[ INVESTIGATION OF THE ELEVATOR

AND RUDDER HINGE MOMENT PARAMETERS

OF AN AIRCRAFT OBTAINED BT

ANALYTICAL, WIND TUNNEL AND FLIGHT

JOHN S. McNULTY
ROBERT N. SMITH
NICHOLAS J. VAfilAKOS
DUOLEVWWXUW^ SCHOOL
 AN INVESTIGATION OF THE ELEVATOR AND
RUDDER HINGE MOMENT PARAMETERS OF AN
AIRCRAFT OBTAINED BY ANALYTICAL, WIND
TUNNEL AND FLIGHT TEST DETERMINATION
by

Major John S. McNulty, USMC

Major Robert N. Smith, USMC
Lt. Nicholas J. Vagianos, USN

*^U Aeronautical Engineering Report No. 421

Submitted in partial fulfillment of the requirements

for the Degree of Master of Science in Engineering from
Princeton University, 1958.
 ACKNO W LEDGMENTS

The authors wish to express their appreciation to Professor

Courtland D. Perkins, under whose supervision and guidance this

investigation was carried out, and to Princeton University for the

equipment and facilities provided at the Forrestal Research Center.

The authors also express their appreciation to Mr. Thomas

E. Sweeney, whose valuable assistance and advice contributed

materially to the timely completion of the wind tunnel and flight test

phases of this report; and to Mr. R. F. Lehnert and staff for their

assistance in the wind tunnel phase and the construction of the

Navion model.

Thanks are also extended to Mr. Robert F. Cooper and the

hangar staff for their assistance in aircraft instrumentation and

maintenance of the test aircraft during this program.

a
 TABLE OF CONTENTS

Page

List of Tables iv

List of Figures vi

List of Symbols ix

Summary xi

Results xii

Introduction 1

Analytical Development 3

Equipment and Apparatus 18

a. Wind Tunnel 18

b. Aircraft Instrumentation 20

Procedure 24

a. Wind Tunnel 24

b. Flight Test 28

Discussion 39

Conclusion and Recommendations 47

References and Bibliography 49

Tables 51

Figures 69

iii
 LIST OF TABLES

1 - Physical Characteristics of the Navion

2 - Navion Tail Wind Tunnel Test. Run L Elevator Hinge Moment

Coefficients at Constant Angle of Attack

3 - Navion Tail Wind Tunnel Test. Run Elevator Hinge Moment

1L
Coefficients at Constant Angle of Attack with Wire at 07c of
.

Horizontal Tail

4 - Navion Tail Wind Tunnel Test. Run IIL Elevator Hinge Moment
Coefficients at Constant Elevator Deflection

5 - Navion Tail Wind Tunnel Test. Run IV: Elevator Hinge Moment
Coefficients at Constant Elevator Deflection with Wire at .07c

6 - Navion Tail Wind Tunnel Test. Run V: Elevator Hinge Moment

Coefficients at Constant Elevator Deflection with Wire at .07c

7 - Navion Tail Wind Tunnel Test. Run VI: Elevator Hinge Moment
Coefficients at Constant Elevator Deflection with Wire at .25c

8 - Navion Tail Wind Tunnel Test. Run VIL Rudder Hinge Moment
Coefficients at Constant Yaw

9 - Navion Tail Wind Tunnel Test. Run VUL Rudder Hinge Moment
Coefficients at Constant Yaw

10 - Navion Tail Wind Tunnel Test. Run DC: Rudder Hinge Moment
Coefficients at Constant Yaw with Wire at .25c on Rudder Fin and
Elevator

11 - Navion Tail Wind Tunnel Test. Run X: Rudder Hinge Moment

Coefficients at Constant Yaw with No Rudder Trim Tab and with
Wire at .25c of Rudder and Elevator

12 - Navion Tail Wind Tunnel Test. Run XL Rudder Hinge Moment

Coefficients at Constant Yaw with a Smooth Trim Tab

13 - Navion Tail Wind Tunnel Test. Run XII: Rudder Hinge Moment
Coefficients at Constant Rudder Deflection

14 - Navion Tail Wind Tunnel Test. Run XIII: Rudder Hinge Moment
Coefficients at Constant Rudder Deflection

IV
15 - Navion Tail Wind Tunnel Test. Run XIV: Rudder Hing Moment
Coefficients at Constant Rudder Deflection with a Smooth Tab

16 - Navion Tail Wind Tunnel Test. Run XV: Rudder Hinge Moment
Coefficients at Constant Rudder Deflection with Wire at .25c of
Rudder and Elevator

17 - Navion Tail Wind Tunnel Test. Run XVI: Rudder Hinge Moment
Coefficients at Constant Rudder Deflection with No Rudder Trim
Tab and Wire at .25c of Rudder and Elevator
 LIST OF FIGURES

Figure No. Description

Analytical

1 Sign Conventions

2. Angle O for Analytical Development

3. NACA WR A-ll (Fig. 2)

4. NACA WR A-ll (Fig. 4)

5. NACA WR A-il (Fig. 6)

6. NACA WR L-663 (Fig. 144)

7. NACA WR L-663 (Fig. 142)

8. NACA WR L-663 (Fig. 154)

Wind Tunnel

9. Diagram of Princeton Subsonic Wind Tunnel

10. Photograph of model in test section

11. Photograph under test section showing rudder

strain gage

12. Side view schematic of model

13. Effect of elevator deflection on horizontal tail lift

14. Effect of rudder deflection on vertical tail force

15. Ch vs 6e

16- Ch vs a

17. Ch vs 6r

18. Cu vs £

VI
Figure No. Description

Flight Test

19. Three view of test aircraft

20. Photograph of test aircraft, Navion N91566

21. Photograph of strain gage instrumented wheel for

measuring stick force

22. Photograph of 6train gage instrumented beam for

measuring rudder pedal force

23. Photograph of rudder pedal strain gage beam

installed in aircraft

24. Photograph of elevator deflection indicating

potentiometer

25. Photograph of rudder deflection indicating

potentiometer

26. Photograph of side slip chute

27. Photograph of side slip angle indicating vane

28. Drawing of sideslip chute installation

29. Stick force calibration chart

30. Elevator deflection calibration chart

31. Rudder pedal force calibration chart

32. Rudder deflection calibration chart

33. Sideslip vane calibration chart

34. Airspeed calibration chart

35. Stick force circuit diagram

36. Elevator deflection circuit diagram

37. Rudder pedal force circuit diagram

via
Figure No. Description

Flight Test

38. Rudder deflection circuit diagram

39. Sideslip vane circuit diagram

40. Elevator deflection vs velocity

41. Stick force vs velocity

42. P.F. vs 6r

43. P.F. vs p

44. Ch vs V cal#
6e

45. Ch a vs Veal.

46. Ch 6 vs p

47. Chp vs 6r

Vlll
 SYMBOLS AND CONVENTIONS

AR aspect ratio

a.c. aerodynamic center

a slope of lift curve

b wing span (ft.)

c chord (ft.)

Cjj hinge moment coefficient

Ch 3Ch/
p 8p
C
H 3 Ch /36

C}w *
Ch/g xt,
= - Ch s in wind tunnel tests

Cl lift coefficient

Fs stick force (lbs)

G gearing ratio (rad./ft.)

H.M. hinge moment (ft. lb.)

i incidence (degreec)

K constant = -G S e C e r| t (rad.ft.2 )

It tail length (distance from aircraft e.g. to a.c. of tail, ft.)

P.F. rudder pedal force (lbs.)

q dynamic pressure (psf)

S wing area (sq. ft.)

