAL _AHIPANY

Philadelphia, PAfor

Ames Research Center

under contractNAS2-11250

NASANational Aeronautics andSpace Administration

Ames Research CenterMoffett Field, California 94035

y

]]

https://ntrs.nasa.gov/search.jsp?R=19880016983 2018-06-29T12:26:49+00:00Z

N

CR177436

FOREWORD

The research and development activity reported in this document

was performed by the Boeing Vertol Company for the National Aero-

nautics and Space Administration, Ames Research Center, underContract NAS2-11250. Mr. M.D. Maisel (Aeroflightdynamics, AVSCOM,

Ames Research Center) was the NASA program manager, and Mr. D.J.

Giulianetti (NASA) was deputy program manager. Other Ames Research

Center personnel who made a significant contribution to this phase

of the program included Messrs. M.K. Betzina, F.F. Felker Ill, L.A.Young, and D.B. Signor. Mr. H.R. Alexander was the Boeing Vertol

Company program manager.

The following Boeing Vertol personnel made significant contributions

to this test program:

D. Ekquist }

K. Farrance

W. Grauer

I. Walton

F. Devlin }N. Ross

K. Bartie }

R. Benson

C. Coleman

M. McVeigh

Wind Tunnel Test Engineering

Instrumentation and Control System Design

Aerodynamics

H. Bishop 1

M. Cawthorne

S. La MonH. Silcox

Stress

S. Botwinik 1

D. Reed

R. Smith

R. Ricks

Dynamics

J. Mayer }W. McLean

D. Podgurski

Wind Tunnel Design and Project Management

Mr. Hank Silcox and Mr. Joseph Mayer merit special mention for their

design of the balance system used to measure rotor performance. Thiswas a major advance in balance technology and the accuracy and relia-

bility of the measured performance data was an important component in

the success of the program.

H.R. Alexander

pRF_.,_ING PAGE BLANK NOT FILM]_I)

iii

CR177436

ABSTRACT

Rotor hover performance data were obtained for the full scale Advanced Tech-

nology Blade (ATB) designed for the XV-15. The ATB rotor thrust-weightedsolidity is 0.10. The test was conducted as part of contract NAS2-11250 atthe NASA-Ames Outdoor Aeronautical Research Facility (OARF). The XV-15

basic rotor (solidity = .089) was also tested. Variations of the ATB tipplanform and cuff planform were also tested. A peak figure of merit of0.806 was demonstrated for the ATB and a value of 0.791 for the XV-15 steelblades. Measurements of the downwash in the wake at O.4R below the disc are

also presented.

KEY WORDSim

Rotor

Hover PerformanceTilt Rotor

XV-15

Rotor BladeRotor Loads

Static Performance Test

iv

CR177436

TABLE OF CONTENTS

LIST OF FIGURES

LIST OF TABLES

LIST OF SYMBOLS

1.0 SUMMARY

2.0 INTRODUCTION

3.0 DESCRIPTION OF TEST INSTALLATION

3.1 Test Stand3.2 Motor and Drive System

3.3 Balance3.5 Hub and Controls

3.6 Rotors3.6.1 Advanced Technology Blade Rotor

3.6.2 XV-15 Rotor

4.0 INSTRUMENTATION

4.1 Instrumentation - General

4.2 Rotor Balance Instrumentation4.3 Blade and Hub Instrumentation

4.4 Wake Rake4.5 Anemometer4.6 Acoustical Measurements

5.0 DATA ACQUISITION AND REDUCTION

5.1 Data Acquisition

5.2 Data Processing

6.0 TEST RECORD AND DATA ACCURACY

7.0 ROTOR PERFORMANCE

7.1 XV-15 Metal Blade Performance

7.2 Baseline ATB Performance7.3 Performance of ATB with Extended Cuff

7.4 Performance of ATB with No Cuff7.5 Performance of ATB with Swept Tip and Extended Cuff

7.6 Performance of ATB with Square Tip and Extended Cuff

7.7 Configuration Performance Comparisons

7.8 Theory - Test Comparison

vii

xiii

xiii

6

99

13

13

1322

22

22

22

2227

27

27

27

27

29

30

50

5O

62

6262

76

7676

96

V

CR177436

TABLE OF CONTENTS (Continued)

7.8.1 Rotor Wake Model

7.8.2 Airfoil Behavior at High Angles of Attack

7.8.3 Spanwise Flow Effects

8.0 ROTOR AND CONTROL SYSTEM LOADS

8.1 XV-15 Metal Blade

8.2 Baseline Advanced Technology Blade (ATB)

8.3 Alternate Configurations: Advanced Technology Blade

9.0 ACOUSTICS

10.0 CONCLUSIONS AND RECOMMENDATIONS

10.1 Conclusions10.2 Recommendations

11.0 REFERENCES

Appendix A - Corrections for Effects of Wind

A.I Introduction

A.2 Correction for Wind Effects

A.3 Data Reduction and Correction Procedure

Page

96

99

99

99

101

101

125

125

128

128129

129

130

130

130

132

vi

CR177436

Figure No.

1.1

1.2

3.1

3.2

3.3

3.4

3.5

3.6

3.7

3.8

3.9

3.10

3.11

3.12

3.13

3.14

3.15

3.16

3.17

LIST OF FIGURES

Title

XV-15 Tilt Rotor Research Aircraft in

Cruise, Transition, and Hover

Baseline ATB Mounted on NASA-Ames

OARF (Outdoor Aerodynamic Research Facility)

Advanced Technology Blade

Pretest Predictions and OARF Test Results for

XV-15 Metal Blade and Baseline ATB Rotor

Plan View of NASA-Ames OARF

Three-View of Test Stand

Gearbox Operating Envelope

Six-Component Rotor Balance System

Schematic of Balance System

Baseline ATB Chord Distribution

Baseline ATB Twist Distribution

Baseline ATB Thickness/Chord Distribution

Baseline ATB Chordwise Stiffness Distribution

Baseline ATB Flapwise Stiffness Distribution

Baseline ATB Torsional Stiffness Distribution

Baseline ATB Pitch Inertia Distribution

Baseline ATB Mass Distribution

Comparison of Pretest Frequency Predictions withPer Rev Crossings and Spectral Analysis Peaksfor Collective and Cyclic Modes

Advanced Technology Blade Basic Material Assembly

Baseline ATB with Alternate Tip and Cuff Configurations

Planform, Twist, and Airfoil Distributions for XV-15

Blades

Paqe

5

7

8

10

11

12

15

15

16

16

17

17

18

18

19

20

21

23

vii

CR177436

Figure No.

4.1

4.2

6.1

6.2

6.3

6.4

6.5

6.6

6.7

6.8

6.9

6.10

6.11

7.1

7.2

7.3

7.4

7.5

LIST OF FIGURES (Continued)

Title

Summary of Instrumentation

Wake Rake Details

Test Schedule for XV-15 and ATB

XV-15 and ATB Test Run Log

Check Calibration Results:

Check Calibration Results:

Check Calibration Results:

Check Calibration Results:

Check Calibration Results:

Check Calibration Results:

Check Calibration Results:

Check Calibration Results:

Thrust (From Rotor Balance)

Thrust (From NASA Load Cells)

Torque (From Rotor Balance)

Torque (From NASA Load Cells)

Thrust with Torque Load

Applied (From Rotor Balance) 45

Thrust with Torque Load

Applied (From NASA Load Cells) 46

Torque with Thrust LoadApplied (Corrected for Friction

Torque and AFFLEX Interaction) 47

Torque with Thrust Load

Applied (From NASA Load Cells) 48

Effect of RPM on the Correlation of Balance Thrustwith Load Cell Thrust

XV-15 CT vs. Cp

XV-15 Figure of Merit

XV-15 Thrust Coefficient vs. Collective Pitch

Variation of Cp with CT3/z for XV-15 Rotor

Variation of Induced Efficiency Factor with ThrustCoefficient for XV-15 Rotor

Paqe

24

28

31

32

41

42

43

44

49

51

52

54

55

56

viii

CR177436

Figure No.

7.6

7.7

7.10

7.11

7.12

7.13

7.14

7.15

7.16

7.17

7.18

7.19

7.20

7.21

7.22

7.23

7.24

LIST OF FIGURES (Continued)

Title

Effect of Tip Mach Number on Thrust/Power

for XV-15 Blades

Effect of Tip Mach Number on Figure ofMerit of XV-15 Blades

Tip Vortices of XV-15 Metal Blades

Contracted Wake Shape of XV-15 Rotor Deduced

from Tip Vortex Photographs

Distribution of Downwash Velocities for Various

Thrust Coefficients for XV-15 Rotor

CT vs. Cp for Baseline ATB

Figure of Merit for Baseline ATB

Thrust Coefficient vs. Collective Pitch for

Baseline ATB

Variation of Cp with CT3/2 for Baseline ATB

Distribution of Downwash Velocities for Various Thrust

Coefficients for Baseline ATB

CT vs. Cp for Baseline ATB with Extended Cuff

Figure of Merit for Baseline ATB with Extended Cuff

Thrust Coefficient vs. Collective Pitch for Baseline ATB

with Extended Cuff

Variation of Cp with CT3/z for Baseline ATB withExtended Cuff

CT vs. Cp for Baseline ATB with No Cuff

Figure of Merit for Baseline ATB with No Cuff

Thrust Coefficient vs. Collective Pitch for Baseline ATB

with No Cuff

Variation of Cp with CT3/2 for Baseline ATB with No Cuff

CT vs. Cp for ATB with Swept Tip

Paqe

57

58

59

60

61

63

64

65

66

67

68

69

70

71

72

73

74

75

77

ix

CR177436

LIST OF FIGURES (Continued)

Fiqure No. Title

7.25 Figure of Merit for ATB with Swept Tip

7.26 Thrust Coefficient vs. Collective Pitch for ATB with

Swept Tip

7.27 Variation of Cp vs. CT_/2 for ATB with Swept Tip

7.28 CT vs. Cp for ATB with Square Tip

7.29 Figure of Merit for ATB with Square Tip

7.30 Thrust Coefficient vs. Collective Pitch for ATB with

Square Tip

7.31 Variation of Cp vs. CT_/z for ATB with Square Tip

7.32 Comparison of CT vs. Cp for XV-15 Metal Blade andBaseline ATB as Measured on OARF

7.33 Comparison of Figure of Merit for XV-15 Metal Bladeand Baseline ATB as Measured on OARF

7.34 Effect of Tip Shape on CT vs. Cp for ATB with

Baseline Elliptical, Swept, and Square

Tips vs. XV-15 Metal Blade

7.35

7.36

Effect of Tip Shape on Figure of Merit for ATBwith Baseline Elliptical, Swept, and Square

Tips vs. XV-15 Metal Blade

Effect of Cuff on CT vs. Cp for Baseline ATB with

Elliptical Tip-Comparison of Truncated (Baseline),Extended, and No Cuff vs. XV-15 Metal Blade

7.37 Effect of Cuff on Figure of Merit for Baseline ATB

with Elliptical Tip-Comparison of Truncated (Baseline),Extended, and No Cuff vs. XV-15 Metal Blade

7.38 Effect of Tip Shape on Cp vs. VTI - RPM, and MTI P -Comparison of Baseline EllipticalPTip and Swept Tip

7.39 Effect of Tip Shape on C! vs. Collective - Comparison ofBaseline ATB Elliptical _ip, Swept, and Square Tips

Page

78

79

8O

81

82

83

84

85

86

87

88

89

90

91

92

CR177436

Figure No.

7.40

7.41

7.42

7.43

7.44

7.45

8.1

8.4

8.5

8.6

8.7

8.8

8.9

8. i0

LIST OF FIGURES (Continued)

Title

Effect of Tip Shape on Induced Efficiency Factor -

Comparison of Baseline ATB Elliptical Tip with

Swept and Square Tips

Effect of Cuff Shape on Induced Efficiency Factor -

Comparison of Truncated (Baseline), Extended, and

No Cuff

Comparison of Downwash Distributions for XV-15Metal Blade and Baseline ATB Rotor

Tip Vortices of Baseline ATB with Extended Cuff

Comparison of Calculated and Measured Performance

on the Advanced Technology Blade Rotor

Local Lift Coefficients, C_, at Various RadialSections on a Rotating Propeller (Reference 7)

XV-15 Metal Blade:

Moments vs. CT

XV-15 Metal Blade: Pitch Link Loads vs. CT

Baseline ATB: Hub Spindle Resultant Flap Bending

Moments vs. CT (Runs 32 and 36)

Baseline ATB: Hub Spindle Resultant Flap Bending

Moments vs. CT (Run 50b)

Baseline ATB: Hub Spindle Steady Flap Bending

Moments vs. CT - Effect of RPM

Baseline ATB: Hub Spindle Alternating Flap Bending

Moments vs. CT - Effect of RPM

Baseline ATB: Hub Spindle Steady Chord Bending

Moments vs. CT - Effect of RPM

Baseline ATB: Hub Spindle Alternating Chord

Bending Moments vs. CT - Effect of RPM

Baseline ATB: Hub Spindle Steady Flap Bending

Moments vs. RPM

Baseline ATB: Hub Spindle Alternating Flap Bending

Moments vs. RPM

Hub Spindle Resultant Flap Bending

93

94

95

97

98

100

102

103

104

105

-I06

107

108

109

II0

111

xi

CR177436

LIST OF FIGURES (Continued)

Figure No.

8.11

8.12

8.13

8.14

8.15

8.16

8.17

8.18

8.19

8.20

8.21

8.22

8.23

g.1

9.2

A.1

A.2

Title

Baseline ATB: Hub Spindle Steady Chord BendingMoments vs. RPM

Baseline ATB: Hub Spindle Alternating Chord

Bending Moments vs. RPM

Baseline ATB: Pitch Link Loads vs. CT (Runs 32 and 36)

Baseline ATB: Pitch Link Loads vs. CT (Run 50b)

Baseline ATB: Steady Pitch Link Loads vs. CT -Effect of RPM

Baseline ATB: Alternating Pitch Link Loads_vs. CT -Effect of RPM

Baseline ATB: Steady Pitch Link Loads vs. RPM

Baseline ATB: Alternating Pitch Link Loads vs. RPM

Baseline ATB: Flap Bending Moments at O.IOR vs. CT

Baseline ATB: Flap Bending Moments at O.IOR vs. CT

Baseline ATB: Steady Flap Bending Moments at 0.85R

vs. CT - Effect of RPM

Baseline ATB: Alternating Flap Bending Moments at0.85R vs. CT - Effect of RPM

Effect of Blade Sweep (Lag Angle) and Tip Shape on SteadyPitch Link Loads

Typical Near-Field Noise Data for XV-15 Metal Bladeand ATB

Typical Far-Field Noise Data for XV-15 Metal Bladeand ATB

Effect of Wind on Induced Power

Effect of Wlnd on Hover Performance

Page

112

113

114

115

116

117

118

119

120

121

122

123

124

126

126

133

134

xii

CR177436

Table No.

