*10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval...

117
*10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND DEVELOPMENT CENTER Bethesda. Md. 200134 AERODYNA'MIC CHA.RACTL..STICS OF THE CLOSE-COUPLED CANARID AS APPLIED TO LOW-TO.MODERATE SWEPT WINGS VOLUME 2: SUBSONIC SPEED REGIME by APPROVED FOR PUULIC RELEASE: DISTRIBUTION UNLIMITED, C' AVIATION AND SURFACE EFFECTS DEPARTMENT RESEARCH AND DEVELOPMENT REPORT Ja- nuary 1979 DTN~3RDC-79/002 'U

Transcript of *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval...

Page 1: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

*10

cc DAVID W. TAYLOR NAVAL SHIPRESEARCH AND DEVELOPMENT CENTER

Bethesda. Md. 200134

AERODYNA'MIC CHA.RACTL..STICS OF THE CLOSE-COUPLED CANARIDAS APPLIED TO LOW-TO.MODERATE SWEPT WINGS

VOLUME 2: SUBSONIC SPEED REGIME

by

APPROVED FOR PUULIC RELEASE: DISTRIBUTION UNLIMITED,

C' AVIATION AND SURFACE EFFECTS DEPARTMENTRESEARCH AND DEVELOPMENT REPORT

Ja- nuary 1979 DTN~3RDC-79/002'U

Page 2: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

PAGESARE

MISSINGIN

ORIGINALDOCUMENT

Page 3: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

MAJOR DTNSRDC ORGANIZATIONAL COMPONENTS

COMMANDE R00

TECHNICAL DI RECTORF7 01

OFFICER-IN-CHARGE OFFICER-IN.CHARGECADEOC ANNAPOLIS ,0

CRROK05 04;IsDEVELOPMENT

DkPARTMENT

SHIP PERFORMANCE AVIATION ANDDEPARMENT - - SURFACE EFFECTS

15PR NDEPARTMENT 1

STRUCTURES COPTATONDEPARTMENT - - MATHEMATICS AND

17 LOGISTICS DEPARTMENT

SHIP ACOUSTICS 1PROPULSION ANDDEPARTMENT AUXILIARY SYSTEMS

S,19. EPARTMENT27

ENGINEERING - INSTRUMENTATION

DEPARTMENT 28 DEPARTMENT 29

I I I i I I I I I I ,

Page 4: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

UNCLASSIFIEDSIECURITV CloWF7ICATIOp4 Of' THIS PAOE (When Date Entered)

R' PORT DOCUMENTATION PAGE BEFORE COMPLETING FORM

r~ HNSr-c9/21 2. GOVT ACCESSION NO, 3 RECIPIENT'S CATALOG NUJMBER

~ ,AROYAIC .CHARACTERISTICS OF THE,.CLOSE- /iaCOUPLED CýAIARD AS.APPLIED TO MOW-TO- *jiB1 4 - /

M$ODERATESWEPTINS S. PERFORMING OftG. REPORT NUMBER

JVOLUME 2. SUBAONIC SPEED REGIME,* Aero Report 12577. UTO~f) -~'---- U CN.AT GGRANT NUM1191(e)

/David -u./Lacey

S.PRFRIG RAIIATION NAME 5 GRS 0 R~rM1EETPOET TASK

David W. Taylor Naval Ship Research Air A, "Pp" 4O6UNTNU0

and Development Center Program Tfletaent 6224IN

Bethesda, Maryland 200F4 Work Arui M 1 4]-21-0

II. CONTA060.IPO OFFICE- A-Wr j-.6001

i-.~~IS J..1.11 OF 0-________

Naval Air Systems Command UCASFE

AIR 320 UCASFE

Washington, D.C. 20361 094ASSW7tCA-rioN7OVIOWRWIINT

1S, 011TAISUTION TATEMENT7 (.t Ohl1.Report) 1 ,

APPROVED FOR /ULC R FASE: DISTRIBUTION TdNLIMITED

17. 01ST R IHaU TI1N 17 A T fM E T h~I I. sh~e f(. nt 14* 4411 1 ft¶ ork " 2t'. Nit~Sllernt ttarn~pnpetl

16 SUP161.1MENTARY NOTES

12 KEY WOROS (Continue on reverse side If oecessary and Identify. by block i'umbet)

Canards, subsonic, lift, drag, pit~ching moment, research models10. AiS IE T Con tinuean re verse aid. II meceseery end iIdentity by block number)An analysis of the' effects of canard size, shape, position ind deflec-

tion on the aerodynamic characteristics of two general research modelshaving leading edge sweep nnglem. of 25 and 50 degrees is presented. Theanaiysict qumlrn.ri?,,es the findinpi; of four experimental subsonic wind-tunnel.programs conducted at the David W. Taylor Naval Ship Research and Development.Center between 1970 and 1974. The analysis is based on four canard geome-tries varying In planform from a 60-degree delta to a 25-degree swept wing

-Continued on reverse side)

DD IAi31473 E-omor IFINov as isoS OLtTF. UNCLASSIFT ED-~ . -SECURITY CLASSIFICATION OF TAsS PAGE (ften bt~e Entered)

Page 5: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

UNCLASSIFIED• O4Il Y CLASStIC&TION OF TP41S PAGE(W*h- Del. Enit**d)

(Block 20 continued)

""Ihigh aspect ratio canard. The canards were located at seven differentpositions and deflected from -LU to 25 degrees.

Significant findings Include: the excellent correlation between canardexposed area ratio and changes in lift, drag, and pitching moment; the deL-rimental effect of positive canard deflection; and the optimum longitudualposition for each canard shape for maximum improvements in lift and drag.It is further concluded that the favorable aerodynamic changes caused byitLerference of the close-coupled canard are not significantly dependent onwing leading edge sweep or wing leading edge modifications.

L

UNCLASSIFIEDlllCLrU ITY CL SI FI,'CATION OFi TMISI pAGE(Wlhle DOMli I•'¢li, d,)

L-7

tzr

Page 6: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

FOREWORDThis report summarizes the findings of close-

coupled canard research performed by the Aviation

and Surface Effects Department of the David W.

Taylor Naval Ship Research and Development Center.The work was performed between 1970 and 1974 and was4

funded by the Naval Air Systems Command (AIR 320).Ft The purpose of the report is to provide a summary of

the aerodynamic findings obtained from a series of

wind tunnel evaluations involving three general re-

search models and the F-4 aircraft. The report Is

presented in four volumes--Volume 1: General

Trends; Volume 2: Subsonic Speed Regime; Volume 3:

Transonic-Supersonic Speed Regime; and Volume 4:

F-4 Phantom II Aircraft.

lii

Page 7: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

TABLE OF CONTENTS

Page

I[ST OF FIGURES ........................ ........................... vi

LIST OF TABLES ................. ........................... ix

NOTATION ..................... ............................... x

ABSTRACT ....................... ........................... .. 1

ADMINISTRATIVE INFORMATION ....... ..............................

INTRODUCTION ........................... ............................ I

LIFT ............... .......................... ..... ............ 4

SIZE .............................. .............................. 5

SHAPE ........................... .............................. 6

POSITION .......................... ............................. 9

DEFLECTION .................... ........................... . . 1.6

WING LEADING EDGE CHANGES ........... .................... .. 25

PITCHING MOMENT ................ ........................... .. 28

SIZE ............................ ............................... 30

SHAPE ........................... .............................. 35

POSITION .................. ............................. .. 35

DEFLECTION .................. ............................ .. 43

WING LEADING EDGE CHANGES .................... . 50

DRAG ....................... ................................ 51

SIZE .......................... ............................... 514

SHAPE ......................................................... 58

POSITION ................ ............................. ... 60

DEFLECTION .......................... ........................... 70

WING LEADING EDGE CHANGES ................. .................... 79

FLOW VISUALIZATION ............... ......................... .. 84

SUMMARY ...................... .............................. .. 89

SIZE ...................... ............................... .. 90

SHAPE ......................... ............. ................. ... 9)

v

V1

~?

Page 8: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Page

IPOSITI[ON ........................ ............................. 90

1)FFI, ('IUT [ ON . . . . . ... . 9 1

WING LEADING EDGE ChANGCES . . . . . ... .

ACKNOWLEDGMENTS ............. ......................... * . . 92

APPENDIX - MODEL GEOMETRY ............ ..................... .... 93

REFERENCES .................. ............................. .... 101

LIST OF FIGURLS

1 - Sketch of Models........ ... ................ ... 2

2- Canards .......................... ..................... .... 2

3 - Geometrically Similar Canards ................ .......... .. 3

4 - Canard Positions ..................... ................. .... 3

5 - Typi'al Lift Characteristics ............ ........... ...... 4

6 - C versus Canard Exposed Area Ratio ................... 5L20

7 - Incremental Lift Characteristics of the VariousCanard Shapes ....... ............................. 7

8 - Lift Characteristics of 25- and 50-Degree WingModels ........... ............................. . . 8

9 - Position Effects of Incremental Lift. .............. 10

1O - Maximum Incremental Lift Variation i.th CanardPosition ......................... .......................... 13

1. - Incremental Lift at 32-Degree Angle of Attack ... ......... ... 15

12 - Variation in Lift Coefficient at a - 20 Degreesversus Canard Deflection Angle 6. ..... ............... ... 17

c

1.3 - Incremental Lift Variation with Canard Deflection ......... ... 19

14 - Variation of a ACL with Canard Deflection ... .......... ... 22

15 - Variation of a AC with Canard-Wing Vertical Gap ....... .... 23

vi

Page 9: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Page

16 - Vrlration of' a AC /AC with Caaard-WnIngVert EI al (iap . . .M 1. . . . . . . . . . . . . . . . . . . . . 24

17 - Geometry of Wing Leading Edge Droops and Radii ............ ... 25

18 - Effect of Wing Leading Edge Droop and Radiuson Lift Chara.teristics ......... .................... .. 26

19 - Lift Characteristics of 25-Degree Model Modifiedwith -9-Degree Wing Leading Edge Droop .... ............. ... 27

20 - Incremental Lift due Lu Canard for Basic and-9-Degree Droop, 25-Degree Wing Model .... ............. .. 28

21 - Typical Pitching Mov.ent Characteristics .... ........... .. 29

22 - Variation of Pitching Moment Coefficient atZero Lift C with Canard Deflection

Ang o , 6 .. . . . . . . . . . . . . . . . . . . . . . . . . .. . 31

23 - Zero Lift Pitching Moment versus ProjectedArea Ratio ................ ........................... ... 33

24 - Zero Lift Pitching Moment versus ExposedArea Ratio ................ ........................... ... 33

25 - Zero Lift Pitching Moment Variation withCanard Exposed Volume Coefficient .. ............... 34

26 - Neutral Point Shift Variation with CanardExposed Volume Coefficient ........ ................... ... 34

27 - Incremental Moment Variation with CanardShape ................... ............................. .. 36

28 - Incremental Moment Variation with CanardPosition .................. ............................ .. 37

29 - Variation of a ACM32 with Canard Position ... ........... .. 40

30 - Incremental Moment Slope Variation withCanard Position ............... ........................ .. 42

31 - Correlation of Neutral Point Change withCanard Exposed Volume Coefficient ........... ............... 43

vii

Page 10: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Pt_

Page

K' - Inc rcml'rnta I Moment Change dche to Canard Didl-I ,k_:t Lon . . . .i. . 44

3 1 - Va LLaL i ouis o1 a Ac with Canard VcflecLtiu. ..... ..... 48

34 - Canard Control Power . . . . . . . . . . . . . . . . . . . . . . 49

35 - Moment Characteristics of Basic 25-Degree Wingand 25-Degree Wing with -9-Degree Droop .... ............ .. 51

36 - Lncremental Moments due to Canard on Basicand Modified 25-Degree Wing ......... .............. ..... 52

37 -Typical Drag Characteristics ........ .................. .. 53

38 - Typical Lift-to-Drag Ratio Characteristics ...... ........... 54

3) - Effect of Deflect tun on Zero Lift Drag C. ... ........... 56D)0

40 - Variation of Canard Drag Coefficient withCanard Lift ................. .. ........................ 57

41 - Variation of (L/ID) with Canard Exposed AreaRatio ........................... ............................ 57

42 - I lt-to-Drag Ratio Varlation with Canard,'-lh. . . . . .................... 59

43 - Tndu,',d Drag F,'actor ................... .. .................... 61

44 - LI ft-to-Drag Ratio Variation with CanardLIlcitionl .... . ...................... ....................... 62

45 - E,'fett of Canard Post tinn on MI.nimum Drag(:ucfl•i'c nt. ................ ......... . 66

4h - I.dticLed I)rag Ira tor Var Iint-Ion wlLE h (CaIIILrdI'OStIOt[n fr tihe 50-Dogrve WIng . . . . . . . . . . ..... 7

47 - [iLduced l1rIg Factor Vrlmntloin witLi Canardtucuoltton fur Lhl 25-I)egrc*'u WInl. ........ .......... 69

48 - Lm..iiM H of HItlnli •i Irid Jlk. d Drag Vm.ti or . .. . . , . . . . . . .. . 71

49 - Iw lot-tn-Iri, Rat jun Vxir llloin wl ii C('anmrdl) (,- l eci t lnm . . r . . . . i . n. . . t . . . . . .. . 7 2

A 0 - ý1t 1. 1mliu Drop Vitr io l Itlm w ith (Itimn rd 1)(If lt, 11 . . . . . . ... 71.5

Vill

Page 11: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Page

51 - 1, f fec t of :;i nard rl'f Icct it on 0 i1 InducedDrag F ict.or . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

52 - Effect of -9-Degree Droop on Lift-to-Drag Ratio ......... 80

53 - Incremental Change in Lift-to-Drag Ratiodue to -9-Degree Droop .............. ..................... .. 83

54 - Tuft Patterns of the 50-Degree Wing ..... ............... ... 85

55 - Effect of Canard on Unseparated Flow Region ......... ......... 86

56 - Effect of Canard Shape .............. ..................... .. 88

57 - Effect of Po,41lton ................ ....................... .. 88

58 - Research Aircraft Fuselage ............ ................... .. 96

59 - Platform View of the Wings .......... .................... .. 97

60 - P tanform View of the Canards .......... .................. 98

61 - Canard PivoL Locations .............. ..................... .. 99

62 - Wir1 Tutinel Model Componients ......................... 100

LIST OF TABLES

.i - ometrIc, Characteristics of the Wing ..ds .......... ............ 94

2) - rcometric CharacterItLics of the Canards.. ............ .......... 95

LKI

Page 12: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

NOTATTON

AR AAspvct ratto

Ii lv I,

Canard pro-ected span

CD Drag coefficient, D/qSw

CDc Drag coefficient of canard referenced to total canardC exposed area

CD Minimum drag coefficient

C' Canard

CI• I,ift coefflcient, ],/qS

CwU Lift coefficient of body alone

Cl ~~T~ft coefficient of body ond canard i'B+C

GI, MaxLimunI Lift coefficienta

iIX

'Lift coefficient of wing iand bodywB

C ift coefficient evaluated at minimum drag value

C, 2fL coefficient evaluated at a - 20 degrees

C2LIft coefficient evaluated at a - 32 degrees

{: -l~ift curve slope of canard, referenced to(i canard exposed area .