V velocity (mph)

W average gross weight of aircraft (lbs.)

IX
Greek Terms

a angle of attack

|3 sideslip angle

6 angle of control surface deflection

6 down wash angle

Tfy tail efficiency factor

p density of air (slugs per cubic ft.)

X taper ratio

T elevator effectiveness

trailing edge angle

a€ rate of change of downwash with angle of attack

da

Subscripts

c calibrated

e elevator

i indicated

r rudder

t tail, tab

v vertical tail

w wing
 SUMMARY

This investigation was conducted to determine the accuracy

with which it was possible to predict the flight hinge moment deriva-

tives of the elevator and rudder of a light subsonic airplane, using

both analytical and wind tunnel methods. The hinge moment deriva-

tives studied were those for the North American Navion N91566.

The tests were conducted at the Forrestal Research Center of

Princeton University, Princeton, New Jersey.

The values of the elevator hinge moment derivatives obtained

analytically, from the wind tunnel, and by flight testing were in

excellent agreement. The rudder derivatives obtained by flight testing

were slightly larger than the other methods. This was partially due

to the difficulty in producing large amounts of sideslip, and difficulty

in accurate scaling of rudder in the model.

XI
 RESULTS

The result6 of the hinge moments investigation are shown in the

following table :

Elevator Rudder
Ch Cha Ch6 r c hp
6e

1 Analytical (thin airfoil theory) -.0147 -.0073 -.0131 +.0058

2 Semianalytical -empirical -.0092 -.0040 -.0120 +.0045

3 Wind tunnel (without spoiler wires) -.0134 -.0043 -.0080 +.0039

4 Wind tunnel (with spoiler wires) -.0105 -.0030 .0070 +.0041

5 Flight test -.0090 -.0040 -.0160 +.0070

The derivatives on line 1 were based on thin airfoil theory, corrected

only for three-dimensional flow and neglecting other modifying effects, aD. of

which tend to lower the values. These derivatives can be considered to be an

upper limit as the physical airfoil approaches the theoretical thin airfoil.

Line 2 shows the results of a more exact semianalytical -empirical

method which includes all modifying effects such as three-dimensional flow,

section thickness, hinge gap spacing, flap leading edge shape, horns and

Reynolds number. Since all these factors reduce the absolute value of hinge

moments, the results on line 2 are lower than line 1.

The results given on lines 3 and 4 are for the wind tunnel model, run

first without and then with 1/16 -inch diameter spoiler wires on the stabilizer

surfaces. The wires were placed at the 7 or 25 percent chord positions and

XII
used to force a turbulent boundary layer. The results using the transition

wires were considered more representative of full scale conditions.

The results on line 5 were obtained from flight test. The elevator

derivatives were in fairly close agreement with the wind tunnel results, and

in excellent agreement with the semianalytical -empirical method. However,

there was less correlation in the rudder results. The rudder derivatives

obtained by flight test were larger than any of the methods predicted. This

may be partially due to a low order of accuracy of the actual rudder hinge

moment flight test due to the difficulty in producing large amounts of side-

slip with the two -foot chute used for this purpose.

A more thorough discussion of the results is presented in the

Discussion secition of this report.

xiii
 AN INVESTIGATION OF THE ELEVATOR AND RUDDER
HINGE MOMENT PARAMETERS OF AN AIRCRAFT
OBTAINED BY ANALYTICAL, WIND TUNNEL AND
FLIGHT TEST DETERMINATION

INTRODUCTION

It was desired to investigate the accuracy with which it was possible

to predict the hinge moment derivatives of the tail of a light subsonic air-

plane by both analytical and wind tunnel means. The airplane upon which this

investigation was performed was the North American Navion, whose normal

speed range is approximately 80 to 150 miles per hour. The airplane charac-

teristics are listed in Table I.

The hinge moment derivatives were predicted analytically by two

methods. The first was based on thin airfoil theory, which neglected nearly

all modifying effects such as thickness, hinge gap, elevator leading edge

shape, the presence of elevator balancing horns, and the Reynolds number of

the air. Three-dimensional effects, however, were taken into account. This

simplified analysis was, therefore, expected to overestimate the derivatives

on the high side. The second method was a semianalytical -empirical method

using wind tunnel data published by the NACA. This data was then corrected

for the geometry of the Navion tail.

The derivatives were also determined through wind tunnel tests of an

8 :1 scale model of the Navion tail mounted on a streamlined shape, Fig,. 12.

The wind tunnel used for testing the model was the two- by three -foot test
section of the Subsonic Instructional Wind Tunnel at the Forrestal Research

Center of Princeton University. The results from wind tunnel testing for

hinge moments are subject to many possible errors, such as wind tunnel wall

effects, inaccurate simulation of boundary layer conditions, inaccuracies in

model geometry, and friction in the moment measuring devices. Considera-

ble care was taken to rriinimize these errors in an attempt to obtain maximum
accuracy.

The actual Navion was finally instrumented to obtain the full scale

hinge moment data, and the full scale data developed from the appropriate

flight tests.

From a comparison of the results, it was hoped to be able to determine

the degree of reliability offered by the analytical and wind tunnel methods for

predicting control hinge moments in low subsonic flow.

This analysis was conducted during the Spring semester of 1958 at the

Forrestal Research Center of Princeton University, Princeton, New Jersey.

 ANALYTICAL DEVELOPMENT

The analytical derivation of the hinge moment coefficients was

accomplished by two methods. The first was a method based on thin airfoil

theory and the second a semianalytical -empirical method using published

section data and correcting the values for the Navion physical tail character-

istics as given in Table I.

The first method is one derived in Ref. 1 from thin airfoil theory and

is based wholly upon first order approximations. The theory completely

neglects airfoil thickness, hinge gap, elevator leading edge shape, the pres-

ence of elevator or rudder horns, and the viscosity of the air. Three-dimen-

sional effects were, however, accounted for. The resulting hinge moments,

therefore, were expected to be of larger magnitude but the method was in-

cluded in the analysis to show quantitatively the magnitude of errors involved

when the modifying factors are neglected.

The expression for elevator hinge moment as given by Ref. 1 is:

Mh 4p V 2 a 2 (T} 1 a + t!
2 5)

where *11 = f ( ch )
a
= f ch
^2 <
6 )

and "a" is the radius of the transforming circle and is approximately equal

to one -fourth the chord of the airfoil.

For control flaps where the ratios of flap chord to airfoil chord are

not small, the parameters rj. and r|2 are given below:
 = -G + cob0 cos
^1 sin t + 2 sin o - 2 o

o + 2 o cos bo - 2 G sin Co - sin 2 o

The angle o is shown in Fig. 2 and for an airfoil where the ratio

of flap chord aft of the hinge line to airfoil chord is 35 percent, the value

of G is 72° 32.5' or 1.266 radians. Substituting this value of ^ into

the previous two equations, gave the following values for n, and r^ :

r\
1
= +0.1684

ti
2
= -0.2423

The relations between tj

j , n^ and c na , c^g respectively were

obtained by comparing similar terms of the two moment equations :

Mh = 4pV 2 a 2 (t
U a +n 2 6) = l/2 p V 2 S e c e (e^a + c h6 6)

where for the left side of the equation :

a = c/ 4

ce = .35c

ce ce
thus a2 = 1 =
.1225 IT 1.96

and for the right side of the equation :

Se = ce x 1 for the area of airfoil per unit span.