3.1

4.1

9.1

LIST OF TABLES

Title

Rotor Balance Load Range and Accuracy

Placement of Blade Strain Gages

Acoustic Data Recorded During Rotor Tests at NASA-Ames

Outside Aeronautical Research Facility (March 'B4 -

August '84)

14

25

127

LIST OF SYMBOLS

a

A1

B

B1

CT

Cd

C_

CNF

Cp

Cppro

Definition

airfoil section lift-curve slope

lateral cyclic

number of rotor blades

longitudinal cyclic

blade chord

blade thrust-weighted chord,

section drag coefficient

I_cx2dx

section lift coefficient

rotor normal force coefficient, NF/pnR2VT 2

rotor power coefficient, HPxSSO/p_R2VT 3

rotor profile power coefficient, profile power/p_R2VT3

Unit____ss

rad- i

degrees

degrees

ft.

ft.

xiii

LiST OF SYMBOLS (Continued)

CR177436

Definition Units

CP1 rotor induced power coefficient, induced power/p_R_VT3 -

CT

DL

rotor thrust coefficient, T/p_R2VT2

wing download lb.

EIc

EIF

blade chordwlse stiffness

blade flapwise stiffness

Ib.ln. 2

Ib.ln. 2

FM rotor figure of merit, CT312/V_

acceleration due to gravity ft./sec. 2

GJ rotor torsional stiffness Ib.in. 2

Ip blade pitch inertia Ib.in. 2

MT}MTIP

NF

rotor induced efficiency factor

blade tip Mach Number

rotor normal Force lb.

PM

Psi

Pti

r

rotor pitching moment

static pressure at the ith tube in the downwash rake

total pressure at the ith tube in the downwashrake

radial station on blade

in. lb.

lb./ft. 2

lb./ft. 2

ft.

xiv

|

)

R

R e

RM

SF

T

Too

vi

VT

VTIp }

VWIND

X

YM

Of

_w

LIST OF SYMBOLS (Continued)

Definition

blade radius

airfoil Reynolds Number

rotor rolling moment

rotor side force

rotor thrust

rotor thrust out of ground effect

induced velocity

rotor tip speed

wind velocity

wind velocity

nondimensional blade position, r/R

rotor yawing moment

shaft angle of attack

hub gimbal angle

rotor rotational speed

ambient wind azimuth

CR177436

Units

ft.

in. lb.

lb.

lb.

lb.

ft./sec.

ft./sec.

ft./sec.

knots

in. lb.

degrees

degrees

radians/sec.

or RPM

degrees

XV

LIST OF SYMBOLS (Continued)

CR177436

air density

Definition Units

slugs/ft. 3

o local rotor solidity, Bc/_R

OT

6)o

thrust weighted rotor solidity, BCT/_R

blade root collective pitch angle degrees

e.75 blade collective pitch angle at r/R - 0.75 degrees

rotor advance ratio, V/VTI P

Subscript Definition Units

conditions when wind velocity is zero

xvi

CR177436

4

1.0 SUMMARY

This document presents isolated rotor test results conducted in two phasesat Ames Research Center in March and July/August 1984. In March a benchmark

test of the XV-15 steel blades was performed and in July/August the Advanced

Technology Blade, including a number of variations, was tested. Betweenthese two test periods, the facility was occupied by a scaled version of theV-22 rotor. All rotors tested were 25 Ft. in diameter. The V-22 test is

reported in Reference I.

Figure 1.1 shows the XV-15 aircraft in several modes of operation. Figure1.2 shows the baseline version of the ATB mounted on the test stand at NASA-

Ames. Figure 1.3 shows the untwisted blade planform and distribution ofairfoil sections for the baseline configuration.

The performance indices of both the XV-15 rotor and the ATB rotors turnedout to be significantly better than expected. Predicted and test values of

figure of merit as a function of CT for both rotors are shown in Figure 1.4.These results, along with those from configuration variations, are discussedin detail in Section 7.0. Possible reasons for the poor quality of the pre-

dictions are identified, and suggestions are made for improvements in pre-

dictions capability.

The test facility used for this program was the Outdoor Aeronautical Re-

search Facility at Ames Research Center. Major improvements to the power

transmission and performance measuring components of the NASA test rig were

funded under the program. These included provision of a 4:1 reduction gearbox which permitted testing beyond the power levels available in the XV-15

aircraft. A major improvement in the test hardware was the development of a

six-component balance with minimal load interaction, absence of thermaldrift, and direct measurement of rotor thrust and torque. These features of

the program are discussed further in Section 3.0 and in more detail in Refer-

ence 2, ......

The tests provided definitive thrust and torque data for the XV-15 steelblade and the Advanced Technology Blades. Hover performance for both rotors

(and for the V-22 rotor tested using the same equipment) was significantlybetter than that predicted using contemporary theory. The measured peak

figures of merit were all in the region .79 - .B1, whereas predicted valuesdid not exceed .79. In addition, peak performance occurs at a higher value

of CT than predicted, and does not drop off as Fast as predicted at the

higher values of CT.

The airflow velocity distribution was measured at a distance approximatelyO.4R downstream from the rotor plane. Vapor trails of the tip vortices were

generated at high CT'S in some atmospheric conditions and photographs ofthese have been used to estimate the rate of wake contraction and velocity

variation as a Function of distance from the rotor planes. These additionaldata have been used to initiate improvements in prediction methodology.

ORIG_,TAILPAG_ L_

OF POOR QUALYI_

Figure 1.1 XV-15 Tilt Rotor Research Aircraft in Cruise,Transition, and Hover

CR17743!

CR177436

ORIGINAL PAGE Ig

p_ !_OOR QUALITY

Figure 1.2 Baseline ATB Mounted on NASA-Ames OARF(Outdoor Aerodynamic Research Facility)

3

CR177436ii i

!

:z

z

o.

z

z

l--

Z

wJ

o

li

I--(/1I,-.4

®"0

m

m

0m

0C

t.1@

I--

"00_JCe8)

"0,<

O0

@

14.

CR177436

.84

.80

.56

.52

.48

.44

XV-15

TEST

ATBOARF TEST FAIRING

XV-15OARF TEST FAIRING

_.____.__ .....I "_ _ ATB LIFTING-LINE

/ _-"\/_'_._ \ THEORYPREOICTIO_

,, \ ,

/ XV-15 LIFTING-LINE \

THEORY PREDICTION

II

/I

I

/I

!

/I

l XV-15 METAL BLADEE

WRIGHT-PATERSON AFB

ATB WITH BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFE

..... LIFTING-LINE PRE-TEST PREDICTION

Figure 1.4

* VTI P = 740 FPS

** VTI P = 766.2 FPS

*** = 743.5 FPSI I I VTIP I . I I

.006 .008 .010 .012 .014 .016

ROTOR THRUST COEFFICIENT (CT)

-I.004

Pretest Predictions and OARF Test Results forXV-15 Metal Blade and Baseline ATB Rotor

_------_ OARF TEST (FAIRING OF*'*RUNS 15,22,23,25,26)

!.01_

WHIRL TOWER TEST

LIFTING-LINE PRE-TEST PREDICTION* ..-.- OARF TEST (FAIRING OF***"--" RUNS 32,33,36,37 ,SO}

CR177436

Blade bending moments were recorded at a number of stations and criticalstation_were monitored for safety. This included data near the tip which

has been used to improve the mathematical modelling in this region. The

test included eight hours of endurance and structural validation testing

during which control inputs were cycled and blade frequency data wasaccumulated.

Over the test period Ames Research Center engineers monitored the near and

far-field noise levels generated by rotor operation. The Advanced Tech-

nology Blades were found to generate significantly less noise than the XV-15metal blades.

A summary of the rotor performance results along with conclusions and

recommendations are given in Section 10.0.

-. 2.0 INTRODUCTION

The XV-15 tilt rotor demonstrator aircraft has been flying successfullysince 1977 using rotor blades of the original design. These blades have a

rectangular planform and the material is steel. The blade design was opti-

mized for a 9000 Ibs. gross weight aircraft and there was no change to therotor design when the gross weight became 13,000 Ibs.

This, along with a number of other factors including fatigue strength limita-tions of the metal blades, led NASA to initiate a program to develop compos-

ite blades optimized to different performance criteria and exploiting the

range of design options made feasible by composite materials. Other objec-tives of the Advanced Technology Blade program were to demonstrate fabrica-

tion techniques appropriate to highly twisted composite blades suitable for

tilt rotor applications.

This report documents the XV-15 Advanced Technology Blade Rotor Test con-ducted at the Outdoor Aeronautical Research Facility (OARF) at NASA-Ames

during April 1984 and July 1984. The report includes a description of thetest apparatus and instrumentation, a presentation of the results, and adiscussion of the implications of the results. The detailed, fully-cor-

rected test data may be obtained from NASA-Ames 40' x 80' wind tunnel staff

in the form of computer tabulations.

3.0 DESCRIPTION OF TEST INSTALLATION

3.1 Test Stand

The Outside Aerodynamic Research Facility at NASA-Ames consists of a large

concrete pad (Fig. 3.1) with a steel platform at the center of which is

mounted a test stand carrying the propeller test rig. Details of the layoutof the stand are given in Fig. 3.2. The supporting test stand consists of a

horizontal frame carrying the motor and drive system. This frame is sup-ported in front by two braced vertical steel beams and in the rear by a

single, smaller beam. The rotor centerline lies 22 feet above the metal

CR177436

N

CONCRETE PAD

METALPLATFORM

/

- ]

\PROP. RIG

\75 TON

TRAVELLING CRANE

r --- m _ _ "_)MPUTER

INSTRUMENTATION IRACKS

CONTROL ]CONSOLE

ELEVATED ._

OBSERVATIONPLATFORM

CONTROL ROOM MOTOR-GENERATOR ROOM

(BELOW GROUND)

Figure 3.1 Plan View of NASA-Ames OARF

CR177436

_q

"0¢:

cn

W

l'-

qlm

0

@m

I

@l,,

.I=I'-

_N

@im

I.I..

CR177436

platform providing 9.5 feet of clearance between the blade tips and the

ground. The rotor hub, controls, six-component balance, gearbox and elec-tric motor with services are mounted in-llne within a 28-inch diameter cy-

lindrical cowling. The motor housing is mounted on three load cells to

provide rotor force and moment data that is independent of, and supplementalto, data from the main balance. The propeller test stand is described inmore detail in Reference 2. Additional details and strength and safety

analyses are provided in Reference 3.

3.2 Motor and Drive System

The test stand is powered by an electric motor driving through a new 4:1

reduction gearbox. The gearbox is oil cooled; the motor is water cooled.

Gearbox output shaft torque limit is 252,000 in.lb, corresponding to theelectric motor limit of 3,000 HP at 3000 RPM. This was sufficient to test

well beyond the current (design maximum 163,000 in.lb.) operating torquelimit of the XV-15. The gearbox is a Cincinnati Gear Co. epicyclic gearunit with a modified aft case to interface with the NASA motor package. The

gearbox unit mounts directly to the Face of the motor unit and supports therotor balance through the balance mounting ring. The system consists of a

sun gear around which are arranged a number of planets rolling within anannulus providing a coaxial design with power transmission at more than one

point. Fig. 3.3 presents the gearbox operating envelope and shows that thehover RPM of the ATB and XV-15 rotors (565 RPM) is within the available

operating range. Maximum RPM was limited to 625 RPM by the blade retentionstrap. Operating time below 370 RPM is also limited because of gear tooth

and bearing lubrication considerations. However, this RPM is below the

present range of interest. The motors and gearbox may rotate in either di-rection, however all the rotors tested were designed to rotate in the clock-

wise direction (viewed from the rear).

3.3 Balance

The test stand is furnished with a six-component balance. As shown in Fig-

ures 3.4 and 3.5, the rotor balance system is mounted between the hub/stack

assembly baseplate and the transmission (through the balance mounting ring).The balance has two sections. The front section is a multi-flexured,

torque-sensing element which measures the frictional torque of the bearings.The rear thrust-measuring section of the balance system consists of two

flexure plates mounted on either end of cylindrical spacer units. Theseflexure elements measure thrust and also normal force, side force, pitching

moment and yawing moment. The primary torque measurement is made by strain

gages mounted on the drive shaft forward flexible coupling. Additionalstrain gages on the flexible couplings measure the axial load in the driveshaft. This is a function of the axial motion of the main thrust measuring

flexures and amounts to approximately 3% of the total load.

Balance strain gages are of the foil type and are temperature compensated.

The primary sensitivities are in the thrust and torque directions with amaximum error of 50 lb. of thrust and 25 in.lb, of torque. The balance is

designed to withstand the loss of one rotor blade without yielding and has

9

CR177436

280000

240000

200000

Iz

_" 160000

m 120000

80000

40000

MAX ROTOR SPEED " 625 RPM

(AS LIMITED BY BLADE RETENTION)

MAX ROTOR TORQUE" 252,000 IN-LB

BOEING VERTOL

DESIGN REQUIREMENTSMIN & MAX RPM FOR MAX-

TORQUE = 360 - 650 RPM

XV-15DESIGNHOVER

RPM

100 200 300 400 500

OUTPUT SHAFT SPEED (RPM)

600

Figure 3.3 Gearbox Operating Envelope

700

lO

CR177436

(

\

.ORIGINAU PAGE IS

DE POOR QUALITY.

\

\\

\

EQi

,ll

tO

@OC

u

m

o

oee

CIDC0Q.

E0oIx

c_

,It

c4@L_

IJ.

11

CR177436

t

g,

,-r

12

CR177436

infinite life over the normal operating load range. Axial load range is

-400 to 16,000 lb. and _he torque range is 0 to 252,000 in.lb. Table 3.1

summarizes the design load ranges and accuracies for the rotor balance. The

flexible couplings are designed to measure a maximum torque of 252,000 in.lb.

with an accuracy of ± 120 in.lb.

3.5 Hub and Controls

The 3-bladed gimballed rotor hub and upper controls are XV-15 rotor compo-nents defined by BHT Drawing No. 300-018-012. The ATB pitch housing was

designed to be compatible with this hub. The upper controls provided con-trol by collective, longitudinal and lateral pitch. Collective pitch motionwas transmitted through the center of the shaft and was controlled by a hy-

draulic actuator. Longitudinal and lateral cyclic motion was provided

through the rotating swashplate. The motions of the nonrotating swashplatewere controlled by electric linear actuators. Both collective and cyclic

pitch actuator control systems were open loop with the electric cyclic pitchactuators rate-limited to 0.5 deg/sec. Collective pitch motion was limited

to a range of -4 to 25 degrees; cyclic pitch was electrically limited to ±

3.0 degrees and a mechanical stop was provided at ± 4.0 degrees.

The complete hub/stack assembly was mounted on a base plate (actuator plate)which was also the mounting point for the control actuators and the connect-

ing element to the balance system. A slipring assembly with 48 rings wasincorporated within the stack to provide transmission of data from the ro-

tating components to the data acquisition system.

A cowling covered the upper controls and balance and was attached to the

motor casing. The cowling provided weather protection.

3.6 Rotors

3.6.1 Advanced Technology Blade Rotor

The ATB rotor is a three-bladed, 25 ft. diameter rotor with a thrust-weight-

ed solidity (oT) oF 0.10, which is 12.3% more than the solidity of the XV-15steel blades (.089). The blades are of composite construction. Theoretical

blade chord, twist, and thickness/chord distributions for the baseline ATB

are given in Figures 3.6, 3.7, and 3.8 respectively. Sectional propertiesare given in Figures 3.9 through 3.13. Estimated blade frequencies in the

cyclic and collective modes are shown in Figure 3.14 along with test measure-ments. The blades were inspected for fidelity to the design values of twist,

chord, airfoil contour and surface condition, and were Found to be acceptable.