XII

x!

- .!

Page 13: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

C1 t curve slope of ,nnard due to def lect ion

C Pitching moment coefficient, pitching moment/qSM w

MB+ Pitching moment coefficient of body alone

C Pitching moment coefficient of body and canardMB+C

C Pitching moment coefficient of body and wingMWB

C Zero lift pitching moment

C Pitching moment coefficient evaluated at a 3 32 degrees32

M MN N

c Mean aerodynamic chord, inches

D Drag, pounds

i Canaud shape

J Canard position

L Lift, pounds

L/D Lift-to-draj; ratio

(L/D) ma Maximum lift-to-drag ratio

Distance between center of grality of wing and canardc pivot location, inches

P Canard position

q Dynanic pressure, pounds per square foot

xi

L..-

Page 14: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

SC Canard Projected area, squace feet

s. Xa11ard cxpose' area, square feet

S W Wing reference area, square feet

W Wing

Longitudinal distance, inches

z Vertical distance, inches

z t Vertical gap measured from canard trailing edge to

wing upper surface, inches

Angle of attack, degrees

ACL CL - CLw

ACL L - C'C B+C 8

AC Maximum value of incremental liftLMIX

AC3 Incremental lift evaluated at a = 32 degrees3.)

A-CM C M -- C~A-MWB

CMM CM

C B+C B

AC AC evaluated at a =32 degrees

Canard deflection angle, degrees

k Wing leading edge sweep angle, degrees

x 25 Quarter chord sweep angle of canard, degrees

xii

•-:.

Page 15: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

ABSTRACT

An analysis of the effects of canard size, shape,po,0it Ion and dcef lection on the aerodynamic character-lstics of two general research models having leadingedge sweep angles of 25 and 50 degrees is presented.The analysis summarizes the findings of four experi-mental subsonic wlnd-tunnrl programs conducted at theDavid W. Taylor Ship Research and Development Centerbetween 1970 and 1974. The analysis is based on fourcanard geometries varying in planform from a 60-degreedelta to a 25-degree swept wing high aspect ratiocanard. The canards were located at seven differentpositions and deflected from -10 to 25 degrees.

Significant findings include: the excellent cor-relation between canard exposed area ratio and changesin lift, drag, and pitching moment; the detrimentaleffect of positive canard deflection; and the optimumlongitudual position for each canard shape for maxi-mum improvements In lift and drag. It is further con-cluded that the favorable aerodynamic changes causedby interference of the close-coupled canard are notsignificantly dependent on wing leading edge sweep orwing leading edge modifications.

ADMINISTRATIVE INFORMATION

This work was undertaken by the Aircraft Division of the Aviation and

Surface Effects Department of the David W. Taylor Naval Ship Research and

Development Center. The program was sponsored by the Naval Air Systems

Command (AIR 320) and was funded under WF 41432-09, Work Unit 1600-078.

INTRODUCTION

This Is the second volume of a four-volume report summarizing the

close-coupled canard work accomplished at the David W. Taylor Naval Ship

Research and Development Center between 1970 and 1974. This volume summa-

rizes the findings of a series of wind-tunnel programs conducted at subsonic

speeds.

Volume 1 of this report presented the general trends of close-coupled

canards on aircraft of low to moderate wing sweep. It was shown that close-

coupled canards con significantly improve stall angle of attack, increase

j 1

, •::•"~m•• W • • lr • :"• • • '-- • • .. . , ',•. .. . ,4 •'r '-•- •-•r: • •:• • . ... .. . . . ..... . • . ...... ..... . .= • ,.

Page 16: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

the maximum lift coefficient, and reduce drag. The extent to which these

imprOVeCl'ftS occiir Is ai function of canard size, shape, position, and de-

f ltct ion. l'Ihu.se variablcs, as well as the influence of wing leading edge

modifications, are discussed in detail in this volume.

The discussion is based on four wind-tunnel programs conducted in the

DTNSRDC 8 x 10 foot subsonic wind tunnel. Two general research models were

utilized in this program. The models had leading edge sweep angles of 25

and 50 degrees. Skatches of the models are shown in Figure 1. Four canards

S 50.DEGREE SWEPT WING

25-DEGREE SWEPT WING

Figure I - Sketch of Models

of dlffprent planform wore evaluated. The shapes were a 45-degree clipped

delta designated Co, a 60-degree pure delta CV, a 45-degree high aspect

ratio canard C2 , and a 25-degree canard C3 , as shown in Figure 2. In addi-

Figure 2 - Canards

LJon, Four geometrirally similar versions of canard C were evaluated with

projected area ratios of 0.10, 0.15, 0.20, and 0.25. Relative sizes of the

four canards are shown in Figure 3.

2

Page 17: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

- - ---- BODY C'

- A - 24(6.10)iTYP

. . . AIRFOIL SECTION60A00d

II.• PIVOT DISTANCE VAN TIP CHORD MOOTCHOAD CANARD AREASA 6 C 0

IN. C_ IN . CM. IN, CM. N CM. IN, CM,'I

0.10 1.311 31111 2.74 1t.96 0.38 0,97 3.12 7.92 30,8 196.6

0 ,15 1.75 4.44 3. 61 9.93 0.44 1 ,12 4.35 11.02 46.7 294.6020 2.25 5.72 4.4 12.29 o.. 1.42 19140 13.62 slo $ 3.5

026 262 6.66 6.74 14.56 0.66 1.10 6,33 161.0 7ls 4 .0

Figure 3 - Geometrically Similar Canards

The models have seven positions at which the canards can be located

(see Figure 4). Positions are numbered from fore to aft and top to bottom.

Position I Is the highest, most forward and position 7 is the lowest posi-

Lion. Deflection range varied from -10 to 25 degrees. Detailed dimensions

of the models, canards, and positions are given in the Appendix.

NOTE: ALL DIMENSIONS ARE IN INCHES (CENTIMETERS)

I i110100I42.50-'- .0 w4-.----+11.3 12 so36 V

1,35 16,3615.

(26.401 MOMENT

+ + REFERENCEP1 P P3 POINT

1,0012.541

("54•p0 CHORDPLANE 13.5-

Figure 4 - Canaird P'ositions

Page 18: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

The discussion is organized into major topics of lift, pitching moment,

and drag. Subtopics include the effect of canard size, shape, position,

deflection, and wing leading edge changes. The data are primarily pre-

sunted Ls incremontat changes in lilft and pitching moment due to the above

parameters. Drag is presented primarily as lift-to-drag ratio at constant

lift coefficient. Data for both 25- and 50-degree wings are presented to

indicate that the favorable effects of close-coupled cana:ds are applicable

to aircraft of relatively arbitrary wing planforms.

LIFT

Typical variation of lift coefficient with angle of attack for the

25- and 50-degree wing are presented in Figure 5. Data are shown for three

- W ingDLGRIE WING

1.4- 0 W+c_

1,2 -VW C3

'

110-

II.-DEORER WING

0.4 -

0-

-0.2

-a,2 L • , t I I I-4 0 4 a 0 4 8 12 16 20 24 28 32 36

"ANGLE OF ATTACK (DEGREES)

Figure 5 - Typical Lift Characteristics

canard shapes: a 60-degree delta, a 45-degree high aspect ratio canard,

and a 25-degree high aspect ratio canard. For all configurations there is

aI 4izeablc increase in lift when the canard is installed to the basic

4

Page 19: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

wing-body. The increase in lift varies somewhat with the particular canard

size, shape, position, and deflection. These differences in lift are dis-

cussed in thie following sections.

SIZE1*

One of the prime variables of the first canard wind tunnel program

conducted at DTNSRDC was the effect of canard size on the aerodynamic char-

acteristics of the 50-degree research model. Four geometrically similar

canards having projected ar'ea ratios of 0.10, 0.15, 0.20, and 0.25 were

evaluated. Relative sizes of each canard are shown in ligure 3. Data from

this wind tunnel evaluation were limited to an angle of attack of 20 de-

grees. The variation of lift coefficient at 20 degrees is presented in

Figure 6 for seven canard positions.

8./sw

u 01 0 Olb 010| 0 2II e

S 0 11.. 0i

0.0

wLe

o IRot 0.04 0d O6 0go 0.10 0.11 Old 0ai

Figure 6 - CL20 versus Canard Exposed Area Rntio

A compleLe listing of references is given on page 101.

Page 20: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

The effect on canard size varied somewhat with canard position. At

those positions where the canard iw fairly close to the wing, P2. P3, P5'1 thiere is a distinct curvature to the data, however, as the canard is

moved Furthler forward L1h0 varl[ato of t It'C with size becomes I inear.120

Little favorable effect would be expected because the interference is mini-

mized. Included In the figure is the value of the lift coefficient for

the 0.25 canard and wing body if no interference were present.

Comparison between this value and the data shows favorable interfer-

ence for the high canard locations (PI, P29 P3 ), and unfavorable inter-

ference when the canard is in the plane of the wing (P 7 ).

SHAPEIncremental lift is presented in Figure 7 for the various canard

shapes. The canard location is position P for all four canard shapes.3

As stated in Volume I., P3 was near optimum for all canard shapes. The

figure is for both 25- and 50-degree research models and contains Isolated

data for each canard shape. I

The most significant conclusion which can be drawn from the figure

is that the large difference in incremental lift bctween 25- and 50--degree

wing configurations Is in the angle of attack range between 12 and 28 de-

grees. This difference is attributable to the poor stall characteristics

at the 25-degree wing.

A comparison between the canard-off characteristics of the two con-

figurations is given in Figure 8. The 25-degree wing configuration ex-

hibits as expected a higher lift curve slope than the 50-degree configura-

tion but the stall angle of attack is only 10 dagrees for the 25-degree Iwing versus 20 degrees for the 50-degree wing.

The favorable interference between canard and wing, therefore, delays

stall at a lower angle of attack for the 25-degree wing.

Lhis reduction in angle of attack due to favorable interference is

clearly seen in the low cross-over point between the isolated Lanard data

6

Page 21: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

0,6

04 1,/C . / '-

/ "° -I //

0.2 - ISOLATI/ / /°LA7ED

op ~-.0-

25 U,11

/ " /#- " " " i 4E•B

0.4 /

0.2 -' INOLA 0 ISOLATED

- - 1I I

0 1i 24 32 0 a 24A"OLk OF ATTACK DEIUREEB) . ANGLE OF ATTACK (DEGREES)

Figure 7 - Incremental Lift Characteristics ofthe Various Canard Shapes

7

Page 22: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

IT

I

1.6!

1.4-

1.2 50 DEGREES

S1025oDEREESL.• .-- •,,SSW•S•S• ••

U.. .U.LU

0.4

0.2

4 8 12 16 20 24 28 32 36a, ANGLE OF ATTACK (DEGREES)

Figure B - Lift Characteristien of 25- and 50- Degree Wing Models

and the complete configuration. Favorable interferunce occurs at approxi-

mately 12 degrees for the 25-degree wing, whereas favorable interferencu

does not occur until thu angle of attach reaches approximately 16 to 20

degrees for the 50-degree wing. It can be said that the poorur the wing

design, the more the canard can help.

Page 23: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Examination of the data with regard to the individual canard shape

indicates that the 60-degree dltL-i canard G maximizes the increase in ilft

for both win.-ig. 'rhi, hO-dv-rLc 'nanrd Is (lotiely followed by the 45-degree

hi 0,61 aspo't ratio canard a and the 45-degree (1J. ppUd del tl COU

The low sweep canard C3 exhibited the lowest incremental lift for ýhe

25-degree sweep model, and, in fact, the canard appear& to have stalled at

approximately 20-degree angle of attack.

Incremental lift for th, low sweep canard is approximately the same

as the other canards for the 30-degree wing model, It thus appears that

while low sweep canards are Inadequate for the low sweep wing, the low

sweep canard doem delay separatiLn Aufficiently for the wing of higher

sweep if located at the proper pusittion. The effect of position change

on incremental lift for the various canard shapes and the two wing sweepm

id discussed in the next section.

POSITION

Incremental lift versus angle of attack is presented in Figure 9 for

both 25- and 50-degree sweep research models. Data 2 are presented for

seven positions for the 50-degree model and three positions for the 25-

degree model. The data are for the four canard configurations at zero

degrees canard deflection.

Incremental lift was, in general, maximized at 28-degrees angle of

attack for the canards on the 50-degree model and between 20- and 24-

degrees angle of attack for the 25-degree model.

The variation of maximum incremental lift with canard position is

shown in Figure 10. The interference free value of canard lift at the cor-responding angle of attack for each position are also shown in the figure.