Substituting the above values in the equations for the hinge moments, the

hinge moment derivatives were obtained :

 :

:
lW - T]l
= -16843 = +0.687 5 per radian
.245 .245

= +0.012 per degree

ch ^2 .2423
6 = - = -0.9885 per radian
.245 .245

= -0.0172 per degree

Using the normal sign convention of downward elevator deflections

being positive, Fig. 1, the sign of cu becomes negative.

The final two-dimensional derivatives were:

Ct, = -0.012 per degree

c h*: = -0.017 per degree

These two-dimensional hinge moment derivatives were than corrected

to three-dimensional flow using expressions derived from lifting line theory:

o ci

Ch ch T (C ha -c ha )
6

The three-dimensional slope of the lift curve for the horizontal tail

was obtained from Fig. 5-5 of Ref. 2 with the aspect ratio of 4.03. The

lift slope at obtained was .057/ degree. The three-dimensional elevator

effectiveness, T , was obtained from Fig. 5-33 of Ref. 2 by entering with

the area ratio S e / St

Se
14.098 =
~ 43.051 °- 327 5
St
The elevator effectiveness, T , obtained was 0.53. These values of

a and t were then used in the above equations :

Ch 6 -.0172 + .53 (-.0073 + .012) = -.0147

The final results for the elevator corrected to three-dimensional flow were:

ChQf = -.0073 per degree

Che = -.0147 per degree

The hinge moment derivatives for the rudder were also computed by

the method based on thin airfoil theory. In this case the rudder chord was

not a constant percentage of the total vertical tail chord. At the top, the

rudder was 42.6 percent chord, and at the horizontal tail centerline, it was

36.92 percent. An arithmetic average of 39.76 percent chord was assumed

as a mean aerodynamic chord. For a 39.76 percent chord, the angle O

was found to be 78. 10 degrees or 1.364 radians. The parameters rjj and ,%
were found to be:
Tij = 0.235

r\
z
= -0.320

For the rudder:

ce = 0.398 c and a = -E-

2 °e
thus a = * -IS.
0.158 ITT 2.529
 :

Therefore, the hinge moment derivatives for the rudder were obtained as

follows:

°h
a
=
"ors"
=
"OF = °- 743 per radian

= 0.013/ degree

cn
hR &
—0.316
Hi— = '°'l
0.316
Z(
j = -1.013 per
p radian

= 0.018/ degree

Converting to rudder nomenclature the rudder derivatives became:

ch = +0-013 per degree

cjjc = -0.018 per degree

The rudder derivatives were also corrected for three-dimensional

flow. Using an area ratio S r / Svt = 0.469 to enter Fig. 5-33 of Ref. 2,

a T =0.64 was obtained. The slope of the lift curve for the rudder was

obtained using formula 8-19 in Ref. 2 to obtain the effective aspect ratio of

the vertical tail

2 2
b
ARe = 1.55 -zr— = 1-55 (*'!??j. 1-943

From Fig. 8-8 of Ref. 2 the slope of the lift curve was found to be:

a^ = 0.043
 G

The three-dimensional derivatives were then computed as follows:

Cho
n (3 = c^
np —Cyp^" = .013
- 04 3
.0955
,. . = .0058

Ch 6
r
= c
H r
+ T
|- Ch
p
-<- c ^p)_

= - .018 + .64 ( - .0058 + .013 )

a -.0131

The final results for the "rudder corrected to three-dimensional flow

were

Chr = + 0.0058 per degree

Che = - 0.0131 per degree

The second method was the semianalytical -empirical method. Since

section data for the Navion tail sections were not available, it was necessary

to use empirical data for sections that closely resembled the Navion tail

sections and then apply corrections to obtain the final results. This was done

by two different procedures for comparison, since there is no one procedure

that is independent of assumptions.

The first procedure was to take the section hinge moment parameters

for the NACA 0009 airfoil with a 30 percent chord flap and .005 c gap from

Fig. 3. For a percent balance overhang cb/cf of 10.7 percent:

ch
a ~ -0.0065
c h6 = -0.0112
T e „a 8 +0.62
6
ao = ci = 0.094
i a
= 11 degrees
 The hinge moment derivatives were then corrected to an airfoil with

a 35 percent chord flap using Fig. 7. The following ratios were obtained:

ch (
= -0.0089
ce
= 0.35
c
Che -0.0138

ce_
= 0.30
c
0.0130
*6

Ratio s

89
na 75

138 , _,.
Ch = = 1 062
6 730" '

ce
Multiplying by the ratios to get ch^ and ch fi
for 0.35 _L!L
c

ch = (-0.0065) 1.187 = -0.0077

ch = ( -0.0112) 1.062 = -0.0119

6

Since the variation in the hinge moment parameters with airfoil section

arises mostly as a function of the included angle at the section trailing edge,

the following equations were used from Ref. 3, page 5, to correct data to the

Navion tail trailing edge :

Ach ff 0.005 a A0
Ac h& = 0.0078 a T
^
Although the included angle for the NACA 0012 section is 15 degrees, the

Navion tail measured only 13.5 degrees, due to having flat sided elevator
 10

surfaces. Therefore, A(j> was equal to 2.5 degrees compared with <j>
= 11

degrees for the NACA 0009 section.

Ach = 0.005 (0.094) (2.5) = 0.0012

Ach 5 = 0.0078 (0.094) (0.62) (2.5) = 0.0011

Applying the above correction in the sense that larger included angles

increase the derivatives positively:

c^ -0.0077 +* 0.0012 = -0.0065

c n6 = -0.0120 + 0.0011 = -0.0108

The three-dimensional flow corrections were the same as those used

in the thin airfoil method for the elevator.

a_t_ = c_ cL
Ct, = cv, L c h ^a,

ao

Ch 6 = c h6 + r (C ha - c ha )

Evaluating these expressions :

°'° 5
Ch = -0.0065 ( n ] 0.0040
a V 0.094 I)

Ch = -0.010C + 0.53 -0.0040 - (-0.0065) |

= -0.0094
 11

Since the elevator is equipped with horn balances, the hinge moment

derivatives must be corrected for the decreasing effect of the horns. This

was done using Fig. 8 entering the plot with the ratio:

Area horn x mean chord of horn ( 0. 16 13 ) ( 0. 646 )

Area of control x mean chord of control (7.05) (1.28)

°- 10
=
n
9.02
y = 0.0155

The AChQ, va* ue was insignificant but the A Ch* value was + 0.0002.

Applying this correction, the final results for the three-dimensional hinge

moment derivatives were :

C ha = -.0040

Ch 6 = -.0092

The second procedure used in the semianalytical -empirical method

consisted of taking the hinge moment derivatives for the NACA 0009 and

0015 airfoil sections, since data for the NACA 0012 airfoil were not avail-

able. This data was for a 30 percent chord flap, a hinge gap of 0.005c

and a 0. 107 c overhang of the elevator ahead of the hinge line. The value

of 0. 107 c was selected as the average value along the span of the elevator.

The values were obtained from Figs. 3 and 4 and were as follows :

NACA 0009 NACA 0015

c h<* -0.0063 -0.0019

ch 6 -0.0112 -0.0054

c
lct
0.094 0.092

T = a 6 0.62 0.48

4 11° 19°
 12

A linear variation of the parameters was assumed between the 0009

and 0015 sections and the following values were obtained for the NACA 0012

section:

ch a 8 -0.0041

ch
6
= -0.0083

c l« = 0.093

ot 6 = 0.55

S 15° (actually measured as 13.5°)

The trailing edge included angle was measured to be 13. 5° since the

elevator surfaces are flat sided.