The rotor blades were instrumented to record flap, lag, and torsional mo-

ments at selected spanwise positions. Details of the instrumentation are

given in Section 4.0, Table 4.1.

Figure 3.15 is an exploded view of the advanced technology blade with call-outs of the various materials. Figure 3.16 presents the baseline ATB con-

figuration with the alternate tip and cuff sections that were tested.

13

CR177436

Table 3.1 Rotor Balance Load Range and Accuracy

COMPONENT LOAD RANGE ACCURACY%OFMAX.LOAD

ii

AXIAL FORCE (THRUST) -400/16,000 LB + 50 LB 0.3

NORMAL FORCE + 600 LB + 12 LB 2.0

SIDEFORCE + 600 LB + 12 LB 2.0

PITCHING MOMENT + 20,000 IN-LB + 400 IN-LB 2.0

YAWING MOMENT + 20,000 IN-LB + 400 IN-LB 2.0m

ROLLING MOMENT

(FRICTION TORQUE)• ,,r I

+ 15,000 IN-LB + 25 IN-LB 0.16

14

CR177436

I,i-r-

Z

i,i

v

I--

28

24

20

16

12

m

8 --

4 --

0

40 --

30--

20 -

I0 -

00

I I 1 I l I.2 .3 .4 .5 .6 .7

RADIAL STATION (r/R)

I I I.8 .9 1,0

Figure 3.6 Baseline ATB Chord Distribution

.1 .2 .3 .4 .S .6 .7

RADIAL STATION (r/R)

Figure 3.7 Baseline ATB Twist Distribution

15

CR17743(

(J-....4J

0

¢.J

,,,1-F-

.4 -

.3 -

o_ _

o I I , I I I I I I I• [ .2 .3 .4 .5 .6 .7 .8 .9

RADIAL STATION (r/R)

Figure 3.8 Baseline ATB Thickness/Chord Distribution

I1.0

lO00

800

X

oa 600

_ 400

_J ZO(]

00

m

I I I I I I I I.1 .2 .3 .7 ,8.4 .5 .6

RADIAL STATION (r/R)

,9 I.O

Figure 3.9 Baseline ATB Chordwise Stiffness Distribution

]6

CR177436

5OO

40O

e,-i

x

300z

I

"" 200

_JI.z.

1,--4I00

120

100

x 80

z

I60

40

2O

w

I t I I I I i I I I

,1 .2 .3 .4 .S .6 .7 .8 .9 I.O.

RADIAL STATION (r/R)

Figure 3.10 Baseline ATB Flapwise Stiffness Distribution

370

I I I I I I 1 I "--P----_ I• l .2 .3 .4 .5 .6 .7 .8 .9 1.0

RADIAL STATION (r/R)

Figure 3.1 1 Baseline ATB Torsional Stiffness Distribution

17

CR177436

12 -

ZI,-q

z

I

..J

• _ 6 -

c_L_Jz

-r

_- 2 -

0

3.2

2.8

_-_ 2.4z

I

co.J_-_ 2.0

z0h-q

:_ 1.6 -CO

1.2 -

(.r)

.8 -

00

I I I I I I

.1 .2 .3 .4 .5 .6

RADIAL STATION (r/R)

WEIGHTS

I r I "I.7 .8 .9 1.0

Figure 3.12 Baseline ATB Pitch Inertia Distribution

B

I I

.1 .2

Figure 3.13

I I " I I I 1

.3 .4 .5 .6 .7 .8

RADIAL STATION (r/R)

Baseline ATB Mass Distribution

I I

.9 l.o

18

CR177436

v

>-_.)zw

CD"

IJ.

ATB COUPLED ROTOR NATURAL FREQUENCIES

N

:)..OZ

C"

,.,.,l.i.

70-

60-

50-

40-

PRETEST PREDICTIONS_

TEST RESULTS x (e.75

COLLECTIVE MODES

/

• I0°)

/

,/ xx/

/

/ 3/REV

/21REV

0 o

Figure 3.14

I00 2oo 3oo 4o0 5oo 6oo 7uoROTOR SPEED (RPM)

Comparison of Pretest Frequency Predictionswith Per Rev Crossings and Spectral AnalysisPeaks for Collective and Cyclic Modes

19

CR177436

_00"

m

.12

E

!

m

@

m

Io

i

m

0u

0c

I-,

(.1

>

<

tD

@L_

20

CR177436

ILlZm

..IUJ

m

a.l-

IJJZ..J--<"JoIJJ--O_l--<a.m.j

..IIJJ

M.IL

0

W

<0Z

rrI-

a.I-

O.UJ

U)

a.m

ILl

<

0U)

r

U.u.0U.U.

0

U.1,1._

com4_

(0b.

mq_m

c0

b

0

"ot.-

0.!

I,.-

G)4,,*

cL

04dm

,<

J::4_!

mI-.-<

o

(0rn

(D_m

otm

u.

21

CR177436

3.6.2 XV-15 Rotor

The XV-15 blades tested in this program were the same full scale blades that

had been previously tested" by Bell Helicopter Textron Corporation on the

Wright Patterson Air Force Base (WPAFB) whirl tower during the XV-15 devel-

opment program (Reference 4). The planform, twist, airfoil and thickness/chord distributions are shown in Fig. 3.17.

4.0 INSTRUMENTATION

4.1 Instrumentation - General

The instrumentation installed for the ATB and XV-15 steel blade tests is

indicated in Figure 4.1. This shows the type of data measured and whichvariables were monitored for safety. All data was recorded on magnetic

tape. All non-steady state variables were subject to high speed sampling.

4.2 Rotor Balance Instrumentation

The rotor balance was instrumented to measure six components of rotor force

and moment: thrust, sideforce, normal force, pitching moment, yawing mo-

ment, and rolling moment. Thrust and rolling moment measurements were sig-

nificantly more sensitive than the others, as indicated in Table 3.1. Thedrive shaft flexible coupling was instrumented to measure torque and axial

force. The balance rolling moment (bearing friction torque) was subtracted

from the shaft torque to provide the net rotor torque. The drive shaftaxial force was added to the balance thrust measurement to give the rotor

thrust. Balance temperature was continuously monitored by thermocouples.

4.3 Blade and Hub Instrumentation

The instrumented blade was strain gaged to measure torsion, flap bending,

and chord bending at the radial stations shown in Table 4.1. All gages were

mounted on the spar beneath the airfoil contour.

Hub instrumentation was provided to measure control system position (e.75,

At, BI), hub gimbal angles, root collective, pitch-link load, and hub yokemoments. Transducers were installed in the rotating system Chub and drive

system) to measure the following:

(a) Blade pitch angle - measured by a potentiometer mounted on the blade

housing. The potentiometer was a custom-fit resistance element with

wiper arm.

(b) Hub gimbal angle - measured by a potentiometer attached to the gimbalinside the hub.

(c) Hub yoke bending moments - measured by strain gages mounted on the hub

spindles; both flapwise and chordwise gages were provided.

22

CR177436

BLADE STATION - INCHES

20 40 60 80 100 120 140 150

CHORD LINEi

t

3O

2O

I0

0

NACA 64X28/

t/c - %

./ t64X18

TWIST - DEGREES t64X12 t

64X08

20 40 60 80 I00 I 140 150

BLADE STATION - INCHES

-i0Figure 3.17 Planform, Twist, and Airfoil Distributions for XV-15 Blades

nl

23

CR177436

,oooo,ooo iooTToTTT°' ii....0,o°°iTT!!°°iii

o __" '" " I" I I

eo I

_ ..J

\.-,..

X

I/I

Z

w-,.Jv1

J

I: 11I'I

i _ t rS"

J

[

/

CO

!

EL_

u

O

SE

4OL_

U.

24

CR177436

Table 4.1 Placement of Blade Strain Gages

PERCENT RADIUS

FLAP

CHORD

TORSION

10.0 (1) 20.6 29.0 49.0 69.0 84.0

10.0 (1) 20.6 30.0 50.0 70.0 85.0

- - 30.0 51.0 71.0 -

NOTE: 1) ON PITCH HOUSING

2) ON HUB SPINDLE THERE WERE IN- AND OUT-OF-PLANEBENDING GAGES AT 6% RADIUS

3) SEE RUN LOG FOR ADDITIONAL INFORMATION

25

CR177436

(d) Pitch link load - measured by a strain gage bridge on the pitch link.

(e) Flexible coupling torque - measured by strain gage bridges (active andspare) on the forward flexible coupling of the drive shaft.

(f) Flexible coupling axial load - measured by strain gage bridges on theforward Flexible coupling.

(g) Forward shaft bending - measured by (2) perpendicular bending bridgesmounted on the rotor shaft.

(h) Rotor I/rev and 512/rev - measured by a phototachometer on the drive

shaft.

(i) Hub acceleration - measured by accelerometers mounted on support struc-ture near the hub.

The signals from the rotating system were transferred to the Fixed system

through a 48-ring slipring assembly. As configured, the test stand was lim-ited to 10 channels on the slipring. For the hover performance test, the

recorded parameters and their corresponding slipring channels requirements

were as follows:

Parameter Channels Required

*Shaft torque

*Shaft axial load (AFFLEX)

*Pitch housing flap bending @ r/R = .10

*Pitch housing chord bending @ r/R z .10

*Pitch link load

Root collective

*Gimbal angle

Blade flap bending @ r/R = .31

*Hub yoke chord bending

*Hub yoke flap bending

*Required for safety

Although only 10 channels were available on the slipring at any one time,these channels could be reassigned to read other strain gages, if desired.

26

CR177436

4.4 Wake Rake

A wake rake consisting of 22 pitot-static tubes was mounted behind the rotor

disc plane at the station corresponding to the wing upper surface. The pur-pose of the rake was to measure the isolated rotor slipstream velocities and

angles under different rotor operating conditions and to use this data tounderstand the structure of the rotor slipstream and the wing download andits distribution. The wake rake was connected to a Scanivalve to measure

the pressures. The wake rake and the spacing of the pitot-static tubes is

shown in Fig. 4.2.

4.5 Anemometer

A wind speed and direction transducer was installed on a narrow tower ap-

proximately 200 feet north and 200 feet east of the rotor hub centerline.The indicator was on approximately the same level as the rotor hub. The

signals from the transducer were fed to the data acquisition equipment inthe control room.

4.6 Acoustical Measurements

Near-field and far-field noise levels were measured. The near-field micro-

phone represented a point on the side of the fuselage of a typical tiltrotor in hover. Far-field noise was recorded by an array of microphones at

250 ft (76m) and 650 ft (198m) radius at O, 15, 30 and 45 degrees behind therotor disc.

5.0 DATA ACQUISITION AND REDUCTION

5.1 Data Acquisition

The NASA-Ames OARF data system provided signal conditioning and amplifica-tion for 50 data channels. Steady-state data were recorded on digital tape.

A quick-look short-form print-out was provided at the end of each run and a

detailed print-out was processed overnight. A monitor program displayed upto 15 steady-state parameters on the Test Engineer's CRT. Two analog tape

recorders were used for safety monitoring and acquisition of dynamicdata.

Complete details of the assignments of the data acquisition equipment are

given in References 2 and 5.

The following quantities were measured:

Rotor balance thrust, T (lb.)

Rotor balance normal force, NF (lb.)Rotor balance side force, SF (Ib)

Rotor balance pitching moment, PMB (in.lb.)

Rotor balance yawing moment, YMB (in.lb.)Rotor balance rolling moment, RMB (in.lb.)

Load cell axial, normal, and sideforces (lb.)Rotor RPM

Shaft torque (in.lb.)

27

CR177436

TUB_._EE r/R

I .2O2

2 .2C53 .221

4 .265

5 .289

6 .3347 .428

8 .5679 .627

I0 .655ii .72012 .75613 ,80114 .80615 .85816 .90517 .956

18 1.02319 1,07C20 1.18721 1.1722 1.22

5-POINT YAW

(9 PLACES)

PLAIN

PITGT-STATIC

PROBE

(13 PLACES)

Flgure 4.2

O

STA 84.05"

Wake Rake. Details

TUBE DETAIL

28

CR177436

Shaft axial load AFFLEX (lb.)

Hub gimbal angle, B (degrees)Blade root collective, eo (degrees)Lateral cyclic (swashplate axes), A1 (degrees)

Longitudinal cyclic (swashplate axes), BI (degrees)

Blade collective pitch, e.75 (degrees)

Blade flap moment at 31% radius (in.lb.)Blade chord moment at 10% radius (in.lb.)

Pitch housing flap moment at 10% radius (in.lb.)

Pitch housing chordwise moment (in.lb.)

Pitch-link load (lb.)

Ambient wind speed, VWIND (knots)

Ambient wind azimuth, _w (degrees)Ambient temperature (°F)Ambient barometric pressure (psi)

Relative humidity, (percent)

Hub horizontal acceleration (g)Hub vertical acceleration (g)

5.2 Data Processing

The data reduction program (Reference 6) performed the following operations:

(a) Subtracted non-rotatlng zero values.

(b) Converted corrected voltages to engineering units.

(c) Computed rotor forces and moments from load cell readings.

(d) Computed rotor balance forces and moments from balance flexure

outputs.

(e) Corrected rotor balance forces and moments for component interac-

tions through the respective balance calibration matrices.

(f) Corrected rotor balance data for temperature effects, if

significant.

(g) Corrected rotor balance thrust for flexible coupling axial load.

(h) Corrected rotor shaft torque For bearing friction (balance rolling

moment).

(i) Transferred rotor balance data to the reference body axis (rotor

hub centerline).

(j) Corrected rotor torque for wind effects, using the method present-

ed in Appendix A.

29

CR177436

(k) Computed atmospheric data from temperature, humidity, and pressuremeasured at the test site.

(1) From the corrected data, computed rotor parameters (VTIP and MTIP)

and coefficients (CT, Cp, etc.) as well as rotor horsepower and

figure of merit.

Provisions were made to harmonically analyze all rotating parameters (blade,

hub, shaft and control System) and vibratory balance flexure and fixed

system accelerometer data at Boeing Vertol.

6.0 TEST RECORD AND DATA ACCURACY

The chronology of the testing is presented in Fig. 6.1 and the XV-15/ATB

Test Run Log in Flg. 6.2. A rigorous calibration of the rotor balance had

been performed at the place of manufacture before assembly of the propellertest rig at the OARF. Following installation of the rig at Ames, anothercalibration was made which included checks for thermal drift effects on bal-

ance readings and a determination of the interaction between the torque andaxial forces at the flexible coupling. The contribution of the flexible

coupling axial load (AFFLEX) was also determined. This accounts for approx-imately 4% of the net rotor thrust. This check calibration showed that theinstalled balance was behaving to specification and that the data obtainedfrom the load cells was in close agreement with the balance data.

The XV-15 blades were installed, checked out, and testing commenced.