For all configurations, as the canard was moved to the most forward

position, Z/ - 1.5 for the 50-degree wing, X.//c 1.30 for the 25-degreec

wing, the maximum incremental lift dropped off. Similarly, lowering the

canard reduced maximum incremental lift. The only exceptions to the latter

statement were the low sweep 25-degree canard C3 , and the high aspect ratio

45-degree canard C Canard C3 had an increase in maximum lift at the low-

est, most aft position for both 25- and 50-degree wing models. Similarly,

lift was maximized at P 6 for the 45-degree high aspect ratio canard C2.

9

Page 24: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

' IIur! , 9 - l',) iL himI EI" cL' ,c L )I- Ilnur mlI nL , lif t

0.6

S0.4

0,2

ANGLE OF ATTACK (DEGREES)

Figuru 9a - Canard C on 50-1)egree "ing0

1, IOE4

0,2

0 B 15 24 .12

0, ANGLk OF ATTACK MOELUS)

Figure 9b - Canard C L un 50-Degree Wing

i0

Page 25: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Figure 9 (Continued)

0.4 P

0,0

0 8 0 8 0 0 lS24 32

k, ANGLE OF ATTACK (DEGREES)

Figure 9c - Canard C9 on f50-Degree Wing

0,4 -

0.2\

Pt Pp0 16 0 16 0 I 16 24 32

t, ANOLE OF ATTACK (DEGRE•SI

Figure 9d - Canard C3 on 50-Degree Wing

3I

a j-1

Page 26: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Figure 9 (ConLinued)

0.,

0 .4 .-

0.4 ./

0.2 '

- I-

0,II. - --

040 /

0 4 8 12 16 20 24 2R 32 0 8 10 24 32ANGL2 OF ATTACK (DEGREES) ANGLE OF ATTACK (DEGREES)

Figure 9e - Cannrds on 25-Degree Wing

12

Page 27: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

0.4 .. II ,DLAI .J I ,IOAL It I ,

00.75 1.00 1.25 1.50 0.75 1,00 1.25 150 0.75 1,00 1.25 1.50 0,75 1,00 1.25 1, b0

C~!e 1.2/'eU,=' I,'/1

Figure lOa - Lift on 50-Degree Wing

J J

0.4 0Ope

ISOLATWD0.2 "IILAD- ILAtKD

,10 LAW TaIr/

,%.5 1,00 1.25 1.50 0.75 1.00 1 IS 1.0 0.1 100 1.25 1.60 0.75 1,00 1.25 1',0

Figure 10b - Lift on 25-Degree Wing

Figure 1.0 - Maximum Incremental Lift Variationwith Canard Position

13

Page 28: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Examination of the data relative to the interference free value indi-

%avsC, that the incremental lift is approximately double that of the inter-

icrc , O ' 1. i LIt , if the canard is properly I y I cated. Improper location,

i.e., too far forward or too low, reduces the value of the incremental lift

to approximaLely I l/z times the interference free value.

The value of Incremental lift obtained is approximately the same for

both 25- and 50-degree research models for properly located canards. The

only exception is the 45-degree high aspect ratio canard which had signifi-

cantly higher values of incremental lift when mounted on the 25-degree

research model.

The data presented in Figure 10 are not at a constant angle of attack,

hence, they do not represent the maximum lift coefficient obtained by the

complete configuration. The angle of attack where waximum lift for the

c:ompleLe configurations occurred was generally at 32 degrees. Figure 11

presents the canard incremental lift at 32 degrees versus canard position,

thus also showing the influence of canard placement on the maximam lift

cUefficienLt. Included in each figure is the interference free lift for

each canard shape. In Figure 10, the incremental lift was always greater

than the interference free lift. This is not the czse at 32 degrees, par-

ticiularly for the 50-degree research model. For all but the 60-degree

delta canard CI, there are canard locations where the incremental lift is

less than what would be obtained from the interference free value, thus

indicating unfavorable interference. The onset of this unfavorable inter-

ference occurs at Z. /c of approximately 1.4 for canards C0 and C, and ap-

proximately Z /c of 1.2 for canard C3 . Lowering the canard further movedc

the intersection point aft and reduced the lift.

Thr, trends for the 25-d-gree rcsearc'h model are similar to those of

the 50-degree model, although the only intersection point noted is for the

25-degree canard C. This intersection point occurs at approximately the

same Z /c value as that of the 50-degree wing model.c

14i

---- - .2 - -

Page 29: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

0.4-

SIOLATED I•OLATED ISOLATED%

0.75 1.00 1.25 1.50 0.75 1.00 1.25 1.50 0.75 1.00 1.25 1.50 0,75 1.00 1.25 1.50

voltFigure Ila - Lift on 50-Degree Wing

0.2

0 ISOLTE ISOLATE ISA IED

0.

0.75 1.00 1.26 1,50 0.75 1.00 1.5 1.50 0.75 1.00 1.25 1.50 0.75 1.00 1.25 1.50

Figure l1b - Lift on 25-Degree Wing

Figure 11 - Incremental Lift at 32-Degree Angle of Attack

1.5 •_• .I

Page 30: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

DEFLECTION

Canard deflection has direct Influence on the maximum lift generated.

This inf-lue nc e, either favorab• c or unfavorable, Is dependent on canard

position and size.

The variaticn of lift coefficient at 20-degree angle of attack are

presented In Figure 12 for the four sizes of the 45-degree truncated delta

canard C 0. In general at low deflection angles, 6c -< 10 degrees, there is

little change in the lift coefficient for all positions and sizes. As de--

flection is increased different trends occur. The smaller canards S c/Sw

0.10, and 0.15 located in the high positions PIV P2 9 P 3 exhibit little

change in lift coefficient with increasing deflection angle. As the canard

size is increased or moved closer to the wing, there is a decrease in lift

coefficient with increasing deflection angle. This reduction in lift

occurs primarily at P 6 and P 7 for the smallest canard Sc /S - 0.10, and

Positions 3, 6 and 7 for larger sizes. Thus, for canards which might be

used for control purposes, i.e., removed from the wing, there is little

lift loss due to the canard. Canards which are located close to the wing,

however, exhibit lift losses with increasing deflection angle. The previ-

ous discussion is based on data at 20-degree angle of attack. At higher

angles of attack it should be remembered that when the canard is moved

longitudinally away from the wing the likelihood of canard stall increases

and, thus, the above discussion is not likely to hold.

The effect of canard deflection on the incremental lift characteristics

of the four different canard shapes is presented in Figure 1.3 for both 25-

and 50-degree research models. Data are presented for canard deflection

angles of + 10, + 5, and 0 degrees for the 50-degree wing and 0 and -10

degrees for the 25-degree wing model. Canard positions represented are

Positions 3 and 6 (Z /c / 1.0).

Canard deflection has little effect on the incremental lift character-

istics at low angle of attack and, as reported in Volume 1, CL is approxi-LImate]y 1/2 C1 of the isolated canard. At higher angles of attack distinct

differences in incremental lift appear as shown in Figure 14 for the case

of 32-degree angle of attack. Examination of the figure reveals the large

16

Page 31: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

.-

04

LIDI

"i l ' I I, i

C, - 4

- -I I.i'I / .

LIU

4-4

MI IIi "17

,I • I I

"-,I

-h7

Page 32: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

,.*1

II1111 ~a 'I18t

Page 33: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Figure 13 - Incremental Lift Variation with Canard Deflection

0.4

5 (DEGREES)- 0-~ 5

0.2 -- 10

00

WOW- -10

,r ANGLE OF ATTACK (DEGREES)

l IFiga;re 13a -Canard GO on• 50-Degree Wing

19

-,

Page 34: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

t+ +

I --'° I !' /

0 0

0

-j

r, 0U 0 00m

01

Page 35: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Figure 13 (Continued)

( IDEGREES) h, IDEUREESý.,4

,% 1+ 10

- K -I0 2

0-

0. -1-

0

I I .L.________L____________

0 Is 24 32 a B Is 24 32SANULE OP ATTACK (DEGREES) ., ANGLE OF ATTACK (DEOREES)

Figure 13d Canard C on 50-Degree Wing

/ o-.0.4 //

O'C,

o, JI

0 21 4 32ANGLE OF ATTACK IMAM)

0.4 PA c i

Figure Ile - Canards on 25-Degree Wing

20.

oxA I I I• . I ,

O I 2 4 i 32a l / H i i M. . .

Page 36: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

LU

ww 9

I 41WW40

--- r4P-4

~6_ _ _ _

2,

a:

do 0_j

66

0 b0

- U.~cc

- I ,

22a

Page 37: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

dependence of position on lift at equal canard deflection angles. This is

seen most clearly for the 60-degree delta canard C1 and for both 25- and

50-degree model s. When thLI canard is at the low position P 6' positive de-

fiuctions cause a severe lift loss v.ead negative deflections cause a lift

gain. Similar trends occur for both the high aspect ratio canards C and2

C3 on the 50-degree wing model.

Figure 15 utilizing the data from Figure 14 presents Incremental lift

versus the canard trailing edge gap measured between the wing upper surfaceand the canard trailing edge. The gap was made nondimensional with respect

to projected canard span.

0.6

0 C20.4 - o 1O.. ... ...-J V

0.2-

0 0.04 0.08 0.12 0.16 0.20 0.24ZT/balnard

Figure 1.5 - Variation of A Ct3 with Canard-Whng

VcrLical Gap

Without taking into account differences in canard lift curve slope, a

pattern of incremental lift versus gap height can be seen whltch is somewhat

similar to a ground effect plot albeit in an invested sense. In true

ground effect, CL increases with respect to proximity to the ground. I1na

the case of the canard, close proximity of the canard to the wing causing

a lift loss.

2.3

Page 38: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

By taking the Incremental momenr data presented in a later section at

the ammv angle of attack and dv [ding by the correspondding InCrementalrift, FigurL, 16 has been dowlopvd-

4

3 0 C2

0 C3

SiI I I I

0 0.04 0.08 0.12 0116 0.20 0.24ZT/ibanhord

Figure. 16 VarI.aticn of ACM/AC with Canard-Wing Vertical Cap

It is seen that at gap ratios greater than 0.1 the ratio ACM/ACt in

approximately I which is where the canards are locatel, ie,, ,/c - I,.

However, as gap height is decreased AC /AC moves rapidly forward Indicat-M 11

Ing that the canard Is very highly loaded (typical of ground effect) and

the wing ig unloading. This behavior of the wing Is perhaps due to the

canard downwash having an unfavorable effect on the lift. It thus appears

that for good high angle of attack characteristics the canard should be aLleast 0.10 canard spans above the wing plane.

24

Page 39: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

WING LEADING EDGE CHANG;ES

The reisearch models LitlItzed In the previous dlscussLonHs had symmetrl-

al enad log edigs and tho nfortieil Iedidf og edge rad IL. assoc.lati.-d WItLI tOLe

MA008 airftul. It is well known that increases in performance can be ob- ...

tained by suitable changes in wing leading edge radius or droop. In order

to examine the effect of such changes on the aerodynamic characteristics,

three radius changes ond four leading edge droops were evaluated onl the

50-degree model. Details of the radius changes and droops are given In

Figure 17.

LEADING EDGE DROOP LEADING EDGE RADIUS(RADIUS * RI) (DROOP - 0 DEGREE)I R0 DEGREE 0.0044a

R2 0,[o088co - 3 DEGREES

U ~ DEGEESR 4 -0,0176cS '• • 6 DEGREES

I 15 PERCENT CHORD

"•-• • -•ll •9 DEGREES

Figure 1.7 - Geometry of Wing Leading Edge Droops and Radii.

,II25

L M. --. . ;-;T~ • Y - ,'-;:L:;•: X .m " =::.--•••,..,-:'•':

Page 40: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

1AL tVer,40-i .3igle OF at t~nk is p ri4ntLd In F'l~gure 1.8 for the 50-

d~grUL' wing hot I withI and wilithot th c~ anard I-or the varying rad ION and

d run 11) i H; ThO curd f. x Lw 4 5-deg,_LL r L trunci ed de.Ito C0 I ocat ed at pos it on

1, ankld L li'grt-L diit IOt hii A!; r~Lm bV ncel In the fiIgurO, iielthur radiuH

change nor drOOP CdUuSe any apprecitable change in lift for either canard-on

or -of f conf igurat ions.

1ANA1113 r.ANAUDi

A A

n1.4

1) 4 1 I 21 2A vs 12 16

1AN(tIL (IF At TACK WORKI~~

Figurc l8a - lrouPs

1.4

.2 -

I, 0 01

L I _II 4 It 12 Is 20 24 09 * 36

ANIIIA OF1 At tACK IL11468PI

Figure 18b - Radius

Figure t8 -IlffL'c.t oII Wing Lunding Edge Dlroop andRadl~us on Lift. ChrdictL'rts~tk

26

I - ~ ...ZX

Page 41: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

T~Li h8i ck Of Cihange W.11- Hot the case for the 25-degrue ruNaui I11UMLL

as 41iowil in F iguri.' 14. In Llie case of thu 2 ¶-dvgruv model, OILe -9-degree.