The derivatives cj, and cw were then corrected to a 35 percent

chord elevator using Fig. 5 and the following values were obtained:

c hQ ,
= -0.0052

ch s -0.0090
s

The derivatives were then corrected for the difference of trailing edge

included angles which was 1.5 degrees. The corrections were:

Ach = 0.005 (0.093) ( 1.5) * .0007

Ach 6 = 0.0078 (0.093) (0.62) (1.5) = 0.0007

and applied so as to increase the hinge moments negatively, since the trailing

edge angle was corrected to a smaller value. The corrected two-dimensional

hinge moment derivatives were :

ch a = -0.0059

c h6 = -0.0097
 -
13

The derivatives were corrected to three-dimensional flow by using

the same expression as was used in the previous procedures :

C L<X
a

Ch 6 = c h& + r (Chft - c hQ()

Inserting the numerical values :

Ch
«
= •- 0059
(w) = - - 0036

Ch 6 = -.0097 + 0.55 [-0.0036 - (-.0059)

= -.0084

Applying the horn balance correction of A Co, = 0.0002, the final three
6

dimensional hinge moment derivatives were :

Ch a = -0.0036

Ch = -0.0081
6

This method gave results approximately 10 percent lower than by the previous

method.

The hinge moment derivatives for the rudder were calculated by the

first procedure. The rudder does not have the simple geometric relationships

that the elevator has ; therefore, it was necessary to take average values for

the section characteristics.

 ) )

14

The vertical tail characteristics at the top plane of the rudder were

as follows:

Airfoil section NACA 0012.04 Mod

Airfoil chord (not modified) 19.779"

Rudder % chord ( not modified ) 28 .9 %

Airfoil chord ( modified 24. 66S 1"

Rudder % chord ( mqdified ) 42. 9 %

Percent overhang to rudder chord 5. 63 %
Rudder chord 10. 6 '«

included angle at trailing edge 8. 8°

The characteristics of the vertical tail at the bottom hinge of the

rudder were as follows :

Airfoil section NACA 0013.2 Mod

Extended airfoil chord (not modified) 47.23"

Rudder % chord ( not modified 29.75%

Extended airfoil chord (modified) 53.45"

Rudder % chord (modified) 37.07%

Percent overhand to rudder chord 6.69%

Rudder chord 19-8"

Included angle at trailing edge 10.3°

 15

The section characteristics for the rudder were estimated as the

arithmetic mean of the rudder top and bottom sections. The resulting values

obtained were:

Airfoil section NACA 0012.5 Mod

Airfoil chord 39.057"

Rudder % chord 39.99 %

cb
Percent overhang to rudder chord = 6.16 %
cf
Rudder chord 15.2 "

Included angle at trailing edge 9.05°

The section hinge moment parameters for this averaged section

obtained from Fig. 3 were as follows:

ch - -0.0063

chc -0.0130
o

r= a 5
= +0.608

c, = 0.095
1 a
4>
= 11 o

The hinge moments derivatives corrected to an airfoil with a

39.925 percent chord flap using Fig. 7 were:

C h<* — -0.0101
ce
= 0.40 =
c Ch -0.0145
6

Ce
-0.0075
= 0.30 ~
c C -0.0130
*6
 16

Ratios

ch
n = = 1.347
a 75

= 145.5 _ .
11n
Ch 1 119
6 -130- *

Multiplying by the ratios

cha = ( 1.347) ( -0.0067) = -0.0091

c h& = ( 1.119) (
-0.0130) = -0.0145

Correcting for the trailing edge included angle:

Ac h a = 0.005 a Q A0

= 0.005 (0.095) (- 1.95) = -0.0009

Aa.o = 0.0078 a Q T A0
0.0073 (0.095) (0.608) ( -0.95) = -0.0009

Applying the above corrections :

ch -0.00910 - 0.0009 = -0.0100

a
c = -0.0145 - 0.0009 -0.0154
hs

The three-dimensional flow corrections were the same as those used

in the thin airfoil method for the rudder.

The three-dimensional derivatives were derived as follows :

Ch
a =
c
ha ^L =
(-.0100) _L<!i! = -0.0045
a .095

ch 6 = c h6 + T C hQ, " ch
<
a )

= -.0154 + .64 (-.0045 + .0100)

= -.0120
 17

Rewriting the above using rudder nomenclature the results for the vertical

tail were :

C ho = -C h = 0.0045

Ch 6 = C h& = -0.0120
 )

18

EQUIPMENT AND APPARATUS

Wind Tunnel and Wind Tunnel Model

The wind tunnel used for running the model was the 3-D side of the

Subsonic Instructional Wind Tunnel located at the Forrestal Research Center

of Princeton University. The test section of the 3-D side is a rectangular

closed jet section, two feet in height and three feet in width. This side is

capable of velocities up to approximately 97 miles per hour. (See Ref. 4

and Fig . 9 •

The wind tunnel model was an 8: 1 scale model of the horizontal and

vertical tail surfaces of the Navion, mounted on a streamlined shape. The

streamlined shape retained the original fuselage cross -section characteris-

tics for approximately two horizontal tail chord lengths and then faired into

a rounded nose, as shown in Fig. 10. The elevator and rudder were hinged

with ball bearing hinges to avoid any influence of hinge friction on the meas-

ured hinge moments.

The elevator could be set to any predetermined angle by the turning

of a rod accessible at the nose of the streamlined body. The aft end of the

rod was attached to a sliding mechanism upon which a small strain gage beam

was attached. A link rod connected the other end of the beam to a lever arm

inserted into the elevator hinge line shaft at the model center line, Fig. 12.

Moments applied to the elevator were transmitted by the lever arm through

the link rod to the strain gage beam where they were measured by the strain

gage. By specifying a strain gage beam of as short a length as practical,

 19

minimum deflections were encountered, which in turn kept elevator angle

deviations to a minimum, as the aerodynamic loads were applied.

The rudder was controlled with a torque tube extending downward

through the center of the turntable upon which the model supporting strut was

fastened. A strain gage beam attached to a shaft at the lower end of the tube

was calibrated to detect the rudder hinge moment. The rudder angle could

be set by rotating the beam and tube to different angles and locking them at

the desired angle, Fig. 11. Two universal bearings at each end of the tube

permitted the rudder moment to be obtained at different angles of model pitch.

The model was mounted on a turntable of 27 -inch diameter by a

streamlined hollow steel strut of nine -inch length. The model had only

freedom of pitch with respect to the supporting strut. Pitch was controlled

by adjusting the length of a supporting turnbuckle attached between the after

portion of the body and the turntable.

The model could be yawed about the rudder hinge line, by rotating the

turntable. The rudder hinge line was vertical and passed directly through

the center of the turntable when the model had a zero pitch angle. At any

other pitch angle, the misalignment caused a tilt of the rudder hinge line,

which was corrected by the universal bearings on either end of the torque

tube.

The two strain gage beams were connected to a Carrier amplifier

which provided the bridge circuit and balancing components for the strain
 20

gages and, in addition, amplified the signal to provide the necessary sensi-

tivity. The hinge moments were thus read in terms of microamperes on the

meter attached to the amplifier and converted to moments by using the

calibration data.

Aircraft Instrumentation:

The airplane used for the flight tests was a "Navion" manufactured

by North American Aviation, Inc\ This airplane is a four place, low wing

monoplane, powered by a single Continental E 185, 205 horsepower engine.

Fig. 19 and Fig. 20 show a three view drawing and photograph of the aircraft.

The pilot stations are situated side by side and each is equipped with a set of

rudder pedals and a control yoke. The two aft passenger seats were removed

and a wooden platform installed to mount the necessary instrumentation

components.