Initial results indicated that the rotor performance was lower than

expected. This was caused by an improper pretest procedure for obtainingR-cals in which the collective actuator was moved to maximum stroke andinduced a false load indication in the balance. When this was understood, a

new check calibration was performed using the minimum collective setting for

zeroes. The results are presented in Figures 6.3 t_rough 6.11. The check

calibration was made with thrust and torque loads applied singly and incombination. The maximum applied thrust was 7000 lb. which corresponds to a

CT of .01 at hover RPM. At this condition the error'In rotor thrust wasonly 0.03 percent, as read from the balance (Fig. 6.3). The load cellresult was 0.4 percent off (Fig. 6.4). Some hysteresis is evident in both

systems. Figures 6.5 and 6.6 present the variation of the differencebetween the actual and measured torque for a range of applied torque levels.

The torque balance error is 0.3 percent (Fig. 6.5). The load cell checkcalibration shows considerable hysteresis compared to the balance (Fig.

6.6). Figures 6.7 and 6.8 present the variation of the difference between

the applied and measured thrust at various torque levels, for a constant

applied thrust of 7000 lb. Figures 6.9 and 6.10 show the variation betweenthe applied and measured torque with a constant applied thrust of 7000 lb.The errors are essentially the same as for the case with zero torque

applied. Figure 6.11 shows the effect of RPM on the correlation of balance

and load cell thrust.

30

L_

l.U

>-,<

<I

CR177436

mI-'-,<

"0c:

tO

I

X

0tim

¢)

"o

(.ICO

I-.

,t,m

Im

om

1.6

31

CR177436

32

CR177436

i!

i

r-.

--, o

.J

c

-I

c

rr

mi-<"oc

LOtim

I>x

_N

@

.__tL

33

I Q ),. o,,a:= z

LM

IM ,m I_J 0

Qe'3o

°Wm _" Z i,-

_z

mQ

|P

I

|

CR177436

=:

o=..J

n-4mW®

I-.

rn

<

tt_1ira

I=)x;

qD

t.,

..=U.

34

CR177436

C_C r-I,,-1 OC:

35

+

%

0 I _. _ p,_

°; _ . _

4 .,

il•_ ®t

,,N ,,g

t l m! s i!-oE_

36

ORIG/NAU PAGE IS

OF POOR QUALITY

CR177436

C

==

o_.J

n-

.<

_0

[X ¸

Im

I.I.

CR177436

°

!

!B[

. _ ._ __• _ _._.

. .o_ _

.... _ z

_

" i[/s, __

Iw I I I I

c

=.

-J

I-

m ¸

I

X

I*i

37

CR177436

C

=:

--I

I-

tO

x

38

CR177436

!

i

o _ _

. _" _ g' ?

°_._ _

_ -_[_ _o"

"||||1

w..J

I---

uJI-

u.

o

i,i

,<

.1{U),<Z

p..<

m).-.<

n

l0

,<

I-Ilu

Ikl

0I

39

CR177436

lU

4[ co coa

aE,c

IQ

i

°

!Q-

"o!==i

--I

Cm

tOim

I

C_i

L..

u_

40

CR177436

BALANCE THRUST ERROR Vs. APPLIED THRUST

RUN 20® INCREASING LOAD

G DECREASING LOAD

i000 4000 5000 6000 )0

APPLIED THRUST LOAD (LB)

(FROM SINGLE AXIS LOAD CELL)

Figure 6.3 Check Calibration Results: Thrust (From Rotor Balance)

41

Figure 6.4

CR177436

LOAD CELL THRUST ERROR vs. APPLIED THRUST

RUN 20

_INCREASING LOAD

DECREASING LOAD

2O

_ .

' 0 I I / I I I I I I. looo/2000_ooo4000500060007000/

_-lOJ[-- _PPLIED THRUST LOAD (LB)

(FROM SINGLE AXIS LOAD CELL)

_ -20_-3o

Check Calibration Results: Thrust (From NASA Load Cells)

42

CR177436

120 -

i00 -

"" 80 -

,_J

i,-._ 60 -

" 40 -

I

"' 20 -

I-.-

_ 0

BALANCE TORQUE ERROR vs. APPLIED TORQUE

RUN 19® INCREASING LOAD

0 DECREASING LOAD

NOTE: TORQUEC : CORRECTEDTORQUE FROM ROTOR BALANCE

-20

-40 --

-60 -

_ "-' I i 'I

000

®

Figure 6.5 Check Calibration Results: Torque(From Rotor Balance)

43

CR177436

LOAD CELL TORQUE ERROR vs. APPLIED TORQUE

RUN 19& INCREASING LOAD

DECREASING LOAD

NOTE: QLC = TORQUE FROM NASA LOAD CELLS

120

100

i000 2000 3000 5_bo 6000 7000 8000

ACTUAL TORQUE (FT.LB.)

9000

Figure 6.6 Check Calibration Results: Torque(From NASA Load Cells)

44

CR177436

._J

]--

..rp-

40

30

20_J

.._]

_- 10

I

)--

= 0-r-)-.

'-_ -10z

..J<:

_- -20O

-30

BALANCE THRUST INTERACTION DUE TO TORQUE

WITH 7000 LB APPLIED THRUST LOAD

RUN 2O

® INCREASING

O DECREASING

I00_/2o0o 30'00_000 s_o'_OOO-_80-0-0-_oo

- APPLIED TORQUE (FT.LB.)

Figure 6.7 Check Calibration Results: Thrust with, Torque Load Applied (From Rotor Balance)

45

CR17743!

LOAD CELL THRUST INTERACTION DUE TO TORQUE

WITH 7000 LB APPLIED THRUST LOAD

RUN 20

INCREASING

DECREASING

40

30

20==-r.).-

t-,w

-_ 10a.

,.c

I

_- 0f,.,r):D

.n,-

-10._I

'-'-20,,={:C_

-30

I I I i I I I I I

I000 2000 3000 4000 5000 6000 7000 8000 9000

APPLIED TORQUE (FT.LB.)

Figure 6.8 Check Calibration Results: Thrust withTorque Load Applied (From NASA Load Cells)

46

CR177436

BALANCE TORQUE ERROR vs. APPLIED TORQUE

WITH 7000 LB APPLIED THRUST

RUN 2O

® INCREASING

140

120

100

80

no,-,, 60e,,

+_40

20

,3

d: 0X

Ro

-60

-80

NOTE:

0 DECREASING

TORQUEC = ROTOR TORQUE CORRECTED FORFRICTION TORQUE (RMRBI) AND

AFFLEX INTERACTION

RMRBI = ROLL MOMENT (CORRECTED) FROM

ROTOR BALANCE

' ' ' '

0 8000 9000

_ APPLIED TORQUE (FT.LB.)_

-i00 -

Figure 6.9 Check Calibration Results: Torque with Thrust Load Applied(Corrected for Friction Torque and AFFLEX Interaction)

47

CR177436

LOAD CELL TORQUE ERROR vs. APPLIED TORQUE

WITH 7000 LB APPLIED THRUST

NOTE:

140 -

120 -

100 -

RUN 20

INCREASING

DECREASING

QLC = CORRECTED TORQUE FROM NASA LOAD CELLS

&

80

"-"-J 60.m

N

x

+ ,_40

la.

.....,l...-z

, _ 20

g- -J

_ Z

_ _ -20_. u.l

.Ic-40

-60

-80

-I00

1000 2000 4000 5000 6000 7000 8000 9000

APPLIED TORQUE (FT.LB.)

Figure 6.10 Check Calibration Results: Torque with Thrust Load Applied(From NASA Load Cells)

48

CR177436

O0

$

(_7) l_n_Hl 773_ OV07 - lsn_H1 _NV7V8

tm

J:I--

00

0_1

J_

W

i-.

I'-

0

C

m

0

0

@

0

@

0

a.

0

@

@

i1_

49

CR177436

The rake to measure rotor induced velocity was installed and calibrated to-

ward the end of the XV-15 blade testing and only a limited amount of suchdata is available for these blades.

At the completion of the benchmark testing on the XV-15 steel blades the rig

was handed over to the V-22 program (formerly known as the JVX) for rotor

performance and download testing of a scaled rotor and semispan wing instal-lations. This V-22 test program is reported in Reference I.

At the conclusion of the V-22 testing the rig was refurbished and an inter-

mittent problem with the force readout from the drive system flexible cou-

pling gages (AFFLEX) was resolved. The AFFLEX signal is a measurement ofthe thrust force in the drive shaft when this is stretched or compressed by

flexure motions In the main balance. This component of thrust was measured

by bridges located 180 degrees apart in the flexible coupling so that i perrev components of force would cancel. One set of gages was found to be

malfunctioning and these were disconnected. The AFFLEX signal was thenrecalibrated with the rotor in the azimuthal location where the i per rev

component passed through zero.

In subsequent testing the rotor was set to this position while pre- and

post-test zeroes were being taken.

The Advanced Technology Blades were installed and testing commenced in thebaseline configuration (i.e., elliptical tip and truncated cuff). This was

followed by configuration variations which included a full airfoil cuff,

swept and square tips, cuff removed, and changes In blade sweep. A checkcalibration of all measuring systems was performed at the conclusion of

testing. This confirmed that accuracy was to the same standard of excel-

lence as at the beginning of the test program.

7.1

7.0 ROTOR PERFORMANCE

xv-15 Metal Blade Performance

The performance of the XV-15 metal blades at the nominal operating tip Machnumber of 0.69 is presented In Fig. 7.1 as a plot of rotor thrust coeffi-

cient vs. rotor power coefficient corrected to zero wind conditions. The

data was gathered during six separate runs and the data scatter is small. Amean line was faired through this data and used to calculate the rotor fig-

ure of merit shown on Figure 7.2. Note that this figure of merit curve al-

ways falls below the line faired through the individual values of figure ofmerit, calculated from each test point. This is the correct method for de-

fining the average rotor figure of merit; the average thrust - power rela-

tionship for the rotor is first determined, then quantities, such as figureof merit, which are functions of this relationship, may be computed. Peak

figure of merit for the XV-15 rotor Is 0.791 at a thrust coefficient of

0.0105. Also shown in Figure 7.2 is the performance of the XV-15 rotor astested on the Wright Patterson AFB whirl tower in 1973 (Reference 4). The

data has been adjusted for the effects of the tower. This comparison shows

that the shape of the curve is the same. The peak figure of merit occurs atthe same thrust coefficient but has a lower value.

50

CR177436

f

I

1,0 ,d"

(I_) IN31gI_30D ISQ_HI _010_

0

I'-0

I

X

b.

m

U.

51

CR177436

.84 0

•80

.76

.72

g_ .6a

_ .64

•60

NASA-AMES O.A.R.F. TEST 910

XV-15 METAL BLADE

_YMBOL RUN RPM VTIP MI'IP VWIND

0 15 587.7 769.3 .690 1.0

22 582.8 762.9 .691 0.4

23 583.7 764.1 .689 1.1

25 586.1767.2690 2626 586.2767.3.68_2_7

0

0

O0 <> OARF TEST FAIRING

/

WPAFB TEST FAIRING

(CORRECTED FOR WHIRL TOWERBLOCKAGE EFFECTS)

.56

.52

m

0

.48 - !.004

I• 006

NOTE:

.008

LINE IS FROM FAIRING OF CT vs. Cp DATA

POWER COEFFICIENT CORRECTED =OR WI._IDVELOCITYAND OIRECTION

RPM, TIP SPEED, TIP MACH NO., AND WIND VELOCITYARE THE AVERAGED VALUES FOR EACH RUN

I I I I•010 .012 .014 .016

I.018

ROTOR THRUST COEFFICIENT (CT)

Figure 7.2 XV-15 Figure of Merit

52

CR177436

Figure 7.3 presents the variation of thrust coefficient with collective

pitch. The collective pitch values have not been corrected to zero windconditions. It is estimated, however, that the correction would reduce the

collective by 1 degree, at most. Also shown on Figure 7.3 is data (withoutcorrection to collective for tower blockage) from the whirl tower test of

Reference 4. It is not known why there is a 4 degree difference between thetwo sets of data. The collective pitch settings recorded in the present

test of the XV-15 metal blades appear to be incorrect, and are presented

only as confirmation of the shape of the curve of CT vs. e.75 Calculationsusing performance codes support the WPAFB values of collective as does

flight test experience.

Maximum thrust was not reached because alternating loads increased rapidly

above a CT value of .0161. Figure 7.3 suggests, however, that a reasonable

value for maximum thrust coefficient for XV-15 would be .0165, i.e. CT/aTMAX= .185.

One measure of rotor induced efficiency is k, as defined by

CT3/2

Cp + k: CPo

where k = 1.0 corresponds to ideal induced efficiency. The value of CPo is

defined by linear extrapolation to zero thrust of the curve of CT3/2 vs. Cp.This data is presented in Figure 7.4 and was used to compute the values of k

presented in Figure 7.5.

The sensitivity of the XV-15 rotor performance to tip Mach number is pre-

sented in Figures 7.6 and 7.7. Tip Mach number was varied From 0.60 to0.73. No well-defined trend is evident although there is a tendency for

reduced performance to accompany increases in Mach number.

The distribution of downwash velocity in the wake of the rotor was measured

by the wake rake described in Section 4.5. The rake was positioned so that

the ends of the probes would coincide with the probable location of the up-per surface of a wing. At 75 percent radius the distance from the rotor

disc to the XV-15 wing surface is O.40R.

In addition to measurements of the wake, a limited series of photographswere obtained of the tip vortices made visible by water vapor condensation.

Figure 7.8 is a typical example and shows clearly the helical path of thevortices from each blade. By measuring from these photographs, M. Maisel of

NASA Ames succeeded in constructing the shape of the outer wake. Figure 7.9

shows that, for CT " .0116, the wake contracts to approximately .79R at .55Rdownstream of the disc. At O.4R where the probe lies, the tip vortex is

located at O.80R. This value is confirmed by the data of Fig 7.10 whichshows the radial distribution of downwash for selected values of rotor

thrust coefficient. Lines have been faired through the data obtained from

the pure pitot-static probes only. The data from the 5-hole angle of attackprobes was considered to be less reliable. At all the values of CT shown,

the edge of the wake appears to lie at 80 percent radius.

53

•02O

NASA-AMES O.A.R.F. TEST 910

XV-15 METAL BLADE

CR177436

.018

.016

.014

•006

0

@

o

o

o

0

o

v/"

WPAFB WHIRL TOWER (REF 3)

• VT = 740

@V T - 600

&V T - 786

0 SYM RUN RPM VTIP MTIP

22 582.8 762.9 .691 0.4

0 23 583.7 764.1 .689 1.1

0 25 586.1 767.2 .690 2.6

0 _ 26 586.2 767.3 .688 2.7

VWIN[

.004

0

-4 0

NOTE: RPM, TIP SPEED, TIP MACH NO., AND WIND VELOCITYARE THE AVERAGED VALUES FOR EACH RUN.

I ,, I I I I I

4 8 12 16 20 24

COLLECTIVE PITCH (e.75)

Figure 7.3 XV-15 Thrust Coefficient vs. C?llective Pitch

54

CR177436

oo

8

55

CR177436

o1

u'lLJJI.-

c_:_

I, I ! ! I I I

..?j..(°dO_ do)• _ _OlOV_ION3TOIJ_3a30naNI

llm

0

o r,.C,.l

o tO

I

XoO,--4 I--

o 0

C

_e

0 @• 0

0

-i_ .=I--

.-,. r.I-- .,-*

v 3

illu.