-- -,EQE DRO CANARD LAnJ~U

1,2 - do

0. BASICLEADING EDGE

0.4 CNR ATP0tlITION 2I f013E i CANARiD DEFLE~CTION

0 4 8 12 18 20 24 28 32

F igki rt 1.9 - I t ('hnroeter tyt:l' (31'I 2 5-lDg rue Model Mod if tucdW it i 9-lg e Wing Leald i g ikLgu Droop

droop (It'm I t 4tal IIby approx im-itLfly 4-dL'grev angl1e of a~ttaok miid 1 ImprOVe!H

a ~ thu maxi~muni tif coeff [I-eent by 0.24. Thu of fect of oiddhltg tHo omviird to,

tho mud lfied modl~d IsN prelsvotud In Figure 19~. MaL- 11r` presented f'ol tho

60-devrc~ delitn cnnard C1 locate'd at p and 0 degreti dul uttion, I oc rv-3

men tIl, I If t versu atSnag.c o f attack. for LhuV hasvui tu and -9-degreo droop

modeI.is is prciiwntc'd i~n Fl~gurp. 20 fur the 60-dlegree canai-rd at PUIkoIt loaN 3

anid 6 for dif I ct ons of LI and -1I I degrees.

le t t or c Lll.i. a aeLu r I st 1tM of Lhu wliiit , modil ted With HIitLe -9-du1gIrue

drmopl) dllayH the' off'ovL or the conard by ;ipproxhimatolIy 4-degree na1g IL' of7

at toek. Au angie~ ofI attack is I n(reaised bey01nd H tol I 1 augI (If : oit tack; 01L,

value or I ncremeiuta.1 I i f 1H appvnximatolIy the i4;.mu ort s I Ight Iy lv~lioihv for

the mcodi fled wing. [t 1I Ho tisNovn that as thei I)OsIC wingý! chilrnl Or l-lH IC

arv Improved the hl"fI tience of the( cLnAimird is cie layod to Iiiglier mng1 iv of'

alac k.

27

Page 42: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

0.6 -___________ _____________________

BASIC LEADING EDGE

0.4 9 DEGREE LEADING EDGE

0.2

0.

0.2

000

eANGLE OF ATTACK (DEGREES)

Figure 20 -Incremental Lift due to Canard for Basic and

-9-DegrLýe Droop, 25-Degree Wing Model *PITCHING MOMENT

The variation of pitching moment with angle of atLack is pres~iited inI

Figure 21. The data are for the same three canard geome'trieo anda positions

as~ those presented In Figure 5 of the section of 11f'-., mainly. the 60-

degree delta, and the 45-degeee and 25-deg~ree sweep hIgh aspec.t ratio on

28

Page 43: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

0.7

, 0,5- •W+q3

0.5 -o, 25 DEGREE WINO

0.3

X 0.2

00.1

0,3-

50 DEGREE WINO

0.2

0.1

0

-0.1

-0.2 - ' I _I I I I-4 0 4 8 12 1( 20 24 23 32 36

', ANGLE OF ATTACK (DEGREES)

Figure 21 - Typical Pitching Moment Characteristics

canards. As with lift, differences occur due to canard shape. The data,

howe2ver, indicate a fairly linear variation of pitching moment over the

angle of attack range when the canards are installed. The influence of

size, shape, and l)Siltion on pitching moment are discussed in this suc.Lion,

Examinatioll of the data presented in Figure 21 indicaLeh that the

pitching momenL behavior of the basic wing-body Is not linear. This

29* -'~ ~ 'fl~..Lt. IX~W*-l s;a-,~.~ ~s

Page 44: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

nonlinear behavior is most pronounced for the 50-degree sweep wing model.

To determine the effect of the canard, incremental pitching moments ACM

wiLl be used almost exclusively La the folLowing discussion. The basic

data from which these incremental moments have been obtained can be found

in the references.

SIZE

The influence of canard size on pitching moment at zero lift C is

presented in Figure 22. Shown is C versus deflection for canard pro-

jected area ratios SU/Sw from 0.10 to 0.25. As indicated, the increase

in moment with deflection is reasonably linear for each canard poasi'ion

and size.

Figure 23 presents zero lift pitching moment at 1.0-degree deflection .1

versus canard area ratio at Positions I and 3. As shown, the data do notintercept the zero value at zero canard size, therefore, indi.asting that '•the canard projected area ratio is too large a parameter for good agree-

ment. The data, when plotted versus canard exposed area ratio, converge

to zero at zero canard size as indicated in Figure 24. Data are presented

for each of the seven canard positions evaluated and, as shown, linear

fits of the data are obtaincd at each position. As expected, moving the

canard forward increases the pitching moment; not so *xpected is the fact

that lowering the canard reduces the magnitude of the pitching moment

change.

The data from Positions 1, 2, and 3 have been plottad versus canard

exposed volume coefficient anud are presented in Figure 25, Catiard "olume

coefficient is defined as /c/ x S /Sw, where Z, is measured from thb 0.27e

Sposition of the wing to the 40 percent exposed root chord of the canard.

The variation in C with exposed volume coefficient is linear as shown.

The forward shift in neutral point due to canard size is presented in

Figure 26. The parameters chosen are incremental moment slope evaluated

at zero lift versus canard exposed volume coefficient. Data are presented

30

Page 45: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

VNC)

4I C L~

LU 6U

0 1

0 I

on cc w ww U.

a~UJ

'4- C

C.V)

44.

IN II31

Page 46: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

uUu

LUL

I..

CC

No u

0 0.9j

LO U.U

U 6U

CC w

Un 0

0 Ci

LN3W'~OW~ DNIHDJ.Ld LN3WiOWIDJ NIHDJ.ld' WO

32

Page 47: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

El

0 006 010 0,15 0,20 0.26 0,30

lc/swý PROJECTED ARIA RATIO

Figure 23 - Zero Lift Pitching Moment versusProjected Area Ratio

- 0 Pl,4

C P2,.7

P 1,2,3 •ic -10 DEGREES

U.1 0.1

U.

0

P7

0.1

0 0.04 0.08 0.12 0.16SC Sw, EXPOSED AREA RATIO

Figure 24 - Zero Lift PitchinK Moment versusExposed Area Ratio

33

Page 48: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

U.

VcU CANARD__ EXPOSEDVOLUME__COEFFICIENT

0.4 0..212 016 02

0 C Ic CANARD EXPOSED VOLUME COEFFICIENT

Figure 26 -Zero ralf PointcShingMft Variation

with Canard Exposed Volume Coefficient

M 10

c -

to aHilft figu2 e 26 fo theutral Pointr ahit Variaotifonwrosto

wit Caar 1.p5,e Voum Co/Sf0.2)t

at - nd10dere cnad afletio. hevaiaio o ~ACM)234i

Page 49: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

S HAPEThu.I variation t i Incremntal moment with canard shape i!: SihoWL- in

Figure 27. Data are presented for the four canard shapes located at P3 for

both 25- and 50-degree sweep models. Included on each figure are the val-

ues for each Isolated canard shape at P36

As with incremental lift, the most significant changes noted are the

early differoaces between the 25-degree data and the isolated duta. This

is undoubtdly due to the early stall of the 25-degree wing.

At low angles of attack the incremental moment has the same value and

slope as that of the isolated canard data for the four canard shapes andboth wings, thum indicating little if any upwash effects due to the wing.

As angl, of attack is increased, significant differences occur primarily in

magnitude. This is due to the fact that the canard is delaying stall over

the root position of both wings. Thus, the center of pressure has moved

inboard and forward for each wing thereby generating increased, nose-up

moments.

Comparison between the incremental and isolated canard data indicates

reasonable up.reement between the slopes and general shape of the curves for

each canard shape--most notably for the 25-degree sweep model. For example,

the isolated 25-degree higlh aspect ratio canard C has a stall at angles of

attack between 12 and 16 degrees; similarly, the incremental data indicate

a reduction in slope In this angle of attack region. Reductions in slope

are also evident for both the 45-degree high aspect. ratio canard C2 and the

45-degree truncated delta canard C0 and these reductions occur for both

Isolated and incremental data, No reduction in slope is evident for the

60-degree canard isolated data and no reduction is seen for the incremental

data.

It is thus apparent that the general shape of the incremental moment

curve is the same as that of the isolated surface.

POSITION

Incremental moment versus angle or attack is presented in Figure 28

for seven positions for the 50-dogree wing model mnd three positions fnr

35

V *Wu,

Page 50: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

- I

'0.3 50 D8EGREE WING 0,

L.

000 OlP go 0

rr

0 1i 24 20U4

.,ANGLE OP ATTACK (DIGANIE111) •.,ANGLE OP ATTACK 1010RI•ESI

Figure 27 - Incremental. Moment Variatioil with Canard Shape

36

]02 7

Page 51: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Figure 28 -Incremental Mome~nt Variation with Canard Position

0.4

0. 4 - 2

3) 0.4

16 203 24 2 3

04

ANL FATTACK (DEQREES) 2 I 0 2 6 3

Figure 28a Canard C2 on 50-Degree Wing

a , I0 4 a 12 4 a 12 4 a 12 1s 20 24 2 B 32

ANGLE OF ATTAI.CK (DEOREEB)

Figure 28b Canard C3 mi 50-flegrBee Wing

2,

Page 52: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

rEI Figure 28 (Continued)

0A

4~000

M2 - .01

0 4 0 0 4 6 12 16 210 24 26 32• 36

,,ANCILE OP A1'TACI( (OEUF10U•8

F•igure 28d - OCanard COon 50-Dgree Wing :

I

0 ,000

ii0 23--

- ANGLEO A.TTACK. (DEUMI "

Figure 28dL - Canard C 0 n 25-Dgres Wing

""8

• • • ,,•. :•... < "•r•• :•"'••' w'-- "•' lt L• .• "JM • ::•4•, ,:•,i • .•.q•¥•;• :-:••. b••.:l• •; • ,.:-•... .- .0,

Page 53: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Figure 28 (Continued)

0A -I: ~0.3 / -

~.0200

0,2•.1 'A

I t I I I Ia 4 8 12 1 0 24 m a

ANGLE OF ATTACK (DERESI)

ligurw 28f Ci.nard C2 on 25-1IXgr-e Wtng2

0,.4

(2)

0,10

0 II 1 24 32,., ANGLE OF ATTACK (DEUREES)

Figure 2Hg - Cniird C2 on 25-Degree Wing

39

Page 54: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

the 25-degree wing configuration. The data are for Lhe four canards at 0-

degroo ronivrd dcf lec tion. III gene.ral., the inc rementa I moments behave as

iexpu Li Ld ;I t low itng I es oF at t ac k, that Is, moving the canlard forward in-

(r'I Lhe hibi' itii nt~a momtflt I or ;i I coot igurahIlotlS4. At higher angiles

of attack, the effectiveness of the forward position drops off and the

incremental moment is Often l~ess than that generated by the canards at fur-

ther Lift positions. This is shown in Figure 29 where incremental, pitching

Figure 29 -Variation of AC with Canard PositionM32

0.5 P7 pl,2,3

04- P4,5,6

0.3

0.4

0.3

0.2 I I0.9 1.0 1.1 1.2 1.3 1.4 1.5 0.9 1.0 1.1 1.2 1,3 1.4 1.5 *qo/-Uka/I

Figure 29a - Moment on a 5O-Degree Wing

40

Page 55: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

F 5 __F_ I" ýIj gtrV2(CO I II nued)

010

0.24

0.3 . . . . . .

LV . g'rv 29~.b - )MV 1ilt. on a 29- Dug tLL WhI I

iiihl~utlit 1I 32-cdogrec' aiig I of LiILacLk 1,4 prusuanud . Anq heýu tic un, maxhi.min

IL [nremnwita I 1Moment OC'Cllr aL 0 IleMooL 1"orwaid powlL ion III 01n]y 0111. CALin

Llinit rcdi.u bo lug Llhe 60-dug rec duIe.1.1 canard C I0Oc~ ed II tho hlug lenL mooLl

For-wird poii)IP V M1X II'll 11unwimc.L In genural. OCcULITUC AL 1) for 11on CI

contig~r(.101 ne 011Ce [XUIIon to) lue ulu iiWas the 25-dugruu high aspeet

ratLi 0 canard C' whe rv minx imum momeicaL occur-rud In the ,[owumL most L )0H1N.1 -F ~Li on.* It [El In Le rvn ig Lo coniliiii cj lt hip the Flicof00Vii r t~t-11 (o31' l I ii]InIiiii

.lift ýoo ff1 ci nvt with unnoa d pansition sholvi II u F Lgurv 10 "Ind the, 11bul'var VI.-

iaLI COf 1 .ncerLemuulta.I 10MoiLllt .1.11 h Lbul Inn LIA1CUN L110 HIMlICHil 0o LAhe CurIIvL'

for uachI canard airo uxt rQt'me. S 1.11111,1r * T1IHa It appen rs Ltm L It IN thle A1 1.1. L

bu Ing gune rat:uid by thV L'Olan~t ri L ir than L he abso Latc oanard rl posi-L oil

which Is the primary do L-vrmlfn lg [actor ail the vmlumeo L gfunerated.

Re turn Lug Lo F'tgure 28 fL bi Heun tha t thW .InlI'' rilWietA It.Lp [LelLg 11101i1011

81.0 ) )C. I sl r eLi Ho nLI a .y I I i lu ll r w t I t1 41 ng.1. of 1 11 m i aIt u1Ll.I Ii to nlg 1. u i. f i at La i I t,

app ruximaituly 8 degrteus. ThIN vor: ation of ine.rumenL.u . plitching momilenlt

shlopc Is prc.Henteil II' Figure '30 f'or both 2'j- nflid 50-de0gree iilUdc.I N. III roi-

t rant wILhi the molonunLt data aLt 32-dog rue itngle of a L Lak , t iO voirint ion of

4 1

Page 56: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

ACM hilhaves in a linear manner, In that increasing moment arm increases

incremental pitciliing momlnat slope. As noted, lowering the canard rLduces

t he slope sli -;t'lv, T h'h dati from F igure 30 show a varying degree of slope

change with canard shape as expected, due to the diffcrence8 in canard lift

curve slope ard canard exposed area. The data from Figure 30 have been

divided by the isolated canard lift curve slope and plotted in Figure 31.

against canard exposed volume coefficient.