To obtain the required flight test data, it was necessary to be able to

vary the center of gravity position and to provide a means of producing a

variable amount of sideslip angle. It was also necessary to instrument the

airplane so that it was possible to read stick force, elevator deflection,

rudder pedal force, rudder deflection and sideslip angle.

The center of gravity was shifted fore and aft by using a sliding weight

inside a tube. Basically, the system consisted of an 83 lb. 14.5 oz. cylinder

of lead sliding in a 20 -foot aluminum irrigation pipe of four -inch diameter.

The movement of the weight was accomplished by means of a rachet arrange-

 21

ment at the forward end of the tube driving a 3. 5 -inch sprocket over which

ran a 1/8-inch bicycle chain. A similar sprocket was mounted at the rear

of the tube and both ends of the chain were attached to the weight. Full

travel of the weight produced a change of center of gravity of approximately

nine percent m.a.c.

To produce the sideslip, a pilot chute from a standard Navy 28 -foot

parachute was used. The chute was attached to the right wing tip with a

10 -foot nylon shroud line and a release mechanism which could be actuated

from the cockpit. A second line was attached to the chute and led inside the

fuselage through the exhaust port above the right wing root to an electric

powered winch. The winch controlled the lateral position of the chute and,

thereby, the amount of yawing moment required to produce sideslip. The

chute, having a diameter of approximately two feet, produced a change in

sideslip of approximately three degrees when the chute was moved from its

full out to full in position. Although a larger change in sideslip was desirable

from the standpoint of increased accuracy and could be obtained by using two

chutes, it was not considered safe for take-off, since the take-offs were per-

formed with the chute fully streamed. It was not possible to collapse the

chute for take-off because the built-in spring designed into the chute prevented

its being packed into a can.

The stick force was obtained by replacing the standard yoke by one

which resembled an elongated "H M , with the hub at the center of the cross
 -

22

bar, Fig. 21. On each half of the cross bar were placed two strain gages.

These measured the bending moments applied to the bars by the pilot's

hands on the vertical members of "yoke handles." The strain gage circuit

was so arranged as to yield the total force regardless of the distribution of

forces on the two handles. The force indication was displayed on a micro

ammeter. Since no amplifier was used due to weight and space limitations,

it was necessary to design the required sensitivity of the meter indication

into the system. The system was then calibrated with a spring scale. A
calibration plot is shown in Fig. 29 and a wiring diagram of the circuit in

Fig. 35.

The elevator deflection was obtained using a low friction potentiome-

ter in the tail, which rotated with elevator movement, Fig. 24. Changing

the potentiometer setting changed the reading of a microammeter in the

cockpit. After proper calibration, it was possible to obtain elevator deflec-

tion. A calibration plot and wiring diagram are shown in Figs. 30 and 36

respectively.

The rudder pedal force was obtained by using a single cantilever

beam mounted on the copilot's left rudder pedal and extending to the right.

The beam had a bolt extending ait, perpendicular to the beam to insure a

constant point of foot force application, Figs, 22 and 23. The strain gages,

mounted on the beam, measured the bending in the beam. It was only

necessary to have a beam on the left pedal, since the parachute on the right
 23

wing tip always created negative sideslip, requiring left rudder to counter-

act it. A calibration plot and wiring diagram are shown in Figs. 31 and 37

respectively.

The rudder deflection was measured in the same manner as the

elevator. A photograph of the potentiometer installation, a calibration plot,

and wiring diagram are shown in Figs. 25, 32 and 38 respectively.

The sideslip angle was obtained by using a vane on a boom attached

to the left wing tip, Fig. 27. The vane rotated a low friction potentiometer

inside the boom which gave indications on a microammeter in the cockpit.

Similar calibration procedures yielded the calibration plot in Fig. 33. The

circuit diagram is shown in Fig. 39.

In order to insure accurate velocity measurements, the airspeed

indicator was calibrated by making timed runs over a measured course.

The airspeed calibration chart is shown in Fig. 34.

 24

PROCEDURE

Wind Tunnel Testing

The model was tested in the 3D side of the Student Instructional

Wind Tunnel at the Forrestal Research Center of Princeton University.

The model was run at dynamic pressures as high ao 24. 1 pounds per

square foot. This provided a Reynolds number of approximately 380,000

based on the horizontal tail chord, which was considerably lower than the

2, 800, 000 average at which the airplane flight tests were taken.

The equations used in reducing the data were derived from the hinge

moment coefficient equation:

Ch = C h6 6e + C h(J a
q
™ e

where HM = hinge moment in foot -lbs

q = dynamic pressure in pounds per square foot

Se = area of the elevator behind the hinge line

ce = root mean square chord of the control surface

aft of the hinge line

Taking the difference between two different 6e and a positions,

the following equations were obtained:

ahm 2
ACh = Che6 A 6 1-2 + Ch Aa 1-2 =
a qS e c e

Letting A0?j.2 = 0, and solving for Chg

c h6 -
AHM '-*
q Se c e * 6 l-2
 25

Substituting in the appropriate conversion factors and dimensional

values for the model:

_
-.0003215 A grams
b _
cn
6 ( .22028) (. 14375) q A6°
- .010153 A grams
q ^6p=
'e
(

A similar derivation letting A8 e] - = produced the following

equation for Q15 :

~ "
-.010153
"
Agrams
B "
Ml c
6
1 A<2°

Using the values for rudder area and chord

S = .09456 square feet

c « .1583 feet

the following equations were derived for the rudder:

-. 0184 A grams
Ch
*r 1 A6 r °

+ 0184
. a grams
Ch
P <1
AP°

The wind tunnel data obtained from the test runs were tabulated in

Tables II through XVII. Wind tunnel wall corrections for change of induced

angle of attack and streamline curvature were calculated as follows:

01
final = ^test + AOj + T 2 AO^
 26

where Affj = 6 —g
CL

S = horizontal tail area

C = test section area

The value for 6 was obtained from Fig. 6:30 of Ref. 5. Using k = .4385

for the horizontal stabilizer, 6 was found to be 0. 115. Therefore:

.6726
Aa t = 0.115 ^
5814 = .0133

The value of T £ was obtained from Fig* 6:54 of Ref. 5 using It/ B of

.03478. The value found for T ^ was 0. 10. The complete expression

becomes:
a a
final = test + .0133 C L + .00133 C L

"final = "test + .0146 C L

The corrections for the rudder were obtained in a similar manner.

Using :

6 .133

T = .08
2

S = ,2018 square feet

C = 5.814 square feet

the following equation for the correction was derived:

P final = Ptest + - 005 C L

Both these corrections were extremely small and would not alter

the results; therefore, both were neglectec. Had a larger model been used

in the same test section, the corrections would have become more significant
 27

Amplifiers were used to amplify the signals from the resistance

wire strain gage beams. The two systems were calibrated using weights

acting on the control surfaces at a known distance from the hinge line6.

The elevator and rudder deflections could be read during the run

in order that changes in deflections could be detected and recorded properly.

To force a turbulent boundar layer considered more typical of the

full scale flight condition 1/16 inch wires were taped along the span of the

horizontal and vertical stabilizers at seven percent chord for Runs n, IV,

V and at 25 percent chord for Runs VI, DC, X, XV, XVI. Below is a

diagram showing location of turbulence wires:

Wires at 7% chord Wires at 25% chord

Upon the completion of the testing, the model was removed and the

test section was calibrated with a pitot static tube located at the centerline

of the tunnel.
 .