Ii0a0

clSm

0

ko

o o

I._

Q)t,,,

9o 1.6

4'

56

CR177436

I i

c_

u

E

3

o

_w

I

¢:)

cO

J

J

b

• 0

C_ W

0_L

C_

57

CR177436

NASA-AMES O.A.R.F. TEST 910

XV-15 METAL BLADE

RUN 16

.84 -- SY._MMRP__M.MVTI.....PPMTIP VWIND

._0

.76

.72

X

_- .68e,,

i,i.0

'" .64

I.I.

.60

.56

.52

0 510.9 668.8 .599 3.4

0 565.8 740.6 .663 4.7

& 589.2 771.2 .690 4.8

V 624.8 817.8 .731 4

NOTE: LINE IS FROM FAIRING OF CT vs. Cp DATA

POWER COEFFICIENT CORRECTED FOR WIND VELOCITY

ANO OIRECTION

.48 -I I [ I I , I I.004 .006 .008 .010 .012 .014 .016 .018

ROTOR THRUST COEFFICIENT (CT)

Figure 7.7 Effect of Tip Mach Number on Figure of Merit of XV-15 Bladesi im ill iiii al ii

58

CR177436

ORIGINAI2 PA-GIg Ig

OF POOR QUALITG

Figure 7.8 Tip Vortices of XV-15 Metal Blades

59

CR177436

NASA-AMES O.A.R.F. TEST 910

XV-15 METAL BLADE

1.00

.95

A

p..

90

_J

--< .85

.80

.75 _

I I. I0 .i .2

RUN 25

VTIP= 767 FPS

CT- .01159

WIND- 5 KTS AT 20

LOCATION OF

WING U/S

I L, I I

.3 .4 .5 .6

AXIAL STATION (x/R)

Figure 7.9 Contracted Wake Shape of XV-15 Rotor Deduced fromTip Vortex Photographs

60

CR177436

NASA-AMES O.A.R.F. TEST 910

XV-15 METAL BLADE

Z/R= Q.4 RUN 25

SYMTP CT RPM VTIP MTIP VWIN_D

1:3 16 .00796 585.9 766.9 .690 2.9_7 20 .01053 585.7 766.7 .690 2.4

0 22 .01246 585.5 766.4 .689 3.3

0 24 °01422 586.9 768.3 .690 2.9

V_

0 °

2O

40

V

___6o

80==

IO0

120

140

NOTE: SOLID SYMBOLS DENOTE DATAFROM ANGLE OF ATTACK PROBES

RADIAL STATION (r/R)

.2 .4. .6 .8 1.0 1.2 1.4

_ , i i , __--"--i

jNACELLE

JiiI

V

Figure 7.10 Distribution of Downwash Velocities for Various ThrustCoefficients for XV- 15 Rotor

61

CR177436

The shape of the downwash distribution changes with increasing thrust coef-

ficient, becoming more skewed toward high downwash values Just inside thetip vortex. Outside the tip vortex, the wake-lnduced velocity Is essen-tially zero; the non-zero values of downwash shown are attributable to theambient wind.

7.2 Baseline ATB Performance

The baseline ATB configuration consists of an approximately elliptical tip

planform and a cuff truncated at the trailing edge to permit g!mbal anglesup to 12 degrees at high collective pitch settings.

Thrust versus power coefficient test data is shown in Figure 7.11. The datais shown for four different runs during which tip Mach number was held

constant and for two runs at high and low Mach number. There is remarkablylittle scatter. The solid llne in Figure 7.11 is an estimated mean Faired

through the data. This Faired line is the basis of the figure of meritshown by the solid line in Figure 7.12. The individual test point figures

of merit are also shown in Figure 7.12. As noted in the precedingparagraphs the mean Figure of merit curve falls below the mean of theindividually calculated test points because of the non-linearity of the

Figure of merit function. Peak figure of merit for the baseline blade is

Just under 0.80, and remains high out to the CT obtainable at the powerlimit of the test rig. Note that a maximum value of CT - .022 (CT/oT - .22)was reached at reduced tip speed.

Figure 7.13 presents the collective pitch vs. thrust relationship and showsa change in slope between CT values of 0.006 and 0.008. As will be shown,

consistent, repeatable CT vs. e.75 relationships were obtained for all theATB configurations and are considered to be reliable'

The plot of CT3/2 versus Cp is s_own in Figure 7.14. The linear projectionto zero thrust gives a value of Po equal to 0.000185 compared with a steelblade value of 0.00013. Figure 7.15 presents downwash distributions for thebaseline ATB.

7.3 Performance of ATB with Extended Cuff

The power-thrust relationship for the ATB with the trailing edge of the cuffextended to complete the airfoil section is shown in Figure 7.16. In Figure7.17 data is presented in figure of merit format. It is seen that the cuff

extension has an effect that increases the figure of merit by approximately

0.01. Figure 7.1B shows the variation of thrust coefficient with collective

pitch and Figure 7.19 preseBts the variation of CT3/z as a Function of powercoefficient. The value of _Po deduced from Figure 7.19 is the same as thatobtained from Figure 7.14 for the blade with a truncated cuff (0.000185).

7.4 Performance of ATB with No Cuff

As expected there was a significant reduction in rotor efficiency when thecuff was removed. The results for the cuff-removed configuration are given

in Figures 7.20 through 7.23. A peak figure of merit around 0.77 was Found,and the _Po value is 0.000197. A run was made with the blade sweep angle

62

CR177436

63

, - , /, . . .

CR177436

.84

.80

.76

.72

.68

.FI--

p,,

.64

I.k.

I.&J

,,_ .60

.56

.52

.48

.44

NASA-AMES O.A.R.F. TEST 910ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFF

0 •

• 0#

e

e

SYMBOL RUN RPM VTIP MTIP VWIND

0 32 570.1 746.2 .663 0.9

& 33 570.7 747.1 .662 2.5

0 36 569.1 744.9 .661 2.0

'0 37(TpIs-zo)571.4748.0 .663 4.8

50(rP3-Iz)561.0 734.4 .661 1.8

_2 50(TP40-49)565.5740.3 .661 1.4• 50(TPX3-39)500.5655.2 .585 1.3• 45(TP47-63)625.1 818.3 .731 1.3NOTE: RPM, TIP SPEED, TIP MACH NO., AND WIND VELOCITY

ARE THE AVERAGED VALUES FOR EACH RUN.

POWER COEFFICIENT CORRECTED FOR WIND VELOCITY

AND DIRECTION.

-I.006

I , 1 I , i I I

.008 .010 .012 .014 .016 .018

ROTOR THRUST COEFFICIENT ._(CT)

I.020

Figure 7.12 Figure of Merit for Baseline ATB

I

.022

64

CR177436

.022

.020

.018

.016

.006

.004

i

o

Figure

NASA-AMES O.A.R.F. TEST 910ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFF

/

SYMBOL RU....NN RP.._.MM0 32 5?0.1

33 570.7

0 36 569.1

"E} 37(TP'S-_0)571.4

50(TP 3-12)561.0

50(,.40-_9)565.5

• 50(TP13-39)500.5

• 45(TP47-63)625.I

VTIP MT!p VWIND

746.2 .663 0.9

747.1 .662 2.5

744.9 .661 2.0

748.0 .663 4.8

734.4 .661 1.8

740.3 .661 1.4

655.2 .585 1.3

818.3 .731 1.3

VELOCITYNOTE: RPM, TIP SPEED, TIP MACH NO., AND WIND

ARE THE AVERAGED VALUES FOR EACH RUN.

7.13

I I I I I4 8 12 16 20

COLLECTIVE PITCH (0.75)

Thrust Coefficient vs. Collective Pitchfor _seline AT

65

I24

CR177436

(Z/El3) IN3131_303 ISfl_Hi _010_

66

CR177436

_P

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH _ASELIN[ ELLIPTICAL TIP

AND TRUNCATED CUFF

l,

v

>-F-

,.JW=m,

NOTE: SOLID SYMBOLS DENOTE DATAFROM ANGLE OF ATTACK PROBES

00

20 -

40 -

6O

8O

:3=Z

_: 100 -O

120 -

140 -

160 --

RADIAL STATION (r/R)

.2 .4 .6 .8 1.0 1.2

i , , ,SIDE _I /

"-- NAOEFLLE

RUN 50 Z/R- 0.4SYM C TP RPM MTI.____PVTI___.PVWIND

0 .0_866 7 561.2 .661 734.6 1.9

0 .01242 9 560.9 .660 734.2 2.2

.01647 11 560.5 .660 733.6 2.2

@ .01859 12 560.1 .659 733.2 2.2

.01917 49 564.5 .660 738.9 2.9

Figure 7.15 Distribution of Downwash Velocities for VariousThrust C0efficlents for Baseline ATB

67

1.4

CR177436

I I I I I I I I I ,. I

[io]Ix]_o_o3 lsn_.L _010_

. cO

o

=.

a

0

0L_

o_ =

• U.

_D

000

0

68

CR177436

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND EXTENDED CUFF

.84 F SYMBOLo _ 56T.gRPM7"3_'.IVTIP MTIP_(EST.)VWIND2.2

00

0

0

0

0

0 0

,56

.52

.48

.44

NOTE: RPM, TIP SPEED, TIP MACH NO., AND WIND VELOCITY#.RETHE AVERAGED VALUES FOR EACH RUN.

POWERCOEFFICIENTCORRECTEDFORWINDVELOCITYAND DIRECTION,

_! I L,, I I I I I

.006 .008 .010 .012 .014 .016 .018 .OZO

ROTOR THRUST COEFFICIENT (CT}

Figure 7.17 Figure of Merit for Baseline ATB with Extended Cuft

69

CR177436

=wL_

L_£k.I.;J

0

I"Ul

I'-

I"

I-4

•020

.018

.016

.014

.012

.010

.008

.006

.004

Figure

NASA-AMES O.A.R.F. TEST 910ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND EXTENDED CUFF

SYMBOl. RUN RPM VTIP MTIPVWIND0 55 563.9 738.1 .663 2.2

(EST. )

NOTE: RPM, TIP SPEED, TIP MACH NO., AND WIND VELOCITYARE THE AVERAGED VALUES FOR EACH RUN.

7.18

l l I I I

4 8 12 16 20

COLLECTIVE PITCH (e.7 5]

Thrust Coefficient vs. Collective Pitchfor Baseline ATB with Extended Cuff

70

CR177436

r_

I ! I I

0 0 0 00 C_ 0 0

I I ! I I ! I I

0 0 0 0 _D 0 0 00 C_ 0 0 0 0 0 0

{Z/_I"3} IN31DI-_303 .LSn_Hi _O.L08

\

,,(

*uJ (W

._o

_. 0.o ,'-" f,_o o

(

m

F,,

0 "-

u+

71

I I I i I I I I I I

0 0 0 0 0 0 0 0 0 0

( ) IN'313[-_-g33 lsn_l I HOIO'L"

72

CR177436

t_m

°i_ -

"_,

° '_

° '_

O

0

0

0

4

CR177436

c

.F

I--I--4

:E:

la-D

laJc_

l.k.

.84

.80

.76

.72

.68

.64

.60

.56

.52

.48

.44

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND NO CUFF

SYMBOL RUN RPM VTIP MTIP VWIND

0 53 566.4 741.4 .661 2.g

54 566.0 740.9 .661 3.2

o

o

NOTE: RUN 53, BLADE LAG ANGLE • i°

RUN 54, BLADE LAG ANGLE - 0°

RPM, TIP SPEED, TIP MACH NO., AND WIND VELOCITY

ARE THE AVERAGED VALUES FOR EACH RUN.

POWER COEFFICIENT CORRECTED FOR WIND VELOCITY

AND DIRECTION.

-I I I I I I I I.006 .008 .010 .012 .014 .016 .018 .020

ROTOR THRUST COEFFICIENT (CT]

Figure 7.21 Figure of Merit for Baseline ATB wlth No Cuff

73

CR177436

I'--

Iml

LLW

U

I"Vl

C_"r-

#,,0I--"

I-4

.020

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND NO CUFF

SYMBOL RUN RPM VTIP MTIP VWIND

0 53 566.4 741.4 .661 2.9

54 566.0 740.9 .661 3.2

.018

.016

.014

.012

.010

.008&

.006

•004

.002

0

Figure

i

7.22

NOTE : RUN 53, BLADE LAG ANGLE - 10

RUN 54, BLADE LAG ANGLE - 0°

RPM, TIP SPEED, TIP MACH NO., AND WINO VELOCITY

ARE THE AVERAGED VALUES FOR EACH RUN.

I I _, I I I

4 8 12 16 20

COLLECTIVE PITCH (O.n75}Thrust Coefficle t vs. Collectivefor Baseline ATB with No Cuff

Pitch

74

CR177436

(Z/_I9) iN3I_I_303 ISA_H± _010_

75

CR177436

set to zero degrees and there is an apparent increase of efficiency at thissetting. The apparent increase in efficiency is however, within the scatter

of earlier testing and is not considered to be significant.

Figure 7.22 indicates that the collective pitch required for a given CT maybe less when the blade sweep/droop is reduced to zero. The control systemflexibility accounts for most of this difference. The recorded values of

collective reflect the control setting at the actuator input. Because ofthe nose down pitching moments associated with blade sweep, the control

blade setting is less in this case by the amount of control flexibility

windup. At a nominal collective input of 16 degrees the difference betweenthe swept and non-swept conditions is estimated to be 0.82 degrees. Figure

7.22 indicates a difference of almost 2 degrees suggesting that some addi-tional mechanism may be involved.

7.5 Performance of ATB with Swept Tip and Extended Cuff

Test results For the ATB with the swept tip are shown in Figures 7.24, 25,26 and 27.

The rate of growth of pitch link loads was almost twice that for the base-

line tip and this restricted testing to a maximum CT of .016 (compared witha CT of 0.020 with the baseline tip at the same RPM). However at this value

of CT the figure of merit (Figure 7.25) is 0.795 and is still trendingupward.

7.6 Performance of ATB with Square TIp and Extended Cuff

Test data with the square tip installed (and extended cuff) is given in Fig-ures 7.28 through 7.31. There is a slight reduction in efficiency through-

out the CT range compared with the baseline tip configuration.

7.7 Configuration Performance Comparisons

Figures 7.32 through 7.42 summarize the comparative performance of the base-line ATB and the XV-15 blades, and of the various ATB alternate

configurations. ,

Figure 7.32 shows that at values of CT above .0125 the power required Forthe ATB becomes progressively less than that for the XV-15 blades. Figure7.33 shows the same information in figure of merit format. Figures 7.34 and

7.35 summarize the comparative behavior of the alternate tip configurationand the XV-15 blades. Figures 7.36 and 7.37 summarize the comparative be-

haviors of the different cuff configurations.

It is seen that the baseline design elliptical tip outperforms the alterna-

tives although the trend For the swept tip suggests that it might be better

at CT values beyond 0.016. The cuff comparisons clearly demonstrate thatextending the trailing edge of the cuff to Form a Full airfoil has a signi-ficant beneficial effect on hover performance.

76

CR177436

(I0) J.N31313_303 ISFI_JHJ._010_

77

CR177436

.84

.80

.76

.72

--- .68E

z .64LLC_

_- .60

.56

.52

.48

.44

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH SWEPT TIP

AND EXTENDED CUFF

SYMBOL RUN RPM VTIP MTIP

0 60 566.1741.0 .664(EST.)