The plotted data fit a straight line reasonably well, thus indicating

very little upwash with canard position and that the linear approximation

C X V is reasonabl2i at low angles of attack.•T

Q.02 0 25-DEGREE WING r2

"" . % eS~Ij- ZC 2

0.02 -

I0.01 0

/C3

o . I t I I I , I , I I

0.9 1.1 1.3 1.5 0.9 1.1 1.3 1.5

Figure 30 - Incremental Moment Slope Variation

with Canard Position

42

I = "". .,

Page 57: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

0.3 FLAGGED SYMBOLS 25DEGREE

WING BASED ON POSITIONS 1,2,3

0 co '

VC1

I Dcl

0.2 C2C.

0.11olm

00.1 0.2 0.3

VCe

Figore 31 - Correlation of Neutral Point Change w:ithCanord Exposed Volume Coefficient

DEFLECTION

The vatiation of incremental moment with canard deflection is pro-

sented in Figure 32. Data are presented for both 25- and 30-0egree sweep

models and the four canard shapes. The canards are located at Positions

Sand 6 and the deflection range is from -10 to +10 degrees for the 50-

degree mode] and -10 to 0 degr.ees for the 25-degree model. When The ca-

nards are located in the high position P3, there Ls little change in incre-mental slope throughout the deflection range. An exception to this is the

25-degree high aspect ratio canard C3 at low angles of attack. This canardhas a varying incremental slope which is progressively reauced with in-

creasing canard deflwctiun. This reduction in slope is due to stall of

the canard.

43

-' .n- a ,1

Page 58: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

FI gorc '32 -I IUL MVentalJ. Moment Change due to(Xi iird 1h' f 1.ct ion

0.4 +. __ _ _ __ _0___ _ _

II.

S0.2 01

I~.j

0 4 a 12 16 20 24 28 3ANGLE OF~ ATTACK (DE(AREES)

Figue 3a CaardC 0 an 50-Dogree Wing

0,2

U.4 ,

.0 .

UI;,0,10

0.10

0.1 0 01

A oL p 01 ATAK(LPHE 1

Vigure~ ~ ~ ~~~~d 32 0nrdCo S-I0ge~W

44 1

Page 59: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

FI igure. 32 (Corit Inued)

0.1

1-1

00 a

Figure 32d00 - -~nr --n 50 e inG

450

Page 60: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Figure 32 (Continued)

-~L.0,3

0.10

0.2 ~ ~ - 00.0000Cow o o0

0.0

0 4 0 2 1 0 2 8 3

ANGLE OF ATTACK IDEGREBS)

Figure 32c Canard C2 on 25-Degree Wing

A, (DOAR46

Page 61: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

,'lgtirV '2 (Continued)

wo 0,- - d '- je m mqm m iam

.. ___ __ __ __ __I __ __ __, ___.__ __ ___ __ __ __0 4 . 0 4 .2 12 1 20 24 do 32

.,ANGLE OP ATTACK (OEORME8I)

Figure 32g - Canard C3 on 25-lI)egree Wing

While the data indicated little change in incremental moment slope

with deflection at Position 3, this is not the case at the lower positio•n

P At this position the inc~remental slope is decreasing with Increasingdeflection angle indicating that the interfer~ence between canard and wing

is no l~onger favorable or that the canard is stalled. This decrease, inmoment is seen more clear~ly In F~kgure 33 when incremental pitching moment

at 32-degree angle of attack is presented versus canard deflection angi.l~,i

At the high position P3 incremental moment increases with canard defloc- .

tion, however, at the low position P6the converse is true. This is most•notable for the 60-degree canard CI, and for the 45-degree high aspect ;i,

ratio canard C2. These canards have a reduction in incremental moment sL -

positive deflection angles. Similar trends were noted for the incrementol '

lift at 32 dlegrees due to deflection presented in the previous section,U The incremental lift presented in Figure 14 showed a reduction inlift of approximately 0.28 due to a positive 10-degree deflection for the

60-degree canard at P6. The incremental moment, however, does not indi-

cate a change of this magnitude, for, if all the incremental lift were due,

to the canard stalling, the expected change in moment would he AC1 * i /c

47

Page 62: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

N

t

L

I I•J I ° /-

too

-LLc

48

}:~

Page 63: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

-I'-.4'

NI %( ,�

0I4

I I 11� w/ ,�

I If' a�

I I H �S -�

- a a = - S Cd

rc�II Nj "1

(III II /1

'I 7/ S� ,�

I 8

49

Page 64: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

or approximately 0.28. Thus it appears that while the canard may have

suffered a slight reduction in lift (hence a reduction in moment) the major

I s -8 or I Ift Ls on tili' wing.

'the variation of control power C , with angle of attack is shown inM6

Figure 34 for both 25- and 50-degree wing models. The data are evaluated

from 0 to +10 degrees and from 0 to -10 depreeos.

At P3 no significant differences in'C occur if the value of C is3 M M 6

computed from either positive or negative deflections, although differences

do occur at low angles of attack for the 25-degree sweep canard C3 which,

as noted is due to canard stall. More significant differences occur at P6.

At P6' CM is higher ihen computed with the negative deflection than with

positive, thus indicating a possible canard stall. In general, there are

only slight differences in control power between the 25- and 50-degree

sweep models over most of the angle of attack range.

Control power was not as sunsitivw to canard position for the 25-

degree wing model as was the 50-ddgree sweep model. Control power for the

25-dogrw Hwvwvp model was bas.'d only on negativw deflection and thus the

IinfluVelce Of positive defleut•lon is not known.

WING LLADI.NU LAGE CFIANUJZS

Tlhe effect of a -9-degree leading edge droop on the 25-degree wing

model is preiiented in Figure 35. As fur the previously discussed lift, the

-9-dagree droop delays stall of the basic wing by approximately 4-degree

angle of attack. The incremental moments due to the canard, presented in

Figure 36, reflect this change in staLl hecause In the region between 8 to

20-degrce angle of attack the canard, located on the normal wing, has a

tslight increase in incremental moment when compared with the drooped leading

edge configuration, This increase in moment is not as large as that. which

would be expected from the incremental lift data, i.e., AC - AC /N L iC,

thus indicating that the primary moment change is due to delay of separAtion

on the wing rather than a lift increase on the canard.

50

........................................- ]=,.-R-

Page 65: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Sm ODEGREE DROOP0.101

0 4 8 12 18 20 24 28 32a, ANGLE OF ATTACK (DEGREES5

Figure 35 - Momont Characteristics of Basic 25-Degree Wing

cand 25-Degtee Wing withl -g-Degree Droop

DRAG

The primary aerodynamic influence of the close-coupled canard iA to

delay separation on the wing. This delay in separation results in a size-

able reduction in induced drae at moderate to high angles of attack. Thimreduction in induced drag is seen quite dramatically in Figures 37 and 38, ,

Figure 37 presents the variation of drag with lift coefficient for

the 60-degree delta canard C1 , the 45-degree high aspect ratio cninard C2 ,#

51I

Page 66: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

z 0.3

-- ODEGRqEE D~ROOP- --- 9.REE DROOP

02

a.;p

000j

20 24 23

~,ANGLE OF 4TTACK (DEGREES)

Pimure '36n - iomition 3

S 0.3

0.2 0

0 U. 00

0

4 12 16 20 24 2 32~cl, ANGLE OF ATTACK MEGREES)

F~igure 36b - Position 6

Fi1gure 36 -Incremental Momun~ts duv to Cainat~d onBasic and Modified 25-Dfegree Wing

52

Page 67: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

1.48 wc

1.0

I:: 25.DECIREE WING

0.4

0

0 0.1 0.2 0 0,1 0,2 0,3 0.4 0.8 0,6 0.7 0,e 0,9 1,0CD, DRAG COEFPICIMT

FLigro 37 - ypLcaL Dlrag CharntcteristicH

25- e.nd 50-degree mweep models and the canard is locatud at P 2. As seen in]

the figure the canard coilfigured vehicles have lemm drug at l~ift couffi-

rwspectivL'ly. F'lgUrr 38 prCL1LINt the correspondIng vnriutlon of lift-Lo-drgrati-o versus Lift coeffirient for the data presented In Figure 317, At4

noted in the figures, there is a decrease in maximum l~ift-to-drag ratio

(L/D) when the canard is installed. The magnitude of this reduction it-,max(L/D) in a function of ctinard planiform, position, and de'flection. ThLc

maxinfluence of tnese parameters on lift-to-drag ratio and induced drag wil~l

be discussed in the following sections.

Page 68: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

9

8 - W1'C 1

14

i-J3,

2

1 I

-0.2 u 0.2 0.4 0.6 0.8 t.o 1.2 1.4 1.6

CL, LIFT COEFFICIENT

Figure 38a - Lift-to-Drag Ratio for a 50.,Degree Wing

14

CW3

12 Q 0 W2+CAwg•c

10- w2+C,

2 S" /C/ /C

L 0, 1 .... I , I .... 1., t I L

-0.2 0 0.2 0,4 0,6 0,8 1,0 1.2 1,4 1.6

CL, LIFT COEFFICIENT

Figure 38b - ,.rL-to-Drug Rat Lo for a 25-flegree. Wing

N.1gure 38 - Typical. Lift-to-Drag Ratio Characteristics

54

"---. .... ..,Q 0, ------ ---

Page 69: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

SIZE

The daL-ta alV labl,, on thL1 Vffec t of carnard sixe noi drtig, nrL' HOtllIt411t

L imited. 0111y MW, WfilM -tt ull. [I 'Q rogoi; it I'INSII)MC Uva tI tiL('d ',iii;Ird 0IV,

as a parameter and the angle of attack was limited to 20 degreus, Slgnl.l 1.-

cant data regarding drag was obtained only with t:he two largent c•inards

having projected area ratios of 0.20 and 0.25. Certain trends, hIwvver,

can be obtained from these data. The first trend noted is the b3haviorof the drag developed on the canard due to deflection at zero lift coeffl-

cient. Figure 39 presents this variation of drag versus deflection for the

two canard sizes. As indicated in the figure, rhe shape of C0 veraus de-

flectLon Is approxhI [rt ly parabol[.c in shape suggemtLing that the da•t may

be analyzed in tho form of GD ( + K (C)2

Utilizing the zero Iift moment data due to dot -ect ion and dividing by

the correspondtrng canard diStaMcV ratio, GL 6 has been obtained. The vanr-

ation of C D with (G L )2 Im presented in Figure 40, Data have buen refur-

anced to the caniard LVXl1 Mx.d area. Vor up to approximaitely 1.5 dugrues dQ-

flection, tihe dati approximates a linear fit. At deflectLon LilgloUs grvaLur

than 15 degrees, the slope becomHs Steper indicating canard stall., It

thus appears that as with the incrementa.l. lift anid moment, drag duu to the

canard is a function primarily of expUHed area ratio rather than totnol aroa

ratio, This is further verified by Figure 41, which preseats t0he wvritat:ion

of aircraft lift-to-drag ratio at a lift coefficient (if .1.0) versus exposrd

area ratio. As shown, t~he change is linear albeit based on only throch datat

points, Data, however, from the 1'-4 •ircraft where can:vrd expoesed rut los

of 0.05 and 0.10 wore voaluatfed a1.s) exhibit this increae Lin I i

ratio with canard exposed area ratio. Thus, for at lertst up to 20-deg1ree,

angle of attack., a l.inear increase In l.ift-to-drag ratio with caeard vx-

posed area ratio appears to be valid.

55

W w

Page 70: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

'P2

. P5 P1 P3P7. ... . P4 -, pe

0.06 /

0 M42--

CLEAN

0 5 10 15 20 25 0 510152026 0 5 10 15 20 25b, CANARD DEFLECTION 6, CANARD DEFLECTION 6, CANARD DEFLECTION

ANGLE (DEGREES) ANGLE (DEGREES) ANGLE !DEGREES)

Fig-tre 39a - Drag for a 0.20 Ca,,

P2

-P' P1 P3

P- P4 P_

0.00

==0.04 •

0.02

0 5 10 15 20 26 2 5 10 15 20 25 0 5 10 15 20 256., CANARD DEFLEM.TON 6, CANARD DEFLECTION 6, CANARD DEFLECTION

ANGLE (DEGREES) ANGLE (DEGREES) ANGLE (DEGREES)

Figure 39b - D-ag for a 0.25 Canard

Figure 39 - Effect of Deflection on Zero Lift Drag CD DI56

-..------- .--~--.-.- - - --- ~ -j

Page 71: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

0.40 Sc/Sw a 0.26

O" S/S 0.20 I0.3 BASED ON TOTAL

EXPOSED CANARD AREA

0.2I

0.1 10 0,1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

(CL5 6)2

Figure 40 - Variation of Canard Drag Coefficientwith Canard Lift

3 P 3

0 7 WING BODY

LIFT COEFFICIENT - 1.0

0 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.18SC/Sw

Figure 41 - Variation of (L/D) with CanardExposed Area Ratio

57

Page 72: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

SHAPE

The effects of canard shape on maximum lift-to-drag ratio (l,/D)max and

lift-to-drag ratio at lift coefficients of 0.b, 0.8, and 1.2 are presented

Sin Figure 42. Lift-to-drag ratio is plotted versus the quarter chord sweep

angle of each canard 25" Data are presented for both 25- and 50-degree

sweep models, The tanards are located at P 3 at 0-degree canard deflection.

Also presented in the figure are the correspondiog lift-to-drag ratios of

the individual wing-bodies. Maximum lift-to-drag ratio is lower for the

canard configurations than for the wing-body alone for both sweep models.

This is to be expected since a penalty must be paid for the zero lift drag

increase due tu the canard. The penalty in lift to drag varies from 9

percent to 15 percent for the 25-degree wing model and from 4 to 7 percent

for the 50-degree model. These losses in (L/D)max can be reduced by proper

placement and deflection which will be discussed in following sections. As

angle of attack and thus lift coefficient are increased, the canard config-

urations have better lift-to-drag ratios than the basic wing-body. The

lift coefficient where this increase first occurs is approximately 0.7 for

tho 25-degree wing and 0.4 for the 50-degree wing model.