28

night Testing

The flight test procedure used to obtain the data for the elevator was

the same as was employed in Ref. 6. The method is based on the following

development

Elevator Derivatives by Flight Test

ChV
The Cftc derivative for the elevator was obtained first. The total

hinge moment coefficient of the elevator can be written as:

Ch = C ha <\ + C h6 6e C 6t
h&t

If the airplane is flown in steady level flight and without changing the speed,

power, or tab setting, the center of gravity is shifted in the longitudinal

direction, a change in elevator angle and stick force will occur due to the

change in lift now required from the horizontal tail to maintain level flight.

The slight change of wing lift required, which results in a slight change in

wing angle of attack and, thereby, tail angle of attack, AG t

, will be

neglected. In addition, the tab was 6et at zero deflection on the ground and

not changed during the flight tests, and changes of tab angle, A 6^ , were,

therefore, zero. The tail efficiency r^ was assumed to be a constant,

1.0, over the speed range.

Writing the hinge moment coefficient expressions for two different

center of gravity positions, and subtracting one from the other:

 L')

Ch = ch + Ch 6 6 ei + Chs 6t
2 a «ti

% « C hQ? a t2 + C h6 6 eE + C h6t 6t 2

AC hl _ 2 Q^ AQtj.2 + C h6 ^6 ei . 2 + C h6t A6 tl _ 2

Since A^, 2
is essentially zero and A6 t , , is zero, the expression

reduces to:

The derivatives Che can then be expressed as:

* 6
ei_2

The hinge moment coefficient can be found from the expression for

stick force:

Fs = -G • KM
= -G ( Ch q S e c e tit)

Solving for Ch
- s
Ch s
-G q Se ce It

The change in Ch is then given by:

AC h 1-2 -G q Se c nt
e

•.*

s,
"1
-
s-
-391 t - *2 )

G Se ce T!
t
VcaJ
 30

Substituting the values for the Navion in the above equation, the expression

for ACjj becomes:

-23. 1
ACh i-; ( F 8l - F 82 )

Veil

The equation for C^ . then becomes:

-23. 1
ch
2 6 °
Veal ».f e2

Stick force and elevator deflections were recorded for two e.g.

positions, the sliding weight full forward and full aft, for various speeds

from 80 to 130 mph. The values were then substituted in the last equation

and a plot of Cu versus calibrated velocity was produce, I"ig. 44.

6

Ch
a

The Cjj derivative was obtained next by flight testing. The

derivation of the flight test procedure again depends on the difference of

hinge moment coefficients at two different flight conditions. Rewriting the

equation for change in total hinge moment coefficient:

^c hl _2 = c ha ^ tl .2 + ^5-^1-2 + Ch &
t
~ 5t i-2

Again the trim tab was set and left at zero, making A 6ti_2 = °-

Solving the last equation for Ch q,:

ACh 12 _ Ch 6 ^6 ei _ 2
l
Q? A «.
1-2
 31

The flight testing was conducted by flying the airplane at a constant power

setting, tab setting and center of gravity location. At one velocity, the

stick force and elevator deflection were measured. The velocity was

changed by going into a slight glide or climb and the new stick force and

elevator deflection were recorded. This provided information from which

C^ was evaluated at some average velocity, using the previous equation.

In this case A Ch was calculated for two different velocities and was

expressed as:

1
AC h 1-2 -G Se ce Ti t
1-2

391
-G Se c e Tit
V c al Veal

Substituting into the last equation the values for the Navion, the expression

for A C-u lecame:

b(
1-2

s F,
ACh 1-2 23. 1
1

Vv calj
,2 cal 2

The expression for Att t

was developed from the expression:
1 *£

<x< OL. € + w
W l

For a first approximation of the downwash at the tail, the downwash can be

assumed to be the same as the theoretical value of the downwash at infinity:

,0 _
114.6 CL
n a
 32

Therefore the expression for a t can be written as:

"t CL - CL + H - iw
" -TCT -del-

Since CL = ™ and
/
€
± Ji±^_ at the tail
qS dCL n A

a = JL -JL - 114 6 -
w +1t _
+ ,
i
« aw qS OA ^S

/ \
"
_aL
qS {— 1
~
114.6
riA 7
+ xt
" l w

The increment of &t due to a change in velocity can then be expressed

as:

Aft
2W 114,6
1-2 P S \ aa^
V w n A VcITJ
2" V cal T

It was desired to use a more refined value of downwash and the

method presented in Refs. 2 and 7 was used to determine a more accurate

value of de/da. Using the m and r parameters, Figs. 5-9 and 5-10

of Ref . 2 and the curves of Ref. 7:

vertical dist. horiz. tail a.c. to zero lift line

m =
wing bemi-span

dist. root 1/4 chord to horiz. tail a.c.

r
wing semi -span

For the Navion:

m -
- —g—
31/ 12
- n ,,
0.16

r = 15/16.7 = 0.9
 33

From Fig. 5-10, Ref. 2

de
(for A = 3:1) = 0.43
da

de
(for A = 1:1) = 0.35
da

Interpolating for A = 2:1 gave de/da = 0.39 at the aircraft cent erline,

This was corrected for the tail span and wing span using Fig. 21 of Ref. 7:

= 0.395 gave a correction factor of 0.935

>w

Therefore:

d€
= (0.39) (0.935) = 0.365

and

de iL dfl 1
'
= (0.365) = 3.76
dC- da dC L K
.097

Since the equation for A at is of the form:

Aa 391 W d€
1-2 dC
"w cal cal 2

the value of 3.76 may be substituted in for the theoretical value of

de / dCj^ . The final expression for Aa. is then obtained:

Aa *l-2 391 W l
- 3,76
i
aw VZZ?
L
i
Vcaal^
Substituting the physical characteristics for the Navion:

^a = 53,173.3
V cal:
M-2 Veal
 34

The expression for C^ may now be written:

Fs
2
V cal? '
Vcalf - C^ 6 6 ° -- 6 o
-23. 1 1 } 2
ch _
a
1 1

53, 173.3
Vcali^ Veal!

Two methods of reducing the data for Chn> were available. The

first method was to substitute in the expression for Ch/y the flight test

values and plot numberous values of Cjj versus velocity. The second

method was to make two plots, one of elevator deflection versus velocity,

the other of stick force versus velocity, by recording values from numer-

ous flight test runs. Both methods were investigated and it was found that

a definite curve of elevator deflection versus velocity and also stick force

versus velocity could be obtained. These curves are Figs. 40 and 41,

respectively. From these smooth curves, differences of stick force and

elevator deflection were obtained and substituted in the expression for

0^ and a plot of C^ versus calibrated velocity constructed, Fig. 45

Due to the complexity of the expression for C^ , it was found that the

value was very sensitive to the accuracy with which the numerical calcula-

tions were performed.

Data for both derivatives, C^ and C^ <• , were obtained through

the speed range of 80 to 130 mph indicated airspeed. The plots of

C^ and Cjj versus calibrated velocity are shown in Figs. 44 and 45.
oe OL
 .

35

Rudder Derivatives by Flight Test

The flight procedures for the rudder hinge moment derivatives were

comparatively easier to develop, since any of the variables controlling the

total hinge moment coefficient could be made zero. By flying the airplane

at a constant velocity and by varying the wing tip parachute to two different

positions, two different rudder pedal forces and rudder deflections could be

obtained

The biggest problem to overcome in the flight testing for the rudder

derivatives was a safe, efficient means of producing a steady sideslip. With

a multi-engine airplane this can easily be accomplished by varying the power

of the engines on opposite wings, but with a single engine airplane, sideslip

must be produced by some other means. For this investigation it was decided

to use a small, two -foot diameter parachute attached to the right wing tip,

Figs. 26 and 28. The drag of the parachute at the wing tip acting on a lever

arm of 16 feet was estimated to produce about four degrees of sideslip. The

amount of sideslip could be decreased by pulling the parachute inboard, thus

reducing the lever arm.