VWIND

1.9

0

NOTE: CT MAX LIMITED BY PITCH LINK LOADS

RPM, TIP SPEED, TIP MACH NO., AND WIND VELOCITY

ARE THE AVERAGED VALUES FOR EACH RUN.

POWER COEFFICIENT CORRECTED FOR WIND VELOCITY

AND DIRECTION.

_I..006

I•008

Figure 7.25

I I I I I.010 .012 .014 .016 .018

ROTOR THRUST COEFFICIENT (CT)

Figure of Merit for ATB with Swept Tip

I.020

78

•020 -

•018 "

.016 -

.014

•006

•004

-4 0

Figure

CR177436

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH SWEPT TIP

AND EXTENDED CUFF

S__YMBO....__L.LRUJ RPM VTI___[PMTI._..[PVWIN__.__.D

0 60 566.1741.0 .664 1.9

(EST.)

NOTE: CT MAX LIMITED BY PITCH LINK LOADS

RPM, TIP SPEED, TIP MACH NO., AND WIND VELOCITY

ARE THE AVERAGED VALUES FOR EACH RUN.

7.26

4 8 12 16 20

COLLECTIVE PITCH (0.75)

Thrust Coefficlent vs. Collective Pitchfor ATB with Swept Tip

24

79

, . .. .

CR177436

/=D ..J J_l i=.

(_ =b --lr-_ •

__ Cl_ m i _ VJ'

;;-_ _ - i o. _..,,_ _ _:_ 8 I <

::"" I-,- 1,,,- / __ _ ..,,.--_ " "' -_ _ -'1_ o"%. L=l _l(_ "_ • / t"_ _-_ ql--

,-* a.W U.,-, / I1 (M_. z w> t,._-- / .. Z .. ,

•..., _ r,j _Ia.l

"_ x _--= ,-, a _ _,,,-_ l"-I_ I '=" _I_ "=" / .III . I

|" o_ 7-1L i t I I n _ I I !

{_/EI_) 1N3131_303 1S_H1 _010_

8O

CR177436

81

CR177436

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH SQUARE TIP

AND EXTENDED CUFF

"84 I SYMBOL RUN RPM VTIP-;-- 6-2---

63

.80

.76

.72

I-..

,.,,,..i,i

_" .64l,,k.

l,.iJp,,,.

I,-4

.60

.56

.52

.48

.44

MTIP VWIND

565.9 740.8 .661 2.3

571.0 747.5 .662 2.1

&

NOTE: RPM, TIP SPEED, TIP MACH _I0., AND WIND VELOCITY

ARE THE AVERAGED VALUES FOR EACH RUN.

POWER COEFFICIENT CORRECTED FOR WIND VELOCITY

ANO DIRECTION.

_I I I i I I I I

.006 .008 .010 .012 .014 .016 .018 .020

ROTOR THRUST COEFFICIENT (CT)

Figure 7.29 Figure of Merit for ATB with Square TIp

82

CR177436

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH SQUARE TIP

AND EXTENDED CUFF

S_.YMBOL RUN RPM VTIP MTIP

0 62 565.9 740.8 .661

63 571.0 747.5 .662

VWIND

.016

.014

.006

.004

.002_;_ NOTE: RPM TIP SPEED, TIP HACH _IO., AND WIND VELOCITY

-4 0 4 8 12 16 20 24

COLLECTIVE PITCH (0.75 )

Figure 7. 30 Thrust Coefficient vs. Collective Pitchfor ATB with S_

83

- .° ....

CR177436

84

CR177436

\

T_

N _ =-=_m

85

CR177436

.84

.80

.76

.72

i .68.64

,,,,,,,

l,,-t

_- .60

.56

.52

.48

BASELINE ATB

__

//

RPM VTIp MTIp VWIND

XV-15 METAL BLADE 585.3 766.2 .690 1.6

BASELINE ELLIPTICAL TIP 568.0 743.5 .662 2.2WITH TRUNCATED CUFF

NOTE: RPM,VTiP, ANDMTIP AREAVERAGEDVALUESFOREACHCONFIGURATION

CORRECTEDFORWINDVELOCITYANDDIRECTION

.44 -I I I I I, I _ I.006 .008 .010 .012 .014 .016 .018 ,020

ROTOR THRUST COEFFICIENT CCT_

Figure 7.33 Comparison of Figure of Merit for XV-15 Metal Bladeand Baseline ATB as Measured on OARF

86

CR177436

\

!

x

(J'3).LN3131_3303 ISn_Hl _010_

87

CR177436 -;

.84

.80

.76

_-_.72

_ .68

,."r' .64

.60

.56

.52

,48

m

ELLIPTICAL

XV-15

_ \ ....--.-_- _.

_ SQUARE

-L I•006 .008

Figure 7.35

RPM VTIP MTIp VWIND

XV-15 METAL BLADE (SOUARE TIP) 585.3 766.2 .690 1.6

BASELINE ELLIPTICAL TIP WITH EXTENDED CUFF 563.9 738.1 (_63) 2.2

SQUARE TIP WITH EXTENDED CUFF 568.5 744.2 .662 2.5

SWEPT TIP WITH EXTENDED CUFF 566.1 741.0 .664 1.9(EST.)

NOTE: RPI4, VTI P, AND ;4TIP ARE AVERAGED VALUES

FOR EACH CONF_.GURATION

CORRECTEB FOR :_IND VELOCITY AND OIRECTION

I I ,, I I I I.010 •012 .014 .016 •018 .020

ROTOR THR_T COEFFICIENT (CT}

Effect of Tip Shape on Figure of Merit for ATB withBaseline Elliptical, Swept, and Square Tips vs.XV-15 Metal Blade

88

cR177436

I I I I

!

x

! I t ] ! ! obJ

m

IL

89

CR177436 :_

L/.

_'FEXTENDED TRUNCATED

I _.._ k_._.80 I'- XV-15 i// _ • _..

J _-'+1- / .-;;" ..--"\ \------'--NOCUFF

.72

p, •64L,J

/.60- I

I

.56 -

.52 -

.48 -

•44 -|

.006

Figure 7.37

RPM VTIP MTIP VWIND

XV-15 METAL BLADE 585.3 766.3 .690 1.6

----TRUNCATED (BASELINE)568.0 743.5 .662 2.2

EXTENDED 563.9 738.1 .663 2.2(EST.)

NO CUFF 566.2 741.2 .661 3.1

I•008

NOTE: RPM, VTIp, AND MTI P ARE AVERAGED VALUES

FOR EACH CONFIGURATION

CORRECTED FOR WIND VELOCITY AND DIRECTION

I I I I I I

-•010 •012 .014 .016 .018 .020

ROTOR THRUST COEFFICIENT CCT}Effect of Cuff on Figure of Merit for Baseline ATB withElliptical Tip - Comparison of Truncated (Baseline),Extended, and No Cuff vs. XV-15 Metal Blade

90

CR177436

.0014 -

.0013 --

.0012 --zw

0011 --0 •{..}

= .0010 --o

<D

.0009 -

.0008 --

I720

I555

,I

.650

Figure 7.38

SYMBOL RUN

0 55 2.2

s6 3.2

/T 56 3.6

[] 60 1.9

_7 61 4.0

BASELINE ELLIPTICAL TIP WITH EXTENDED CUFF

v / v" J v"

SWEPT TIP WITH EXTENDED CUFF

CT - .012 (CROSS-PLOT FROM CT vs. Cp)

SWEPT TIP

/ELLIPTICAL TIP

•NOTE: VTIP, MTIP, VWIND, AND RPMARE AVERAGED VALUES FOR EACH RUN

I I I I I I

740 760 780 VTI P 800 820 840

I ! I I I I I I565 575 585 595 605 615 625 635

I I RPM I I I

.670 .690 .710 .730 .750MTIP

Effect of Tip Shape on Cp vs. VTI P, RPM, and MTI P-Comparison of Baseline Elliptical Tip and Swept Tip

gl

CR177436

Figure

p-. ¢.j

)--Z

LLILLI.Ll

(.I

l--t_

C¢:

p--

L-4

7.39

.020

.018

.016

.014

.012

.010

.008

ELLIPTICAL TIPSQUARE TIP

!PT TIP

.006

.004

Effect of TipBaseline

I4

RPM VTI_.._PPMTI.__PVWIND

;ELINEELLIPTICAL TIP 563.9 738.1 .663 2.2

WITH EXTENDED CUFF (EST.)

SOUARE TIP WITH_._568.5 744.2 .662 2.5EXTENDED CUFF

SWEPT TIP WITH 566.1 741.0 .664 1.9EXTENDED CUFF (EST.}

NOTE: RPM,VTIp,MTIP ANDVWIND AREAVERAGEDVALUESFOREACHCONFIGURATION.

VALUESLISTEDFORSOUARETI° ARE AVERAGES

I OF RUNS621AND63. I ! I8 12 16 20 24

COLLECTIVE PITCH (e.75}

Shape on C T vs. Collective- Comparison ofATB Elliptical Tip, Swept and Square Tips

92

CR177436

I I Ic_J C_l

$3

\

d3)

I'--

(._LL

I'-- ¢'-,,)

__.lhl

LdZ

,..,IL_J "r

,_ o--.+

I I I I

• e D •

' _O.L3V.-IXON3IOIJJ3 (133R(]NI

NC_

O@O0

0

@

I '-

_g0

"(,9

o _®

,,,=

"@l--u. c

,., ,(_ ww• I-'-

,,.,.= _r,,F- 0'1.-

_ ,,- o

0.._

N

0

L_

m

t_

93

CR177436

I

Pq

\\

\

! I

,,.4 ,.4Z/_;'L3

z_/,.(°d3 - do)

%

\

\ \

0

\\\\1/

I ! I I

- _! '_IO.L3V.-IA3_31_I_3 (]33naNI

¢,j

• b

0oO Z

- ,-, 1.6"0

CX

o,,,o .

i

q N-e. o

,9- =®_ o'_

"°1o. _ ,=

_ ,.oN o_ =,.

-- 0

• tm

m

U.

c_0

g4

CR177436

0

20

40

v= 60

120 -

14°I160

,30

=

I00

RADIAL STATION (r/R)

0 .2 .4 .6 .8 1.0 1.2 1.4

- I SIDE

- /XV-15 METAL BLADE

CT- .00796

j XV-15 METAL BLADECT- .01246

BASELINE ATBELLIPTICAL TIP AND

CT .00866

BASELINE ATB WITH ELLIPTICAL TIP

AND TRUNCATED CUFF CT- .01242

RUN TP CT RPM MTIP VTIp VWIND

•----.-- XV-15 METAL BLADE 25 16 .00--7-96585.9 .690 771.5 2.9

25 22 .01246 585.5 .689 770.9 3.3

R-- ATB ROTOR 50 7 .00866 561.2 .661 734.6 1.9

50 9 .01242 560.9 .660 734.2 2.2

Figure 7.42 Comparison of Downwash Distributions forXV-15 Metal Blade and Baseline ATB Rotor

95

CR177436

In Figure 7.38 the effect of tip shape on power required is shown as a Func-tion of tip speed, RPM, and tip Mach number. A comparison is made betweenthe swept and the baseline elliptical planforms. The extended cuff config-uration was used for this crossplot. At the normal operating RPM (nominally

565), and at maximum RPM (625), the elliptical tip maintains a slight advan-

tage at CT = .012.

Figure 7.39 compares the baseline elliptical, swept, and square tips on a CTvs. collective basis. As in Figure 7.40, the extended cuff was used for this

plot. As expected, the baseline elliptical tip has slightly better perfor-mance than the other tip configurations.

Figures 7.40 and 7.41 present the effect of tip shape and cuff configurationon induced efficiency as a function of rotor thrust coefficient.

Figure 7.42 compares downwash distributions for the metal blade and thebaseline ATB at nearly similar thrust coefficients.

The photograph In Figure 7.43 show tip vortices for the elliptical tip withthe extended cuff configuration. (Compare with Figure 7.8).

7.8 Theory-Test Comparison

The predictions for the XV-IS metal blades and for the Advanced Technology

Blades are compared with measured performance in Figure 1.4. The figure of

merit is generally underestimated at high values of _ The predicted per-formance of the XV-15 metal blades and the baseline blades was calculated

using a current lifting-line/blade element program. The program, which cor-relates well with low twist helicopter blades, appears to overestimate the

induced power for highly twisted propellers/rotors. Possible reasons forthe discrepancies are discussed in the following paragraphs.

7.8.1 Rotor Wake Model

Because over 70 percent of the power absorbed by a hovering rotor is wake-

induced, the successful prediction of performance depends on how well theeffects of the vortex wake are modelled. An accurate wake model is One in

which the strength and positions of the vortices forming the wake are cor-rectly represented. The current wake model is semi-empirlcal and contains

many correction factors determined by correlation of the analysis withmeasured helicopter rotor and prop-rotor performance. While the modelyields practical results in cases where the rotor geometry and operatingconditions are within the range of the empirical factors, extensions to

configurations outside the data base are less reliable. This semi-empiricalwake representation is outdated and is currently being replaced with a

modern wake representation based upon experimentally observed rotor wakestructures following Landgrebe, Kocurek, and Gray.

A comparison of the effect of the different wake representations on thecalculation of figure of merit for the ATB rotor is presented in Figure7.44. With the current wake, the peak figure of merit level is underpre-

dicted and performance at CT values greater than O.OOg is also underpre-dicted. When the empirical wake is replaced by the Kocurek wake model,

96

CR177436

Figure 7.43

ORIGINAI-_ PAGE Ig

DE POOR QUALITY

Tip Vortices of Baseline ATB with Extended Cuff

97

CR177436

.84 --BLADE ELEMENT/MOMENTUM THEORY

\_--

//

ATB BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFF

TEST DATA FAIRING

/

_! IMPROVED

ANALYSIS

%%

CURRENTANALYSIS

\

I I I I _ , I.010 .012 .014 .016 i-08 .020

ROTOR THRUST COEFFICIENT (CT)

Figure 7.44 Comparison of Calculated and Measured Performanceon the Advanced Technology Blade Rotor

98

CR177436

prediction of peak figure of merit is improved. Note that neither modulepredicts the high figures of merit at high thrust coefficients. Also shownare the results of a blade element/momentum analysis (using the same airfoil

data) which somewhat overpredicts the performance but yields a better over-

all shape for the curve.

Although there has been much work on experimentally observed rotor wakestructures for helicopter rotors, there have been no tests conducted spe-

cifically for highly twisted rotors or propellers. Some progress in this

direction was made during this test as described in sections 7.1 and 7.7.Further detailed experiments using pressure instrumented blades and lasersto measure the detailed wake structure should be conducted for representa-tive rotors. These results can be mathematically modelled and incorporated

into suitable analysis techniques.

While the introduction of more realistic wake models may improve the pre-

diction techniques, other areas also require attention.

7.8.2 Airfoil Behavior at Hiqh Angles of Attack

Highly twisted tilt rotor blades operate in hover with the root sections ator beyond stall angles of attack. Two-dimensional airfoil data obtainedfrom the wind tunnel usually does not define the post-stall lift, drag, and

pitching moment behavior because testing is rarely conducted beyond stall.It has been shown by the OARF results and elsewhere that the root area can

influence rotor performance significantly. Additional test data both atmodel and full scale is therefore required on representative airfoils at

high angles of attack to establish the basic shape of the post-stall be-havior for the root section.