At low lift coefficients C < 0.6 the loss in lift-to-drag ratio isL

clearly a function of the quarter chord sweep angle as shown by the linear

variation. At hIgher lift coefficients this is not the case because the

amount of variation with canard quarter chord sweep angle is minimal for

the 50-degreu wing model. The canards on the 25-degree model exhibit a

nonlinear behavior with canard sweep angle at a CL of 0.8 (a w 10 degrees),

L/D increased with decreasing sweep angle up to X of 40 degrees and then

decreased. This behavior is due to the early stall of the 25-degree canard

C3 when located on the 25-degree wing model.

The improvement in lift-to-drag ratio is due to a reduction in induced

drag. Using the definition of drag as

CD C + K C

D~ I LO.L

CL

58

Page 73: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

IJ_-.-

25.DEGREE WING

13

L/DMAx WB

12

50-DEGRE t WING

L WING BODY

10 CL 0 'B WB W

9C L -0 -8 W F ,,• •.•,, '', •..'. . . . L/D M A x ,

e o -L/DMAX 9

C,7 7

"CL Os 0W ,CL 0'a

63 CL

CL~C 1.2BWB,

2Oc ClC3 c ?). C2 l , C012 C1

20 40 60 0 20 40 60

A, SWEEP (DEGREES) X, SWEEP (DEGREES)0.25 CHORD 0,21 CHORD

F i g~ u re 4 2 - L if t- Lo - D ra g [Ra L :Lo V a.rl a.t'.lo n w i th C n aa rd S ha pe -

59?

Page 74: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Figure 43 has been developed foir the four canard shapes on each model as

well n; each wing-hodv. The val ti l (i C T is approximately zero for each

tont i tIrat ion. At the value 0f CL [or (L/D))mx (c• '% 4 degrees), the in-

duced drag factor is higher than the basic wing-body for all configurations.

This is due to the unfavorable interference of the canard on the wing.

Then the angle of attack is increased, the canards have a lower induced

drag factor than the basic wing-body. This reduction occurs at an angle

of attack of approximately 8 degrees. No strong influence of shape on the

induced drag factor is seen for the 50-degree wing model,

Shape does influence the induced drag on the 25-degree wing. At low

lift coefficients, the 25-degree canard C3 had the lowest induced drag and

the 60-degree canard CI the highest. As angle of attack is increased, their

trends were reversed and the C3 canard had the highest induced drag and the

C1 canard the lowest.

POSLTiON

The eflect of canard position on maximum lift to drag ratio (L/D)max

and lift to drag ratios at Lift coefficients of 0.6 and 1.2 are presented

in Figure 44. Data are presented for both 25- and 50-degree sweep models.

Shown for eaich canard is the position where the canard exposed root trail-

Ing edge lnitially overlaps the exposed wing root leading edge.

For all configurations maximum lift to drag ratio occurs at a position

forward of" the exposed overlap position. In general the maximum value oc-

currud at P2, however, the 60-degree delta canard had a maximum value at P

for the 50-degree wing model. Lowering the canard reduced (L/D)max except

tL the cNIsu iof the 6(1-degree canard as mentioned above. A comparison with

dota for the basic wing-body shows only small losses in (L/D)max for each

canird if the canard Is at the optimum position. These losses in (L/D)max

were on the order of 3 and 7 percent for the 50- and 25-degree sweep models,respecLLvely. As Lift coefficient is increased, canard configurations have

Less drag than the basic wing-body for the 50-degree wing model. LifL to

drn; ratio In still optimized at positions forward of the canard wing over-

hip. As wlth, ([,/I))mx, lowering the canard reduced the lift to drag ratio.

Page 75: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Oigue 43 ,

0.H

0,7

CAI

D. . .| 10 1 ,4 N.N

0.6

CL., LIFT COEFFICIENT

Figure 43b - Drag for a 50-Degree Wing

G .9 o] .

0.1 '

! 0P3

0 I l 1 1 ,,' 0.4 0A8 O , , 1,4 1.0

UtL, LIFT COEiFFICIENI'D

Figure 43a Drag for a 25-Degree Wing

16

C, •

Page 76: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

i'tgurl2 44 -LI f L-to-Draig Ra"tio Variat [nWI Lf) Canard Posi8tion

1,2,3WING BODY

- - -4,5,0CANARD-WING OVERLAPL 37

CO9 19.0 - - w

8 - WO wo.

7 7.0

OCLfO 0, L '6 (86.0cc~

9 0

4 - co t4.0 (1

3 L-1.2 3.0 -

2 2.0 -CL .2

0.0 1 1

019 1.0 ill 1.2 1.3 1.4 1.5 0.9 1.0 1.1 1.2 -1.3 1,4 1.5

[U gurc 44a - Canard G on Figure 44b -Canard Co5 fl-Degree Wing 5 0-Degree Wing

62

Page 77: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Figure 44 (Contl-iued)

1,2,3 " WING BODY" P4 ,5, 6 CANARD.WING OVERLAP

SP7

10 10

-9 9

Fiue4 e- Cnr 2onFg r - Caad-3o

4A1

3 63

2 2

0 -- 01-I I0.9 1.0 1.1 1.2 1.3 1,4 1. 5 0.9 1.0 1.1 1.2 1.3 1.4 1.5

Figure 44c - Can rd C on F~igure 44d - Canard C on50-Degree Wing 250-Degree Wing3

63

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

Page 78: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Figure 44 (Continued)

P,3 3 WING BODY

O P6 CANARD.WING OVERLAP

WB L/Dm"

12 ~ I..DMAX

CL

/ /C3]

i 8 0 ,,

.c~

2.1

ii 4 -

S~CL - 0 ,e8

C' 0NOTECL 1,2 NOTOBTAINED ATPO .SITION 2

0.8 1.0 1.2 o's 0, 10 1,2 0,8 10 1.2

Figure 44e - Canards on 2 5 -Oegree Wing

i 6

6 4 :

Page 79: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

.:i . . .. . .. . = .i i • ..- ... - --- -- ------ ...- --V-WR-ZR'• • - • •¸' ' •

At CL 1.2, only slight changes occur in the lift-to-drag ratio, the loca-

tion for maximum value of L/D has, however, moved aft of Lhe overlap Junc-

ture. Lift-to-drag ratio at this C tended to be lower at the forward and

low positions.

The effect of position on the minimum drag coefficient C0 is presented

in Figure 45. Position has a minimal effect of C but moving the canard

and downward increased minimum drag. 0

As stated earlier the major effect of the canard on drag is to reduce

the induced drag component of the total drag. The effect of canard posi-tion on the induced drag factor KI, is presented in Figure 46 for the 50-

degree model. The influence of canard position on induced drag is somewhat

dependent on canard geometry. The 60-degree delta canard C1 has the small-est change in K due to position. Minimum K1 occurred at P over most ofthe angle of attack range.

Moving the canard aft increased induced drag. The three other canards

show a greater dependence on canard position. Maximum induced drag occurredat the most forward positions P1 and P4. Minimum induced drag occurred at

P2 for the 45-degree sweep canards C0 and C2 and P3 for the 25-degree canard

C 3 For all canards lowerlng the canard increased the induced drag factor.

Figure 47 presents similar data for the 25-degree wing model. More

variation of K1 with position is evident for the canards on the 25-degree

wing model than for the 50-degree wing model. At low lift coefficients K

did not vary significantly for the 50-degree model, whereas for the 25-

degree model these differences are more evident. At low CL P2 clearly has

a lower induced drag factor for all three canards. As lift coefficient Is

increased the induced drag factor at P increases to larger valijes than2

those obtained at Positions 3 and 6 for the 45- and 25-degree canards, C.

and C3. and Position 3 for the 60-degree canard CI. This increase in K

was relaLively large for both C2 and C indicating a large loss in effec-2 3

tiveness of the canard at P The increase in K for the 60-degree canard

was relatively small. Over most of the angle of attack range evaluatod,

lowering the canards to position P6 increased the induced drag factor.

65

Page 80: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

0.02 -• - --- -•_ _ __ __ _=- - -'

V 13

0.01 ,:aC p

MV{EGRIE WING

0.02 _________

0

II I I I

0,9 1.0 1.1 1.2 1.3 1.4 1,5 0,9 1,0 1.1 1.2 1,3 1.4 1,5

0.02 - _______

-, m a, -, 4 a

0 L I I . .p

25-DEGRfEU WING0.02 p

wepe•/ W8N

j 0.0 I I I

0.8 0, 9 1.0 1.1 1.2 1,3 0,8 0.9 1.0 1. 1 1, 2 1, 3

Fi1 00gure 45 - Effect of C.mird rloitihn on Minimum ,, ng Coeff-iient

ii Z•/ilI/I//r/f/fl// WE •/II,///z//'//H,/H~//b6W

Page 81: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Av 'tý -aFigure. 46 Induced Drag F~actor Variation with Canard

Position for tha 5O-Degrem Wing

0.4 ICAidAt4

0.2

0 00 0.4 nfl. 0A I. .O 1 2 1. 1.8CL0. UrTUM CIN

Figure 46a -CanardC

0.3 - --

Ug -

01

0 Cýt

0.1 D M 01".CI.

11. 1A 1. INI0(:L 0400 001? CIE11

Page 82: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

r •

Figure 46 (Continued)

Q.5

03

j !2 D U:o 0.1 1" j 1*4 1,0

Ct, LIF T tCULPPICILNI

lFigure 46c - Canard C2

or

0.4

. I , ,

PM, 01"C .i T I I t I

Figure 46d - Canard C3

68

Page 83: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

0.6

0.5

0.4

• 0,20,13

p6

0.2

01,1

0-

0.1

I L I I Io 0.2 0.4 0.6 0,8 1,0 1.2 1,4 1.0

CL, LIFT COEFFICIENT

Figur., 47 - Induced D)rag Factor V.riaition with CGnnrdPosition For the 25-De1gruo Wing

69

=MO .

Page 84: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Figure 48 indicates Lh1L minimum level of induced drag obtainable for

ecwh canard. The fivure is a Locus of the minimum induced drag factor

bas (I io ll' canard posit ions. Miininum induced dr;/, ka 1'btali:cd foIr tic

25-degree ciirard at low lift coeflicionts for both research models regard-

Less of canard position.

The 60-degree canard iad the highest value of induced drag at low lift

coefficients and the lowest value at lift coefficients near (C L)mx. T.e

Intermediate canards C0 and L2 have values between the 25- and 60-degree

canards. Thus, for goou low llft performanct! characteristics, i.e., range

and enduratnce, the low sweep canard Is best. When maneuvering capability

is; the dominant design factor, the highly swept canard gp.nerates the best

performance. The Intormediate canards are good compromises having lower

Lnduced drag 1han tile 60-degree delta at low lift cOeffic Lnts and slightlV

higher values of induced drag near maximum l ift. ExaminaLion of the figures

naoicates that while the range of Induced drag between the 25- and bO-degree

canards is not large for the 50-degree sweoip mo'-el, large differences in

induced draig pIus the peor stL:',l characteristics ux. ýbitad by the 25-degrec

canard preclude its use on tLhe 25-degree swept wing model.

Dh, IECTI ON

PossIt i ve caard def Iect ions ri:duce (L/D)1X s igni f icanitly, wh,._reas

small negative deflection improves (1/D)ml * This bch.iv.i or is illustrated

I.n Figure 49 which presents the varia-ion of lift-to-drag ratio versus ca-

nard deflection. DaLa are presented for Positicnv 3 and 6 for a canard

defl-cciion range frow -I( to +10 degrees in 5-degree increments for the SD-

aegrec 'WeLt IrOdc I and at -10 and 0 degrees for thlc 2g-,logree sweep model.

At btoth positions and for all canard S.a;lO;es, a l()-degvre deiflect[un

claUses a loss in (I,/D) of approx matel 40 percent. The ititliuenco ofZl.I~X

nc-gattivi canard deflection oi (LI/D) Max varied somewhat with canard posi-

ton 11nd shapc . For nearly all configurations on the 50-degree wing model,

a cans rd - leOction of -') dcgrees increased (lI/D) MI. lI'e sole exception

Lo Li, is was a sl ight decrease iii (LID)MAx when the 45-degree high aspect

riiti I rinar(I C2 waa; at 11

)70

Page 85: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

0,6

321 INCRIEA&INCANARD LEADING0.5 21EDGE SWEEP

0.4A

S0.3

01

S0 WEI

0.1 0 c.

00C

CC,

0.4 2 0.4 0. 0.8 1.0 1.2 1.4 1.6 1.eCL, LIFT COEFFICIENT

Figure 48a -Drag for a 50-Degree Wing

06-

, 0.4

0.2 INCHE CA LzIINCESN A~U I SWL

0 0.2 n,4 0.6 0.9 10 12 1.4:L, hF r COEPFICIENT

FigurLe 48b - Drag for a 25-i)grec Wing

Figurc 4M - Iocus of Minimum Induced Drag F'actor

71

Page 86: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Ft gurc 49 L- ft-LO-Drag Ratio Var Lat ionwi t h Canard I)ef I uv ti on

•.• L/DMAx WING BODY

10 10

I B L/DMAX88 8 - ' 8C)7 0 7

-- ix 6CL O.6

CL 0-0

0 0 c,4 4-~

CLI

2c2 - .2

" 0 2 .I 2 - P___-10 .0 10 -10 0 10

Sc, CANARD DEFLECTION 6C, CANARD DEFLECTIONANGLE (DEGREES) ANGLE (DEGREES)

Figure 49a - Canard C on Figure 49b - Canard C on50-Degree Wing 0 50-tlegr e Wing 3

IL

72

2111H I

Page 87: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Figure 49 (Continued)

10 10

___________ 9 -,L/DMAX

U -- •' "WING BODY

weo

3- 3

2 62

-10 0 10 -10 0 106c, CANARD DEFLECTION bc, CANARD DEFLECTION

ANGLE (DEGREES) ANGLE (DEGREES)

Figure 49c - Canard C., on Figure 49d - Canard C on

50-Degree Wing 50-Ihegree Wing

14

13 L-UMA%- WING 3ODY

12

iii A °"10

0 ,.........