It was necessary to use a small electrically powered winch to control

the chute position. A toggle switch for operating the winch was mounted on

the flight test instrument console between the pilot and the observer. With

this arrangement, the observer could more easily control the amount of

sideslip while applying left rudder on the strain -gage riggeu rudder pedal,
 36

and observing the microammeters which indicated sideslip angle, rudder

deflection, and rudder force.

The technique for obtaining data for C, was to fly at 100 mph in
P
smooth air at a constant altitude with a fixed power setting, 22. 5 inches of

manifold pressure and 2050 rpm. The pilot held the airplane in steady

flight at 100 mph while the observer held a constant rudder deflection,

using the strain -gage instrumented left rudder pedal. Simultaneous read-

ings of 6ideslip angle, rudder deflection, and rudder force were recorded

when the pilot called out that the airspeed was steady at 100 mph. Then the

chute was pulled inboard, reducing the sideslip angle and rudder force as a

constant rudder deflection was maintained. When the airspeed and the

three microammeters were steady, the data were recorded again. This

procedure v/as repeated for different rudder deflections.

Writing the expression for the pedal force difference:

AP.F. = -G * A KM = -G q S r c r r^ A Ch = K q ACh

where K = -G S r cr T|
t

Then AP.?. = K q < Chp A(3 + Ch 6 A 6r + Che A 6t )

Since the only rudder tab is a fixed tab, this makes A 6 t = 0.

Then solving for Chg :

AP.F. K q (C h6 A6 r
-
C h, )

P K q A P
 37

In flight testing for CQg » the rudder deflection was made zero for each

reading, therefore ^ 6r = 0. The expression for C^ became:

Ch
A P.F.
f
K q A p

Substituting the appropriate physical characteristics for the Navion:

G = 1,38

S, 6.21 sq. ft,

1.27 ft.

^ 1.0

the final expression for Ch A became:

391 AP.F, -35.6 ri r2

(-11.0) V ca f A(3 cal LP: Pz'

where P.F. = Fr = Rudder Pedal Force

Ch,

The derivative C^c was te6t flown by flying at one constant

velocity and taking two readings of pedal force and rudder deflection by

varying the chute position. In this case the aircraft was maintained at

constant sideslip, that i6 A£ = 0, with the rudder. The expression for

Ch then became:
6
r

- 35. 6 r l
Ch 2
&r
Veal _ 6_ °
 Data for both derivatives, Cjj and C^ , were obtained for the

speed of 100 mph indicated velocity only, but at varying angles of sideslip

and rudder deflection. It was felt that a better comparison of results could

be made with the wind tunnel by this procedure than if the tests for the

rudder were made at varying velocities. Also it was felt advisable to keep

the velocity low to avoid the possibility of a chute failure.

It was found that the rudder force and rudder deflection were quite

steady when flying in smooth air, but that the sideslip indicating vane was

very sensitive to any disturbance. This became very important in talcing the

data for Che since these runs were made at constant sideslip, while

varying rudder deflection and rudder forces as the parachute was moved

from the full out position to the inboard position.

 )
39

DISCUSSION

The results of the hinge moments investigation are shown in the

following table:

Elevator Rudd er

c h6e Ch Ch C hp
6r

1 Analytical ( thin airfoil theory -.0147 -.0073 -.0131 +.0058

2 Semianalytical -empirical -.0092 -.0040 -.0120 +.0045

3 Wind tunnel (without spoiler wires) -.0134 -.0043 -.0080 +.0039

4 Wind tunnel (with spoiler wires) -.0105 -.0030 -.0070 +.0041

5 Flight test -.0090 -.0040 -.0160 +.0070

The results on line 1 were based on thin airfoil theory and were

corrected only for three-dimensional flow. All other effects were neglected,

These include airfoil section thickness, hinge gap spacing, bluntness of the

control flap leading edge, presence of balancing horns and turbulence of the

boundary layer. Any increase in any of these factors tends to lower the

absolute value of the hinge moments. The results on line 1, therefore,

should represent the maximum hinge moments possible if the actual airfoil

is made to approach the theoretical thin airfoil configuration.

The results on line 2 were based on experimental data published by

the NACA in Refs. 3 and 8. All the factors which were neglected in the

line 1 analysis were accounted for in line 2, and the results should be an

accurate representation of the actual hinge moments. The values obtained

 40

by this method are all smaller, by varying amounts, than the values

obtained by the method of line 1.

The results on line 3 were obtained from the wind tunnel tests of

the 8:1 scale model of the Navion tail. The derivatives Cv,
n and Cun
a P
obtained from the wind tunnel tests were in excellent agreement with those

obtained by the semianalytical -empirical method of line 2. However, the

value of Cjj was 46 percent higher than the semianalytical and Cu

6e 6
r
was 33 percent lower than the semianalytical.

Three major factors which may have accounted for the discrepancy

between wind tunnel and theory are: 1 ) the difficulty encountered in

attempting to maintain the proper gap spacing in a test model that has dimen

sions which are comparatively small; 2) the difficulty in accurately

determining the slope of C^ versus the independent variables 6e , d ,

6r , and (3 at the zero value of the variable; and 3) the difference in

the Reynolds numbers at which the model data and the theory test data were

obtained.

The NACA tests were based on chord lengths of approximately two

feet, whereas the Navion model average chord was approximately five

inches, and both tests were conducted at approximately the same dynamic

pressure of 20 pounds per square foot. The Reynolds number difference

was 1, 500, 000 for the NACA and 380, 000 for the Navion model. Since the

pressure distribution over the chord of the airfoil section changes with
 41

Reynolds number in the low Reynolds number regime, a variance of hinge

moment can be expected. See Ref. 9.

The results on line 4 were obtained from the wind tunnel using

spanwise spoiler wires of 1/16-inch diameter, located on both the upper

and lower surfaces of the horizontal stabilizer and on both side6 of the

rudder fin. Elevator Runs II, IV and V were made with the wire located

on the horizontal stabilizer at "Seven percent chord and Run VI at 25

percent. Stall buffeting occurred at angles of attack larger than five de-

grees when the wire was located at seven percent chord. Moving the wires

to 25 percent delayed separation until seven degrees angle of attack.

Stalling drastically increased the derivative Cj. , due to the lowering of

the pressure on the upper surface caused by the wire induced flow separa-

tion. The rudder was affected similarly by the wire at 25 percent chord,

except the stall buffet was delayed until 10 degrees. The delay in stall

may have been aided by the dorsal fin.

For angles of attack and yaw below buffet, the wires, due to an

increase in the boundary layer turbulence, caused a decrease in the deriva

tives Cvn , Cv,

n and Cv.
n . The derivative Cv~
n was not appreci-
6 tf 6r (3
e
ably affected.

The wind tunnel results using the spoiler wires should give a closer

agreement with the full scale airplane since the effect of the wires was to

simulate an increase of Reynolds number . The wind tunnel model

 42

was operated at a Reynolds number of approximately 380,000 and the air-

plane at 2,800,000. Therefore, the addition of the wires was desirable

in attempting to duplicate the full scale conditions.

The results on line 5 were obtained by flight testing the Navion.

Both elevator derivatives were in perfect agreement with the semianalytical-

empirical values. Cn was 15 percent lower than the wind tunnel test
"e
using spoiler wires, whereas "Cjj had a value between the two wind tunnel

te6ts. However, the rudder derivatives by flight test were larger than the

values obtained by all other methods.