7.8.3 Spanwise Flow Effects

There is evidence that two-dimensional airfoil data is not entirely appli-

cable near the root of a rotating blade. Figure 7.45 from Reference 7

shows that propeller blade sections near the root appear to have extendedlift-curves and higher lift-curve slopes than those obtained from two-dimensional wind tunnel tests. The mechanism suggested for this improved

performance is boundary layer thinning arising from centrifugal spanwisepumping and the effect of Coriolis forces on the boundary layer. Develop-ment of a method to account for this effect is recommended and could be

conducted as part of the previously recommended experiments on pressureinstrumented blades.

8.0 ROTOR AND CONTROL SYSTEM LOADS

Rotor and control loads were continuously monitored throughout the hover

testing of the XV-15 Metal Blade and Advanced TeChnology Blades to ensurethat static and dynamic limits were not exceeded. In general, testing was

limited by steady spindle flap bending at the extremes of the thrust range.

While oscillatory loads did not limit performance testing, it was necessary

99

CR177436

RADIALSECTION

r/R= 0.8 _,

0.7(

0.6(o.5(0.4i(Z

"3--4 B._

.-q

---WIND TUNNEL

-15 -I0 -5 0 s 10 15

ANGLEOF ATTACK(_)(DEG.)

20 25 30

Figure 7.45 Local Lift Coefficients, C_, at Various Radial Sectionson a Rotating Propeller (Reference 7)

lO0

CR177436

to use cyclic to control flapping and flap-induced loads to acceptable lev-els. Even with cyclic control, transient bending loads frequently exceededthe endurance limit of the hub yoke spindle necessitating frequent damage

counts to determine the percentage of spindle life used. (Total life used

did not exceed 5 per cent.)

The sign convention used in reporting the measured loads is as follows:

+ Flap Bending -- Compression In the blade upper surface

+ Chord Bending -- Compression in trailing edge

+ Pitch Link Load -- Consistent with blade torsion leading edge up

It should be noted that, in any given run, substantial scatter is present in

the loads data because of cyclic adjustments and variations in wind direc-

tion and magnitude. However, although not ideal for correlation studies,the data indicate general trends.

8.1 XV-15 Metal Blade

The rotor system, blades, hub, and controls were essentially the same as

previously tested on the Aero Propulsion Laboratory Whirl Stand at Wright-Patterson Air Force Base and documented in Bell Helicopter Report No.

300-099-010 (CR 114626), Reference 4.

Figures 8.1 and 8.2 present a summary of the measured steady and oscillating

yoke spindle and pitch link loads as a function of CT. Comparing the datafrom the previous testing at Wright-Patterson with the current Ames testing

does not indicate any significant differences in the measured loadings.

The upper end of the thrust range was generally limited by blade stall asevidenced by an increase in rotor noise, a rapid increase in oscillatory

pitch link loads, and difficulty in controlling gimbal angle. In the pre-liminary run-ups, operation was limited by oscillatory loads in the hub yoke

spindle. Bending moment allowables initially imposed on the spindle for

this test, (± 20,000 in.lb, as compared to ± 58,000 in.lb, in the previouswhirl test) were based on later knowledge of the endurance limits for thetitanium material and were routinely exceeded during spln-up and shut-down.This limitation was overcome by utilizing an S-N curve for the spindle toallow short time exceedance of the ± 20,000 in.lb, endurance limit. A run-

ning damage count was maintained to ensure safety of operation.

8.2 Baseline Advanced Technoloqy Blade (ATB)

Figures 8.3 through 8.22 present a summary of the measured steady and oscil-lating yoke spindle, pitch link, and flap bending loads as a function of CTand RPM.

I01

NASA-AMES O.A.R.F. TEST 910

XV-15 METAL BLADE

CR177436

140,000

RUN22

5

23

MAXIMUM--0--

ALTERNATING

-_

120,000

_ 80,000

I;i 60,000

.,-,,.

_ 40,000

¢m

20,000

0

MAXIMUM

(STEADY + ALTERNATING)

- o_°ALTERNATING

0 .004

Figure 8.1

.oo8- .o12 .o16 .o2o

ROTOR THRUST COEFFICIENT (CT)

XV-15 Metal Blade: Hub Spmd_ Resultant FlapBending Moments vs. C T

.024_

102

CR177436

100

-200

-300

-400

NASA-AMES O.A.R.F. TEST 910

XV-15 METAL BLADE

ROTOR THRUST COEFFICIENT (CT)

.008 .012 .016 .020 .Oz,

0 _kLTERNATING

A mO

° o

<>__0 MAXIMUM

(STEADY ÷ ALTERNATING)

%

L_

Figure 8.2 XV-15 Metal Blade: Pitch Link Loads vs. C T

I03

CR177436

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFF

140,000 -

120,000

.-. q

_'_'l.O0,000

_60,000

W

40,000

RU.__NNMAXIMUM ALTERNATING

32 _ --_--

36 _ --0"--

MAXIMUM

(STEADY + ALTERNATING),,_

A

0

2o,ooo- /_ ,,_-__

00 .004 .008 .012 .016 .020 .024

ROTOR THRUST COEFFICIENT (CT)

Figure 8.3 Baseline ATB: Hub Spindle Resultant Flap BendingMoments vs. C T (Runs 32 and 36)

I

104

CR177436

160,000

140,000

..-.120,000

_J

g

I00,000

'_ 80,000

_=6o,oooIJ.I

I.iJ.-I

40,000

20,000

L__

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE ELLIPTICAL TIP

- AND TRUNCATED CUFF ?/

RUN MAXIMUM ALTERNATIN_ /50b "_ -

m

m

m

MAXIMUM /

m

- /

ALTERNATING I

.004 .008 .012 .016 .020 .024

ROTOR THRUST COEFFICIENT (CT}

Figure 8.4 Baseline ATB: Hub Spindle Resultant Flap BendingMoments vs. C T (Run 50b)

105

CR177436

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE ELLIPTICAL TIPAND TRUNCATED CUFF

120,000

ee_..I

80,000

v')l,--z_JX0

40,0oo

z

er_

e_

_ o

p-

..J

_-40,000

.004 .012 .016

ROTOR THRUST COEFFICIENT (CT)

.020 •024

Figure 8.5 Baseline ATB: Hub Spindle Steady Flap Bending Moments vs. C T- Effect of RPM

106

CR177436

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE ELLIPTICAL TIPAND TRUNCATED CUFF

SYM RPM VTI__._P

500 654

•_. 571 747

--Z_--. 626 819

.004 .008 .012 .016 .020 .0240

0

ROTOR TBRUST COEFFICIENT (CT)

Figure 8.6 Baseline ATB: Hub Spindle Alternating FlapBending Moments vs. C T -Effect of RPM

107

• , . . . ,,

CR177436

n__J

z

v1I.-

i.iJ

_=Zi,ie_

-'1-(J

>-r_,,¢laJPu'}

-J

Z

0.V1

-r"

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFF

SY..__HRPM VTIp

"E}-- 500 654

•"0-" 571 747

--A-- 626 819

80,000 -

0,000

40,000

20,000

• .,_/

.024

-20,000 k ROTOR THRUST COEFFICIENT (CT)

Figure 8.7 Baseline ATB: Hub Spindle Steady Chord-Bending Moments vs. CT-Effect of RPM

I08

CR177436

30,000

25,000

Z

v

2o,ooo0X

rv,

_=i5,ooo(..}

_i,I0,000

Z

5,000

0

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFF

SYM RPM VTI.____P

500 654

•_. 57! 747

-A--626 8i9 Io/ /

• 004

Figure 8.8

.008 .012 .016 .020 .024

ROTOR THRUST COEFFICIENT (C-)Baseline ATB: Hub Spind_ Al_ernat_g Chord Bending

Moments_ vs. C T- Effect of_RPM

I09

CR177436

NASA-AMES O.A.R.F. TEST 910ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFF

I CT - .010 1

20,000 F

15,000 L

i /ii0,000 I

I I I300 400 500

ROTOR SPEED (RPM)

I60C

Figure 8.9 Baseline ATB: Hub Spindle Steady Flap Bending Moment vs. RPM

II0

CR177436

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFF

20,000

z 15,000

.=.,

,-.,10,000

._:_,.,5,00_0

o:

-r

I CT - .0101

z,,

A

,e,

o_ I I I300 400 500 600

ROTOR SPEED (RPM)

Figure 8.10 Baseline ATB: Hub Spindle Alternating Flap Bending Moments vs. RPM

Ill

•. ,. •

CR177436

NASA-AMES O.A.R.F. TEST 910

ATB WITH BASELINE ELLIPTICAL TIPAND TRUNCATED CUFF

30,000 I

,,.,25,000

20,000

oLI

300

A /x

I I I400 500 600

ROTOR SPEED (RPM)

Flgure 8.11 Base,m) ATB: Hub Spindle Steady Chord Bend|ng Moments vs. RPM

ll2

CR177436

20,000

15,000

g

,.=,

1o,ooo(..}

C.gZ ""

Zew,

,=C

.._ 5,000

Z

O-

-r

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFF

. ICT-.°12]

A A A&

O m

I , ,, I I300 400 500

ROTOR SPEED (RPM)

Figure 8.12 Baseline ATB: Hub Spindle Alternating ChordBending Moments vs. RPM

t600

ll3

CR177436

-I00

-200

_J

,,-,,

"_ -300,J

J-4

..J

-r-

p-

-400

-500

-600

0

NASA-_ESO.A.R.F.TEST910ATB ROTORWITHBASELINEELLIPTICALTIPANDTRUNCATEDCUFF RUN MAXIMUM ALTERNATING

32 _ --_--

36 --_ --_

-700

ROTOR THRUST COEFFICIENT (_)

.004 .008 .012 .016

"1 I I l

ALTERNATING

m

.020

I

.02_

I

MAXIMUM

(STEADY + ALTERNATING)

1

Figure 8.13 Baseline ATB: Pitch Link Loads vs. C T (Runs 32 and 36)

114

CR177436

-I00

-200

,_J

-300

.,_I

,._ -400I--

-500

-600

-700

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFF

SYM RUN

50b

ROTOR THRUST COEFFICIENT (CT)

.004 .008 .012 .016

I i I I

.020

I

MAXIMUM(STEADY + ALTERNATING)

f

C) 0 C)

Figure 8.14 Baseline ATB: Pitch Link Loads vs. C T (Run 5Oh)

,024

I

ll5

CR177436

-100

-2oo,,,,.,,,

,,..i

z

-300

!..-

,,,-,

-4oo(,t3

-500

-600 -

-700 -

NASA-AMES O.A.R.F. TEST 910ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFF

SYM RPM VTI_._..PP

500 654

•,-(_),-.571 747

--_-- 626 819

•004ROTOR THRUST COEFFICIENT (CT)

.008 .012 .016 .020

\

\

\

\

\

[]

Figure 8.15 Baseline ATB: Steady Pitch Link Loads vs. C T -Effect of RPM

.024

I16

CR177436

..Jv

--J

-,#Zl--q_J

I--

:Z

1-'-

Z

i,i

,..I

200 -

150 --

100 --

50

(

0

NASA-AMES O.A,R.F. TEST g10

ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFF

SYM RPM VTIE

-'E}'-- 500 654

•--Q-. 571 747

-A-- 626 819

x,,

C)

I, I I I I•004 ,008 .012 .016 .020

ROTOR THRUST COEFFICIENT (CT)

Figure 8.16 Baseline ATB: Alternating Pitch Link Loads vs. C T-Effect of RPM

ll7

I.024

CR177436

NASA-AMES O.A.R.F. TEST 910

A?B ROTOR WITH BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFF

_oo- lcT . .OlOI

ii m

-loo..,.I

.J

P- -200

-300

-400 - I I I300 400 500 600

ROTOR SPEED (RPM)

Figure 8.17 Baseline ATB: Steady Pitch Link Loads vs. RPM

118

CR177436

200

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE ELLIPTICAL TIPAND TRUNCATED CUFF

I CT - .010 1

A

0I I I I

300 400 500 600

ROTOR SPEED (RPM)

Figure 8.18 Basellne ATB: Alternatlng Pitch Link Loads vs. RPM

ll9

CR177436

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE ELLIPTICAL TIPAND TRUNCATED CUFF

-160,000

120,000

RU..NNMAXIMUH ALTERNATING

.004

MAX IMUM

(STEADY + ALTERNATING)

"x /x

ALTERNATING

08 .012 .016

ROTOR THRUST COEFFICIENT (CT)

.020

-80,000

Figure 8.19 Basellne ATB: Flap Bendlng Moments at 0.10R vs. C T

120

•024

CR17743._..__6

160,000 --

120,000 --

-80,000 --

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFF

RUN MAXIMUM ALTERNATING

SOb _ -C_-

MAXIMUM

(STEADY + ALTERNATING

•004

ALTERNATING

-_ __c_/_c___ - _ --_ --_

008 .012 .016 .020

ROTOR THRUST COEFFICIENT (CT)

Figure 8.20 Baseline ATB: Flap Bending Moments at 0.10R vs. C T

.02_

121

CR177436

w

.-I

Z

CO

I--

u')I--Z

Z:

_JC_

>.-

,:CL_JI-"(.n

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR WITH BASELINE _LLIPTICAL TIPAND TRUNCATED CUFF

SYMBOL RPM VTIp

564 738

•_. 592 775--Am 626 819

2000 I

1000

0

-I000

-2000

-3000

•004 .008 _.012 .016 .020

ROTOR THRUST COEFFICIENT (CT}/

/

Figure 8.21 Baseline ATB: Steady Flap Bending Moments at 0.85R vs. C T -Effect of RPM

i i

122

, . . , .

CR1774361

g

C_

NASA-AMES O.A.R.F. TEST 910

ATB ROTOR RITFI BASELINE ELLIPTICAL TIP

AND TRUNCATED CUFF

2000 -

SYMBOL RPM VTIP

•--0--" 564 738

•.-_,-. 592 775

-A-- 626 819

.,:..

"-" 1600

CO

C)

I'--

1200

Z

,.,z,._800<

.JI.I.

<.'3

=:. 400

I'--.,.I

-e

z_ ,A

I I I I i0 i0 .004 .008 .012 .016 .020

ROTOR THRUST COEFFICIENT CCT)

Figure 8.22 Baseline ATB: Alternating Flap Bending Moments at 0.85R vs. C T -Effect of RPM

123

CR177436i

8OO

mm

600

Gele-_

O

.,...

_= 400.,J

:E

I,--

N 200

I,--

THRUST BEARING LIMITII_*l*lnt#Zlll/ll/#/##ll_ll IIIIIIIIIIIIlllllllnlllllllllllllllllll//lll/lllt/I

MONITORING LIMIT-_LLL_ 44#,.U._.U.CJ.4J._ .U.¢4.4._ Hill''" , s _ / I t , _ _ / / H , t / , i / / t / _ / _l , / / / /

/ / BASELINE "

m ./ / ELLIPTICAL TIP

/" / ..... 765 RPM .