C, C

2-

2 !

.10 0 -10 0 .10 0

A,, CANARD DEFLECTION ,• CANARD DCFLECTION A , CANARD DEIFLECTIONANGLI IDIOM$1) ANGLE InEGREM ~ ANGLE MDURFE81

Figure 49e - Canards on 25-1h'gree Wing

73

Page 88: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

A deflection of -I0 degrees caused a reduction in (L/D) for all

maxcanards with the 0x-teption of the 6U-degree delta canard C1. On the 25-

degru.e modul, negative 10-degree deflection caused reductions in (L/D))Mx

tor LlieU 25-dugrou C;i nard at both pusitions. The 45-degree high aspu•ct ratioI

canard C2 and a slight increase in (L/D)max at the low position P 6 and a

decrease at the high position P3' The value of (L/D)max was increased for

both positions of the 60-degree canard %ith negative deflections.

With Increasing lift coefficient the effect of canard deflections be-

comes less dramatic, Positive deflections still cause reduction in lift-

to-drag ratio yet the magnitude of these losses is reduced. Lift-to-dragratio w'is improved by a negative 5-degree canard on the 50-degree swept wingmodel.. Canard deficction had only a small effect on lift-to-drag ratio at

maneuverIng lift coefficients (CL>1.O). Slight losses do occur, however,

due to positive deflection. Negative deflections had little effect on L/D

at C[ - 1.2 on the 50-degree model. Negative deflections had a detrimental

eff'ect on lift-to-drag ratio for canards C1 and C2 on the 25-degree model.

'1h1 1 Lft-to-drag rat to wag slightly increased for the 25-degree canard C3

(in this model.

The losses in (l,/D) and L/D occurring at positive defluctions are

due to an increaue in minimum drag and an Lncrease in the induced drag fac-

tor KI. The slight gain in L/D uccurring at negative duflections is duL to

a decrease in The variation of C with canard deflection is shown in

Figure 50. The curves are roughly parabolic in shape and the minimum value

in general. occurs at 0-degree deflection. The curves, however, have a

slight bHas in that the drag coefficient for negative values of deflection

Is, In general, slightly less than that of the corresponding positive do.-

f lection. 'his bi.as may be due to the downwash of the canard when posi-

tively deflected causing an increase in drag of the wing.

the effect of canard deflection on the induced drag factor (K1 ) Is

presented in Figure 51 for deflections of -5, 0, and 5 degrees in the 50-

degree sweep model and -10 and 0 degrees for the 25-degree sweep model. As

IndicaLed In the figure a postLive 5-degree deflection increases induced

drag signiflcantly over that of the basic wing-body at low lift coefficients.

74

- - ~- -~- - --

Page 89: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

0.03 .-- P3• /

0.02

0.01 c

0 1 L, .I I. I .-10 0 -10 -10 0 -10 -10 0 10 -10 0 100c, CANARD0 DEFLECTION 0 c, CANARD DEFLECTION ",, CANARD DEFLECTION1

ANGLE (DEGREES) ANGLE (DEGREES) ANGLE (DEGREES) ANGLE (DEGREES)

Figure 50a - [)rag for a 50-Degree Wing

0,04

/3 P3_ -- --P| . BASIC

000 -10 0 -10 0 -10h., CANARD DEFLECTION h.,o CANARD DEFLECTION h.e, CANARD DEFLECTION

ANGLE (DEGREES) ANGLE (DEGREES) ANGLE (DEGREES)

Figure 50b -Drag for al 25-Degree Witng

Figure 50 - Minimum Drag Variation with Canard Defluction

75~

Page 90: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Figure 51 - Effect of Uanard Deflection on Induced Drag Factor

0.4

at

0 w

I I _.

I

i

i

O . 0.4 D' s I 1.0 , 1.4 1.6

CL, LIFI COEFFICIENt

Figure 51a - Canard C0 on 50-Degree Wing

0.9

0 A0 0.6 1.0 1 1.4

4j, UFT COMIPIUNT

F-igure 511 - inard C1 on 50-Degre Wing

76

_ __

__

__

Page 91: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Fiue51 (Continued)]

0.4 32J

0.2 0, . .

0.4 -

~,.IN

C3, LIFT COEFFICiENT

Figure 51.c C anard C~ on 5O-!Dugrue Wing

27

Page 92: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Figure 51 (Continued)

0.2 /

0,4 -

- ~ I

0,3 /0

-- I/

". 0,1 0

• //I

0 We ,

0.1 - o ooOi/

02

0,010 . -1o D I

,

0.1 ,do" • ow op -00lo0

0 ,02 0 0,2 0 0,12 014 Oe0, .8 O l 'O, 1,2 1.4CL, LIFT COIIFFICIENT

Figure 51a Canards on 25-Degree. Wing

78• . "h •I

Page 93: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

V]

T1he canard is carryfng much of the t.oad at these lift coofficlents andcl

thus has s gnficaL I induced drag of Its own. in addiLtion, the downwashll

from Lhisi hiLgh loading [,n modify ing the load distrir butimn if tid wt! g iLW :i F-g

unfavorable manner. Lift is being suppressed on the inboard position of

the wing causing a nonelliptical distribution. With increasing angle of

attack the favorable effects of the canard become evident and ;he induced

drag factor is less than that of the basic wing-body.

A negative 5-degree deflection of the canards reduced the induced drag

factor to values below the basic wing-body and O-degree deflection canard

configurations throughout most of the angle of attack range. Positive de-

flection configurations had higher values of induced drag than the O-degree

canard configurations. At high lift coefficients only small differences in

KI occur for either positive or negative deflections when compared to the

0-degree deflection canard configuration.

Negative 10-degree deflections on the 25-degree sweep model reduced

K1 at low lift coefficients. Negative deflection caused an increase In K

at high lift coefficients when the canards were located at 1' 'Thie hin-rease

in K did not occur to any extenL when the canards were located at the lowerLL position nor did it occur to the 25-degree sweet) canard C3. A possible

explanation of this behavior Is that the canard trailing edge gap may le

too laLte and thus the favorable interference effect from the wing may be

diminished somewhat,

WING LEADING EDGE CHANGES

In the earlier section on lift, it was stated that wing ,ludhIig edgo

modifiL.cations can have a beneficial effect on aircraft performanco. The

increases in performance, in gennral, take the form of an Increase In .lift-

to-drag ratio and increases in lift and stall angle of attack. I icrease!4

in lift and stall angle of attack did occur for the 25-dogrec wIng modul.,

howevor, leading edge droop and radius changes and little eiffc.t 1on 1l1e

50-degree wing. Leading edge droop increased .Ift-to-drag ratio r'or both

the 29- and 5U-degree swept wing models. The variation in L/I) with li Isprvsented In Figure 52. Data are presented for a -9-dugre droop with and

79

Page 94: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Figure 52 - Effect of -9-Degree Droop on Lift-to-Drag Ratio

1.4

1.4

',0.6

0.4 %%

- BASIC WB

•m BASIC WB PLUS CANARD

0.2 -"9DEGREE DROOP

-9 DEGREE DROOP PLUS CANARD

0 2 4 a 8 10

L/D, LIFT-TO.DRAG RATIO

I0igure 5ýa - Drug of 50-Owgree Wing

8o

Page 95: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Ftg~iri 52 (Continued)

1.4

1.29DERI DROOP PLUS CANARD 0DGE

1.0

3

WB PLUS CANARD %4%~ -U-OOEREE DROOP

0.4

U.6

8 0 -g DEOPEE DROOP PLUS CANARD 6c 0 DEGREE

-c-10 DEGREES6

0.8S WPLUS CANARD %~Cl P3

0,4

0.2 6c -10 DE001E1S

0 2 4 a 8 10 12 14 1LID, LIFT-TO.DflAG RATIO

I1l.gurvi 52.b - rag of 25-Ikigrei Wing

.81 ...... .....

Page 96: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

without the close-coupled canard install.ed. The data for the 50-degree

wing are based 0n the 45-degree truncated delta canard C0 at 0-degree de-

fI t''ct i[o ll ILoci t ed at I'P

The 60-degree delta canard Cl, located a P31 was used for the 25-degree

wing. The data are for deflection angles of -10 and 0 degrees.

Adding the droop to the 50-degree wing increased lift-to-drag ratio

for both canard on and off configurations throughout the angle of attack

range evaluated 4 < (i < 33 degrees. The incremental change in 1,/I) due to

the addition of the -9-degree droop is shown in Figure 53. At low lift co-

cfficients the droop improved the Lift-to-drag characteristics of the basic

configuration by AL/D ,1\.5 or approximately 14 percent. The gain in L/D

for the canard configuration with droop was approximately tO percent.

These improvements in lift-to-drag ratio remained fairly constant with in-

creasing CL for the canard configuration but were reduced for the wing-body.

The gains in lift-to-drag ratio due to the droop are far more impres-

slve on tho 29-degree wing model than those increases noted for the 50-

dugree model. Maximuin lift-to-drag ratio was increased rrom 12.5 to 14.8

when the droop was installed. The value of lift cuefficlent where (L/D) MIAX

occurred was increased for C from 0.3 to 0.54.

The iamoualt or increase in L/D due to the droop of the canard cont ig-

uritlon was dinepondnt on the canard dvf iection ongle as can be He0n In

Figure 53 which presents the change in L/D due to the droop for the three

configurationa. As shown, the gains in L/D exhibited by thc -10-degree

canard configuration are similar In shape and magnitude to those of the

basic wing-body. The zero degree canard configuration data has th saNme

gener'al shape as that of the -l-degree cAniard con figurati ton but the mag-

nitude or F the .increase INs ýignif.lcantLy small, r. A pialbl OI xpIl.anati n

for these differunces Is thatL the combination or the droolp and dOwnwnshI

froin the canard are having a detr, imental effect on the improvewd i ow cru-

antd by the droop. Similar rnsults were seen on the 11-4 when the inboard

Hi.at; and canaird were both IrisLaltted, 1')ef [iecctiog the cainard ngntfvL.*y,

8 2 .i

Page 97: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

AL/D DUE TO DROOP DR 9 DEGREES2401

II0.3 0,5 0,7 0.9 1.1 1.3 1.5

CL, LIFT COEFFICIENT

Figure 53,1 - hllL f t Drnr' fuor 50-I1)uogrutu W.hig,

VVB C1P3, DR 9 DEGREES

fie -10 DEGREES

6

[ 404 0DEGR~EE\

2

0.3 0.5 0.7 0.9 1.1 1.3CL, LIFT COEFFICIENT

,' 1urc 531h - II'IL L0 1)r '0 1ur 2 5-1I)L'IuL' WIulg

Ilr tiro 539 - I tIrLI' 101n tilI Chliugu In 11 L -L-lriag Rnt 1uLIM! lo -11-lgrovu I)ruoop

83

Page 98: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

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

however, will not change the wing flow significantly at low angles of a,-.

tack--hence, the similarity of improvements in lift-to-drag ratio for the

basic wing-body and the -10-degree canard configuration.

Thus it appears that it is possible to overdo the canard interference

at low angles of attack and the full performance potential of the canard

wing interaction will not be attained.

FLOW VISUALIZATION

The previous discussion presented in this volume has been based pri-

marily on force and moment data. A limited amount of flow visualization

data has also been obtained on the 50-degree sweep model in the angle of

attack range between 0 and 20 degrees. Tufts were installed on canards

and wing atid photographed with a motion picture camera. Results from these

motion pictures, at angles of attack of 10, 15, and 20 degrees, are shown

in the accompanying figures. The canards were, in general, located at P 6.

Sketches of the tuft directions for the basic 50-degree wing with

canard off are shown in Figure 54 for ct - 10, 15, and 20 degrees. At 10

degrees the basic wing has a region of flow near the body where the tufts

are in the streamwise direction (unseparated) and a leading edge vortex

which starts off the body. Increasing the angle of attack to 15 degrees

reduces the streamwise flow area and causes the wing leading edge vortex

to break down at approximately half the semispan. The outboard portion of

the wing is stalled. At 20-degree angle of attack the region of streamwise

flow is very small and most of the surface of the wing is in reverse flow,

The primary influence of the canard is to increase the area of stream-

wise flow (unseparated) and move the point where the wing leading edge vor-

tex begins outward. This increase in unseparated flow is seen in Figure 55.

Figure 55a shows the boundaries of unseparaued flov for the 60-degree canard

located at P6' Data are shown for canard deflection angles of -10, 0, and

10 degrees. Also shown on the figure are the corresponding boundaries for

the basic wing-body. As indicated, the area of unseparated flow is greatly

increased when the canard is installed; in addition, the vortex initiation

point is moved outward. The line of flow is approximately at the location

84

Page 99: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

'AIIsI

Is

Lsm

411

ID 4-4

Lr

85

. .- -

Lu3'I

Page 100: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

INIw *U

* wC3

LL

0

UU

LW wW to

an U

ItL6 8L

Page 101: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Ln

LU

Luu

LU.cc (DII'

LL

ccs-CD L

Cc t

87.

Page 102: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

Lu

LU

LLI LIUA

U, u 0n

o A:

4- 4,4.

LUn

a4 z

bo,4r344*

gj I* I

S W U.*~~~ a. *.

88

Page 103: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

of the canard tip for deflections of 0 and 10 degrees. Negative deflec-

Lions moved this location Inboard, whereas a positive deflection moved the

Iine outboard. rhe region of unseparated flow is relatively constant with

angle of attack for LhL 60-degree canard.

The same trend with deflection occurred for the 45-degree high aspect

ratio canard C and the 25-degree canard C as seen in Figures 55h and 55c.2 3

The separation point is, again, at approximately the canard tip but moves

rapidly inboard with angle of attack. Both of these canards exhibited early

stall on the upper surface and the region of unseparated flow tended to move

inboard as the canard stnll progressed towards the root. The 60-degree

delta canard did not exhibit any pronounced stall and the region of unsepa-

rated flow did not change to any great extent.