SWEPT TIP / _ BLADE LAG ANGLE-1 u

// / / ELLIPTICALTIP// / / 565 RPM

0 .004 .008 i012 .016 .020 .024

ROTOR THRUST COEFFICIENT ..(CT)

Figure 8.23 Effect of Blade Sweep (Lag Angle) and Tip Shapeon Steady Pitch Link Loads

124

CR177436

A comparison of the ATB loads with the XV-15 Metal Blade loads indicatesthat load trends are similar, with the greatest difference being in the os-

cillatory loads. This is to be expected since oscillatory loads are depen-dent on wind conditions and cyclic control adjustments, and it was evidentthat the the oscillatory loads were affected by these.

In general, the ATB blades were tested to significantly higher thrust

coefficients (CT) than the metal blade without encountering any signs ofinstability, flutter, or excessive loads. As expected the steady spindlebeam moments and steady pitch link loads were proportionately higher than

the metal blade for the same CT due to the increased chord of the ATB.

Load trends with RPM followed expected patterns. For the same CT,

increasing the RPM increases blade thrust and blade loads.

8.3 Alternate Configurations: Advanced Technology Blade

Two alternate configurations had a notable effect on blade loads. Thesewere the 0.0 degrees blade sweep (the baseline has 1 degree sweep) and the

swept tip configurations. As expected there was an effect on steady pitchlink loads as shown in Figure 8.23. As noted in sections 7.5 and 7.7, the

swept tip performance data was following a trend which suggested that peak

performance might be better than the baseline ATB when pitch link loadsrestricted the test before peak figure of merit was reached.

9.0 ACOUSTICS

Near-field and Far-field noise levels were measured during the hover test

program. The near-field microphone location represented a point on thefuselage side of a typical tilt rotor in hover. The microphone locationsimulates a point on a fuselage 8 ft aft (2.4 meters) and with 2 ft (O.6m)radial clearance from the tip path plane. Far-field noise was recorded with

an array of microphones at 250 ft and 650 ft radius, at O, 15, 30 and 45

degrees behind the rotor disc.

Figure 9.1 shows comparative XV-15 metal blade and ATB overall near-fieldsound pressure levels as a function of rotor thrust. The ATB noise level is

approximately 2-3 dB lower than the metal blades over the normal operating

range.

Far-field noise data for the 15 degrees aft location is shown in Figure 9.2.

The ATB OASPL is approximately 5-6 dB lower than that from the XV-15 metalblades.

These comparative trends are expected since the tip pressure loading is lessfor the tapered, higher solidity ATB with its more even thrust distribution

over the span.

Table 9.1 summarizes runs during which acoustics data was acquired during

the OARF testing in 1984. This includes JVX isolated rotor runs as well asXV-15 metal blades and ATB test runs. The acquisition and analysis of theacoustics data from these tests was an Ames Research Center activity. Addi-

tional information may be obtained from Ames personnel (M.D. Maisel).

125

CR177436

_" 132 -

41_ 2 FT

MICROPHONE

128 -XV-15 METAL BLADES

MTI P = 0.69>__ 124 -

i_ _

120 -- i/ ,

116 _

_112 r """

d-io8_--0 I l,r, 1

0 2,000 4,000 6,000

\ADVANCED TECHNOLOGY BLADES

MTIP = 0.69

I I I8,000 I0,000 12,000

ROTOR THRUST (LB)

I14,000

Figure 9.1 Typical Near-Field Noise Data for XV-1 5 Metal Blade and ATB

C¢3

"" i00 ---JL_J>

"' 96--,..,.I

= 92

IX

" 88

Z

° 84

--J

u. 0

.

/ 2-.50$'r MICROPHONE

,/

m

_ _ ,=,.= m =,=., J,.=,

m

XV-15 METAL BLADES _"

MTIP = 0.69 "_-__._._'_" _ .

_ ADVANCED TECHNOLOGY BLADES

MTI P = 0.69

I I I I ..... ! ..... ! !2 ,000 4,000 6 ,000 8 ,000 10,000 12,000 14 ,DO0

ROTOR THRUST (LB)

Figure 9.2 Typical Far-Field Noise Data for XV-15 Metal Blade and ATB

126

127

CR177436

10.0 CONCLUSIONS AND RECOMMENDATIONS

10.1 Conclusions

Full scale rotor hover performance was obtained for the basic XV-15 rotorand the Advanced Technology Blade (ATB). The following conclusions can be

drawn from the results:

Io Accuracy and reliability of the performance data is very high asshown by the low level of scatter in the data and repeatabilityof data taken at different times. (See for example, Figure 7.11).

. Both the ATB and the XV-15 rotors performed at levels signifi-

cantly better than anticipated from theoretical estimates.Measured peak values of figure of merit were in the range 0.79 to0.81 whereas predicted values did not exceed 0.79. Peak perfor-mance occurred at higher values of CT and did not drop off as

quickly as predicted (Figure 1.4).

. The performance of the baseline XV-15 rotor is higher than thatmeasured during a previous test of the same blades on the WPAFBwhirl tower when corrected for tower blockage effects (Figure 7.2).

o

.

Of the three tip shapes tested on the ATB blades, the elliptical

tip outperformed a rectangular tip and a swept tip (Figure 7.35).The swept tip and elliptical tips had the same solidity; thesolidity of the rectangular tip was slightly higher. However,

testing of the swept tip was curtailed but did indicate that its

performance might match or exceed that of the elliptical tip at

high thrust.

Testing of the ATB with no cuff, truncated cuff, and full cuffshowed that performance is improved as more blade area is added to

-the cuff region (Figure 7.37).

. A value of CT/OT = .22 was reached with the ATB before loadslimited further testing (Figure 7.11). The corresponding valuefor the XV-15 was .IB, (Figure 7.1).

a

.

Comparison of the ATB and XV-15 blade loads indicates that loadtrends are similar. The ATB blades were tested to significantly

higher thrust levels than the XV-15 without encountering anyinstability, flutter, or excessive loads.

Acoustical measurements show (Figures 9.1 and 9.2) that the ATB

with the elliptical tips was 2-3 dB lower than the XV-15 metalblades in the near-field and 5-6 dB lower in the far-field.

128

CR177436

10.2 Recommendations

A program of research should be initiated aimed at developingbetter wake models for use in tilt rotor hover performance anal-

yses. The program would consist of experimentally determining thewake vortex structure for different rotor operating conditions

(tip speed, collective) for representative blade planforms, twistdistributions, and number of blades. It would also be desirable

to make these measurements with blades having pressure instrumen-tation so that the blade circulation distribution can be deter-

mined. The combination of blade circulation distribution and wake

geometry can then be used to derive wake models for use in hover

performance analyses.

eThere is a need to acquire a better understanding of behavior of

the thick root sections used on tilt rotor blades especially near

and beyond stall. A program of wind tunnel test and analysis todefine the stall and post-stall behavior should be initiated.This would include two-dimensional testing as well as measurements

on the root sections of rotor blades.

1.

.

.

e

.

.

.

8.

11.0 REFERENCES

McVeigh, M. A. and Bartie, K., "V-22 Large-Scale Rotor Hover Perfor-mance and Wing Download Test", Bell Boeing Document D901-99140-1,

March 1985.

Benson, R. G., Ekquist, D., et al, "XV-15/Advanced Technology BladeHover Test Plan for NASA-Ames OARF", Boeing Document D210-12262-I,

December 1983.

Silcox, H. F., et al, "System Safety Analysis for Advanced Technology

Blades (ATB) on NASA-Ames Test Rig", Boeing Document D210-12256,

Volumes 1-5, January 1984.

Hell, S., et al, "Full-Scale Hover Test of a 25-Foot Tilt Rotor", BHT

Report 300-099-010, NASA CR114626, May 1973.

Devlin, J. F., "Instrumentation Requirements for Advanced TechnologyBlade Rotor Test at NASA-Ames OARF", Bell Boeing Document D901-gg025-2,

May 1984.

Walton, I., "Data Reduction Requirements for JVX/XV-15 Large-ScaleRotor Test at NASA-Ames OARF", Bell Boeing Document D901-gg050-3,

October 1983.

Schlichting, H., "Boundary Layer Theory", McGraw-Hill, NY, 1968.

Hoerner, S. Fo, "Fluid Dynamic Drag", published by Hoerner Fluid

Dynamics, 1965.

129

CR177436

APPENDIX A - CORRECTIONS FOR EFFECTS OF WIND

A.1 Introduction

The XV-15 and ATB rotor diameters are 25 Feet. The ATB rotor has a thrust-

weighted solidity of 0.10, the XV-15 rotor has a solidity of 0.089. The ro-tor data was generally acquired in conditions where the ambient windvelocity was not zero. A correction for the effect of wind is therefore

required to arrive at true hover performance. The method used to correctfor the effect of wind on hover performance is presented below.

A.2 Correction for Wind Effects

In a wind of speed V at an angle o to the rotor shaft, the rotor develops athrust T and a normal force NF. The rotor power is

Cp = CPPRo + p CT cos_ - p CNF sina + k _i CT(1)

where _ = V/VT, _i = vi/VT, v_ is the mean induced velocity, and k is acorrection factor for the ideal induced velocity.

During the test the rotor was trimmed to zero flapping. The rotor balancemeasured T and NF. Wind speed (V) and direction (o) were measured by an

anemometer mounted at a height above the ground. Using Hoerner's recom-mended model for the wind boundary layer, (Reference 8), it was determinedthat if the anemometer were positioned at the same height as the hub, the

mean wind speed would be read. The effect of the wind can be calculated as

follows.

The power required to hover in zero wind conditions is:

= + kH " CTCPH CppR0 vlH

Since CPpR0 does not vary significantly with small changes in ambient wind

conditions, we may substitute for CPPRofrOm Equation (1).

The adjusted power for hover in zero wind may then be written:

CPH= Cp p (CT cos_ - CNF sine) + kHViH I kH-ViH CT (2)

130

CR177436

The ratio v, = _i / _iH is obtained by solving

v,4 + 2 v, 3 V, cosa SIGN CT + v,2 V,z - i = 0 (3)

where V, = V/ViH

Equation (3) is solved for V, by iteration using the Newton-Raphson method

V,n+1 = V,n - (F/F')n

where F = v,4 + 2v,3 V, cosa SIGN CT + v, 2 V,2 - i

F' = 4v, _ + 6v, 2 V, cosa SIGN CT + 2v, V, 2 (4)

and a starting value is given by an approximation developed by Wayne Johnson.vlz

v, = i - Ucosalq_Tl = 1 - 0.5V, cosa

Figure A.1 shows the induced velocity ratio (and ideal power ratio) for dif-ferent wind speeds and directions for CT = .015. The effect of not applyinga wind correction is shown in Figure A.2 where true hover performance is

compared to that which would be calculated from measurements made in a 3knot wind. The effect of the wind is substantial, amounting to approximate-

ly 2 points in figure of merit when the flow Is axial.

On the basis of the above analysis the method for wind correction is:

i)

2)

Calculate v, using the full iterative quartic solution.

Calculate the hover power using equation (2). From estimated hover

performance the value of kH is 1.16. It is assumed that k = kHsince only very low advance ratios are involved.

131

CR177436

A.3 Data Reduction and Correction Procedure

The following are the steps in the data reduction and correction procedure.

I. Record the main balance thrust (axial force), rolling moment

(friction torque), normal force and shaft torque. Record meanwind speed (V) and direction (a).

2. Subtract the friction torque from the shaft torque to yield a true

rotor torque.

3. Put data in coefficient form, CT, Cp, etc.

4. Calculate the following:

(a) V. - VI ( VT 4T_)

(b) v. = I - .5 V. cosa

(c) F and F' from equation (4)

(d) V.n+ I = V.n - F/F'

(e) _ = - F/F'

If I_I _ .00001 set v.n = V.n+ 1 go to step (b) and iterate until

< .o00oi.

(f) set v. = V.n_ I

5. Calculate the corrected hover power coefficient from

CPH = Cp - p (CT cosa - CNF slna) + kH (l-v.) ICTI_'zvT

132

CR177436

.80 -

.76 -

.52

NOTES :

-|.006

TRUE HOVER PERFORMANCE

APPARENT I S=HOVER

PERFORMANCE

LIFTING-LINE ANALYSIS

WIND - 3 KTS

VTIP= 779 FPS

CNF= 0

kH= 1.15

I•008

I I I I.010 .012 .014 .016

ROTOR THRUST COEFFICIENT ..(CT)

i.018

Figure A.2 Effect of Wind on Hover Performance

134

I•020

CR177436i

1.0

.98

-r"

0..

"_ .96.1'--

0.

II

itm

'> .94

,l,--

.92

.90

CT- .015

VTIP- 779 FPS

Figure A.1 Effect of Wind on Induced Power

133

I. Rel_ort No. 2. Government Accmaion No.

CR 177436

4. Title irtd Subtitle

Hover Performance Tests Of Baseline Metal and

Advanced Technology Blade (ATB) Rotor Systems for

the

7. Author(s)

K. Bartie, N. Ale_apder, M. McVei_h, S. La Mon,

g. _f_ming Or_n:_tion Name and Addr_

Boeing Vertol Company

P.O. Box 16858

Philadelphia, Pennsylvania 19142

Sgor_oring Agency Name and Address

NASA/AMES RESEARCH CENTER

MOFFETT FIELD, CALIFORNIA 94035

3. RecJOient's CatalOg No.

October 19866. _m,.g O,_an, zaz0o- Code

8. _.,_orm_ng Orginization Report No.

D210-12380-I

10. Work Unit No.

11. Contrlct or Grant No.

_?-1 1 '_13. Type of Rel3o_ and Plriod Covered

F INAL

14 Soonsorincj Agency Code

505-61-51

15 _D_em_tarV Not_

POINT OF CONTACT: TECHNICAL MONITOR, MARTIN D. MAISEL

M/S 237-5, NASA AMES RESEARCH CENTER

MOFFETT FIELD CA 94035 (415) 694-6372

16 Amtrect: Rotor hover performance data were obtained for two full-scale rotor

systems designed for the XV-15 Tilt Rotor Research Aircraft. One rotor

employed the rectahgular planform metal blades (rotor solidity=0.089) which

were used on the initial flight configuration of the XV-15. The second

rotor configuration examined the non-linear taper, composite-construction,

Advanced Technology Blade (ATB), (rotor solidity = 0.10) designed to replac_

the metal blades on the XV-15. Variations of the baseline ATB tip and cuff

shapes were also tested.

A new six-component rotor force and moment balance designed to obtain highly

accurate data over a broad range of thrust and torque conditions is describe,

in the report. The test data are presented in non-dimensional coefficient

form for the performance results, and in dimensional form for the steady and

alternating loads. Some wake and acoustics data are also shown.

7. Key W_ (Suggest_ _v Auth.(s))

ROTOR

HOVER PERFO_4ANCE

TILT ROTOR

XV-15

18. Oistr_tl_ Staten,n(ROTOR LOADS UNLIMITED

STATIC PERFORMA_E SUBJECT CATEGORY 05

- TEST L .

20. Security Cie_f. (of mi| I:age) 21, No. of PMll4 22. _ice"UNCLASSIFIED UNCLASSIFIED 151

"For _le by the NItionll TKhnical Inf_mati_ _lce, SOft,field, Virgiml 22161