A comparison ul the three canards at zero-degree deflection is shown

in Figure 56. At 1.0-degree angle of attack the largest region of unsepa-

rated flow is due to the 25-degree canard with the 60-degree canard having

the smallest region. At 15 degrees there is little difference between the

3 canards. At 20 degrees the 60-degree canard had the greatest influence

on increasing the area of unseparated flow.

The 60-degree canard was also evaluated at PV Data from P2 are com-

pared with data at P6 in Figure 57. Moving the canard upward and forward

to P2 muved the vortex initiation point inwards, however, the surface area

of unseparated flow is relatively unchanged.The limited flow data are in agreement with the results obtained from

the force data in that the effectiveness of the higher aspect ratio canards

declines with increasing angle of attack, and that the 60-degree canardhas only minimal Loss in UffeCtIveness with increasing angle of attack.

SUMMARY

in the preceding analysis it will be noted that the presence of a

close-coupled canard modifies the aerodynamic characteristics of the basic

wing-body on which it is mounted. These changes, which are due to favor-

able interference, occur regardless of canard shape, size, position, or

wing planform. The aerodynamic changes consist primarily of an increase in

89

Page 104: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

stall angle of attack and maximum lift coefficient, and a decrease in drag

at angles of attack greater than approximately 8 degrees. The extent to

which these aerodynamic improvements occur is a function of 'anard size,

shape, positioun, and deficction. A summary of the effects of 'he z'bov-.

parameters is given below:

SIZE

I. Changes in lift and pitching moment at low-to-moderate angles ol

attack (a < 20) are proportional to canard exposed area ratio S /Sw andSC We

exposed volume coefficient S /Sw * c/, respectively, for canards of geo-C W C

emetrically similar planform.

2. Neutral point shift at low angles of attack is proportional to

exposed volume coefficient.

3. Incremental changes in lift-to-drag ratio are proportional to

canard exposed area ratio.

SHAPE

1. Lift is maximized by high sweep canards X nu 60 degrees.

2. The incremental moment characteristics of each canard shape were

similar in shape and magnitude to the isolated characteristics of eachcanard.

3. At low lift coefficients, incremental changes in lift-to-drag ratio

are proportional to the quarter chord sweep angle of the canard for canards

of equal exposed area.

4. Induced drag factor is reduced by the canard

5. Induced drag factor was lowest for low sweep canards at low lift

coefficients and highest at high lift coefficients. High sweep canards

exhibit opposite trends.

POSITION

1. Moving the canard longitudinally forward and downward reduces

maximum lift increments.

2. Maximum lift increments occur when the canard exposed root trail-

ing edge Is slightly forward of the exposed wing root leading edge.

90

Page 105: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

3. At low lift coefficients, moving the canard forward increases

increment~al moment. Maximum Incremental moments, in general, occur at the

sam vt' ) o t I o1n where max i mum I I f L occurs. 43(AcM)

4. At low angles of attack, incremental pitching slope - Is a

linear function of isolated canard lift curve slope multiplied by the ca-

nard exposed volume coefficient.

5. Lift-to-drag ratio was maximized at the same location where lift

was maximized.

6. Moving the canard forward and downward increased winimum drag.

DEFLECTION

1. Neither positive ncr negative deflections of the canard have any

significant effect on lift if the gap between canard trailing edge and wing

surface divided by canard span is greater than 0.1. Reducing this gap ratio

by positive canard deflection caused large reductions in maximum lift.

2. Similar reductions in Incremental moment occurred if the gap was

reduced. The reductions in incremental moment were not as correspondingly

large as the reductions in lift. These characteristics indicate a loss of

effectiveness of the canard on delaying wing itall.

3. Positive deflectioas cause a large increase in drag and a reduc-

tion in maximum lift-to-drag ratio.

4. The increase in drag with positive deflection is due' to an increase

in minimum drag and in the induced drag factor.

5. Small negative deflections (6c ' -5) can increase maximum lift-to-

drag ratio and reduce the induced drag factor when compared with zero-

degree deflection configurations.

6. The gains or penalties in lift-to-drag ratlo at large, lift coeffi-

cients C1 > 1.0 ýire small for either positive or negative deflections.

WING LEADING EDGE CHANGES

1. Increasing the canard-off stall characteristics of the 25-degree

wing by 4 degrees, delayed the favorable influence of the canard by a simi-

lar amount. Maximum incremental lift due to the canard was the same for

both basic and improved configurations.

91

Page 106: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

2. Incremental improvements in lift-to-drag ratio due to installation

of a -9-degree droop were approximately the same for both canard on and off

configurations of the 50-degree wing.

3. A similar -9-degree droop on the 25-degree wing model required a

-10-degree canard setting to obtain the incremental improvement due to the

droop.

4. Neither droops nor wing leading radii change significantly modi-

fied the benefits in lift, drag, or pitching moments due to the canard.

ACKNOWLEDGMENTS

The author wishes to thank Stephen J. Chorney, John R. Krouse, and

Jonah Ottensoser for their help in obtaining and evaluating the data pre-

sented in this report. Additional acknowledgment is given to James H.

Nichols and Dr. Roger J. Furey for their guidance and support.

92

Page 107: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

APPENDIX

MOI)DE C;EOMETRY

The data presented in this report are based on two research models.

The models consist of steel wings and a steel central core. Fuselages are

wooden fairings surrounding the central core. The canards and horizontal

tall are wood and fiberglass fairings built up around a steel spar. Attach-

ment of the canards and horizontal tail is provided by steel plates flush

with the fuselage. Seven canard and three horizontal tail mounting posi-

tLions are provided. Each canard can be rotated through a deflection range

from -10 to +25 degrees in 5-degree increments. Horizontal tail deflection

range is from -25 to +10 degrees. Rotation point for both canards and horl-

Zontal tail is 40 percent of the exposed surface root chord. Moment refer-

ence point for both research models is 0.27 c.

Detailed dimensions of the wings are given in Table 1. Table 2 pre-

sents dimensions of the four canards. Figure 58 shows the common fuselage

shape for both models. Wing planform geometries are given in Figure 59.

Canard geometry is given in Figure 60. Canard locations are presented in

Figure 61. A photograph of the various model components is shown in

Figure 62.

4l

93 Z S. X

Page 108: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

TABLE I - GEOMETRIC CHARACTERISTICS OF THE WINGS

WI(A - 50 [)'grees) W2(A = 25 D)Lgrees)

Alrfoii Section (NACA) * 64AO08

Projected Area, square inches 304 295

Span, inches 35.50 42.00

Chord, inches

Root (centerline) 15.38 12.20

Tip 1.90 1.90

Mean Aerodynamic Chord, inches

Length 10.30 8.30

Spanwise Location from 6.70 7.90Body Centerline

Aspect Ratio 4.15 6.00

Taper Ratio 0.12 0.16

Sweepback Angle, degrees

Leading Edge 50.0 25.0

Quarter Chord 45.5 20.0

Trailing Edge 23.5 -1.5

Incidence Angle, degrees 0 0

Dihedral Angle, degrees 0 0

Twist Angle, degrees 0 0

*64A008 Airfoil swept 25 degrees around 0.27c chord line.

94

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

Page 109: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

TABLE 2 - GEOMETRIC CHARACTERISTICS OF THE CANARDS

"C23

A M " Io I Scot oti ( NACA ) '4A0)08 04A006 64AO08 (14AU)%0

Exposud Area, square Inchcs 39.8 39.8 47.2 /49. 1

Projected Area, square inches 76.0 89.5 76.0 76.0

Exposed Semi-Span, inches 5.74 4.79 7.60 7.60

Total Span, inches 16.28 14.38 20.00 20.00

Chord, inches

Root (centerlLne) 8.73 12.45 6.70 6.12

Root (exposed) 6.33 8.30 5.31 5.00

Tip 0.59 0 0,90 1.,8

Aspect Ratio 3.50 2.31 5.26 5.26

Taper Ratio 0.07 0 0.13 0. 24

SweepbaCk Angle, degrees

Leading Edge 45 60 45 25

Trailing Edge 0 0 22.8 0

Dihedral Angle, degrees 0 0 0 0

S95

Page 110: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

NOTE: VERTICAL TAIL WAS NOT TESTED WITH THE25-DEGREE LEADING-EDGE SWEEP-WING (W2)

IA -- J

(a) '[up View

SECTION A-AD ALL DIMENSIONS ARE IN INCHES (CENTIMETERS)

WIDTH - 4,75 (12.06); HEIGHT = 4.15 (10,54)UPPER CORNER RADIUS = 1,00 (2.54)LOWER CORNER RADIUS 0,25 (0.64)

25.12(63.80)

45.62I (115.87)

(b) Side Vi•,w

Flgure 58 - R•search Aircraft Fuselage,

96

Page 111: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

IJ

1 10.77 (27.35)

SA 50 DEGREES

AR - 4.15" • " 0.12

ww

5.2(15.05)

p -

•\=25 DEGREESAR - 6.00

0.18

AIRFOIL:NACA 64AOWS W2

Figure 59 - Pianform View of the Wings

97

Page 112: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

I A *45 DEGREESIAR - 3.50 A -48 DEGREESx 1 0.07 AR - 5.26AIRFOIL: A a 0.13NACA 64A00a I AIRFOIL:[ NACA 64A00B

I \

tU

AR - 5.26I h - 0.24

AIRFOIL:A - 60 DEGREES NACAI4AFO

AR - 2.31 NAA

AIRFOIL:NACA 64AOO6

Figure 60 - Planform View of the Canards

98

Page 113: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

1614

Sito

994

a..-.

I94

Page 114: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

4 1I

I I

I II II II

I

II II

I i gLIrL' h2 - WI tilt Itiiiiit� I Mtitltj I ( (Hflj)Ot1(Ilt H

I

I I

II

Page 115: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

REFERENCES

1. Lacuy, I).W. and S.,J. Chorney, "Subsonic Acrodynamic Character-

istics of Close-Coupled Canards with Varying Area and Position Relative

to a 50-Degree Swept Wing," Naval Ship Research and Development Center,

DTNSRDC ASED-199, AD 882-702 (Mar 1971).

2. Krouse J.R., "Effects of Canard Planform on the Subsonic Aero-

dynamic Characteristics of a 25-degree and a 50-Degree Swept-Wing Research

Aircraft Model," Naval Ship Research and Development Center Evaluation

Report, AL-91, AD909-459 (May 1972).

J

10

103

Page 116: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

INITIAL DISTRIBUTION

Copies Copies

I OUSD (R&E)/Mr. Makepeace I MIT/Lib

1 DARPA/COL. Krone I N. Carolina State U./Raleigh Lib1 CHONR

704/R. Whitehead I Stanford U./LibNAVMAT 1 Virginia Polytech Inst/Lib

08T23

1 Boeing Aerospace Company/NADC H. Yoshihara1 Terry Miller (V/STOL

Project Office) I General Dynamics/Fort Worth/1 Bill Becker C.E. Kuchar

1 Carman MazzaI Grumman Aerospace Corporation/NAVPGSCOL Nick Dannerhoffer

I Lou Schmidt1 Lockheed-California/9 NAVAIRSYSCOM 2 Andy Byrnes

I AIR 03EI AIR 03PA 2 McDonnell Douglas Corp/3 AIR 320 St. Louis4 AIR 530 1 Jim Sinnett/MCAIR

1 Jim Hess/MCAIR12 DDC

1 Northrop Corp Labs/2 AFFDL I. Waller1 CAPT Bill Sotomayer1 R. Dyer I Rockwell International,

Columbus/LibNASA, HQ/Scientific Tech

Info Branch 1 Rockwell International,Science CenterI NASA Ames/Preston Nelms

I NVought Corporation/H. Diggers3 NASA Langley

I Joe Chambers1 Blair Gloss CENTER DISTRIBUTIONI Bill Henderson

Copies Code NameCalif. Inst of' Tech 10 5211.1 Reports Distribution

Grad Aero Labs 1 522.1 Library (C)

I U. of Cincinnati/Lib 1 522.2 Llbrary (A)

I U. of Maryland/Lib 2 522.3 Aerodynamics Library

105

Page 117: *10 cc DAVID W. TAYLOR NAVAL SHIP RESEARCH AND … · 2020. 2. 18. · *10 cc david w. taylor naval ship research and development center bethesda. md. 200134 aerodyna'mic cha.ractl..stics

DTNSRDC ISSUES THREE TYPES OF REPORTS

i. DTNiRDC REPORTS, A FORMAL SERIES, CONTAIN INFORMATION OF PERMANENT T'ECH,NICAL VALUE. THEY CARRY A CONSECUTIVE NUMERICAL IDENTIFICATION REGARDLESS OFTHEIR CLASSIFICATION OR THE ORIGINATING DEPARTMENT.

2, DEPARTMENTAL REPORTS. A SEMIFORMAL SERIES. CONTAIN INFORMATION OF A PRELIM-INARY, TEMPORARY, OR PR'IPRIETARY NATURE OR OF LIMITED INTEREST OR SIGNIFICANCE,THEY CARRY A DEPARTMENTAL ALPHANUMERICAL IDENTIFICATION.

3. TECHNICAL MEMORANDA, AN INFORMAL SERIES, CONTAIN TECHNICAL DOCUMENTA rIONOF LIMITED U4E AND INTEREST. THEY ARE PRIMARILY WORKIN,3 PAPERS INTENDED FOR IN-TERNAL USE. THEY CARRY AN IDENTIFYING NUMBER WHICH INDICATES THEIR TYPE AND THENUMERICAL CODE OF THE ORIGINATING DEPARTMENT. ANY DISTRIBUTION OUTSIDE DTNSRCMUST BE APPROVED BY THE HEAD OF THE ORIGINATING DEPARTMENT ON A CASE.BYCASEBASIS.