Several factors contributed to the discrepancy in the results obtained

for the rudder. First, the order of accuracy of the elevator flight tests was

greater than that of the rudder. This is a direct result of being unable to

create large enough yawing moments with the two -foot diameter chute.

Small yawing moment changes, therefore, produced only 6mall changes in

the pedal force, rudder deflection and sideslip. With the pedal force points

close together, any error in reading the pedal force microammeter or in

converting to force by using the calibration chart, Fig. 31, created a drastic

variance in the slopes of P.F. vs 6r , Fig. 42, and P.F. vs (3, Fig. 43.

This caused the large scatter of points in Figs. 46 and 47, which lowered

the accuracy of the final average hinge moment derivatives Chg and Chg

for the rudder.

 43

Additional factors also contributed to the discrepance in results

between the flight test and the semianalytical -empirical dialysis. The

analytical hinge moments may be lower than the flight test values because

the proper effective trailing edge included angle was not taken into account.

The rudder has a sheet metal fixed tab extending 1.4 inches aft and extends

along approximately 1/3 of the length of the trailing edge. This tab may

effectively reduce the included angle along that portion of the trailing edge

and if taken into account would increase the analytical derivatives.

Another factor is the fact that it is very difficult to determine the

effective aspect ratio, lift curve slope, and control effectiveness of the

rudder, accurate knowledge of which is necessary in the analytical

analyses.

The discrepancy between the flight test and wind tunnel results may

have been caused by several factors, First, due to the small scale size of

the model, the gap size was larger than that on the airplane, especially at

the lower half of the rudder. Here, the gap was effectively increased by

the cavity in which the rudder torque shaft was attached. In addition, the

fuselage was wider than the rudder at the point where it faired into the

rudder, creating the possibility of turbulence along the lower portion of the

rudder. The supporting struts along the bottom of the fuselage would also

create a turbulent wake which might act to lower the hinge moments of the

rudder.
 44

Another factor may have been the relative surface roughness. The

airplane had a relatively smooth lacquer finish while the model was rela-

tively rough. This would tend to decrease the wind tunnel derivatives.

A third factor may have been the difference in shape of the trim tab

between the model and the airplane. Any difference here would change the

effective trailing edge included angle. The difficulties encountered herein

with respect to the model stems from the fact that the scale of the model

was effectively too small to accurately reproduce the physical characteris-

tics of the rudder. Despite the small scale size, the actual included angles

for both the elevator and rudder were duplicated within a fraction of a

degree.

It is apparent that if a higher order of accuracy is desired a model

of such small dimensions is not completely satisfactory for obtaining hinge

moment derivatives. It is recommended that a larger model be constructed,

preferably not smaller than 3:1 scale, for a tail the size of the Navion. The

model should be a duplicate copy of the full scale tail, and if possible, be

manufactured by the shop which produces the full scale tail. Specifically,

if the control surfaces are made of fabric, so should the model; and if the

surfaces are made out of sheet metal, the model should be made of similar

scaled down material, The importance of this lies in the fact that any

bulging of the surfaces greatly affects the hinge moments. Normally, an

increase in the thickness due to bulging will decrease the hinge moments.
 45

Flight techniques, weather conditions, and airplane instrumentation

are vitally important in obtaining accurate flight test data. Weather condi-

tions must be such that the flight is flown in still air, away from clouds.

It was found that the be6t data was obtained at night or in the early morning.

The slightest amount of thermal activity rendered any data taken unrepro-

duceable, and therefore useless, due to the large scatter in the data.

Instrumentation must be such that readings can be taken with reasonable

accuracy. Trying to read small deflections on a crowded meter scale is

useless. It is strongly recommended, in any future testing of rudder hinge

moments of single engine aircraft, that a larger sideslip producing para-

chute be used. The capability to produce at least five degree change of side-

slip with the parachute is necessary to perform proper testing. A chute of

this size would necessarily have to be contained in a container for take-off,

streamed out during the test and jettisoned prior to landing. In addition,

the capability of mounting chutes on either wing tip in order to produce

right and left sideslip is necessary, if the slope of C^ vs (3 is to be

obtained at zero sideslip.

The large scatter in the Che data is believed to be caused partially

by the difficulty in maintaining a constant sideslip angle. The sideslip angle

was extremely sensitive to any atmospheric disturbance or the slightest

aileron displacement. A larger chute and, therefore, larger sideslip angles

would tend to minimize this type of error.

 40

Since the hinge moments for both the elevator and rudder were

measured at the control stick and rudder pedal, any friction in the control

system would introduce errors in the force readings, algebraically adding

or subtracting depending upon which direction the control was last moved.

In the case of the Navion, the friction in the control systems was estimated

at approximately two pounds. Since this is not considered an excessive

amount, reasonable data accuracy was obtained for the elevator by repeat-

ing runs several times and by using average values of control forces. This

procedure thus largely eliminated the error due to friction. In the case of

the rudder, small pedal force changes due to the small sideslip chute,

allowed the friction to become significant, increasing scatter and possibly

reducing the accuracy of the results.

The only factors affecting the hinge moments in this low speed

investigation were boundary layer effects and physical dimensions. At

higher speeds where compressibility and aeroelastic effects become signifi-

cant the additional complications may increase the difficulty of predicting

the hinge moment derivatives.

 47

CONCLUSIONS AND RECOMMENDATIONS

In attempting to determine control surface hinge moments, a careful

analysis must be made of all the variables affecting the hinge moments; and

unless this is done, incorrect results will be obtained. An analytical calcu-

lation must, therefore, account for all these influencing factors in the form

of corrections. A wind tunnel investigation must duplicate the actual model

in all details, and flight conditions must be simulated as closely as possible.

Flight tests must be conducted in perfectly smooth air, and instruments

must be easy to read for satisfactory results.

In wind tunnel testing it is recommended that a large enough scale

model be used to enable authentic duplication of the original, especially with

respect to surface characteristics, trailing edge angles, the trailing edge

itself, and the hinge gap size. It is extremely important to duplicate bound-

ary layer conditions.

In flight testing, a data recording scheme would be superior to a

visual reading of instruments. A larger chute, capable of producing at least

five degrees of sideslip and mountable on either wing would increase data

accuracy. A rapid means of changing the center of gravity location for the

elevator derivatives, is a recommended convenience leading to increased

efficiency and accuracy. Locking the trim tab at the required position by a

positive external locking mechanism eliminates the tendency of the tab to

deflect, thereby eliminating erroneous stick force data. Friction can be

 48

eliminated by taking force measurements at the control surface, or if the

friction is small, take repeated runs and average out the friction error.
 49

REFERENCES AND BIBLIOGRAPHY

1. Durand, William F.: Aerodynamic Theory, A General Review of

Progress, 1934, J. Springer, Berlin.

2. Perkins, Courtland D. , and Hage, R. E.: Airplane Performance

Stability and Control, 1949, John Wiley and Sons, New York, N. Y.

3. Crane, Robert M.: Computation of Hinge Moment Characteristics

of Horizontal Tails from Section Data. NACA Wartime Rep. A- 11,

1945.

4. Williams, For man A. , Carter, R. , Landis, H. , and Wilkinson, R.:

Calibration of the Princeton University Subsonic Instructional Wind

Tunnel. Princeton University Rep. 288, 1954.

5. Pope, Alan: Wind Tunnel Testing, 1954, 2nd Ed., John Wiley and

Sons, New York, N. Y.

6. Horner, Richard E. , and Butman, P. M.: A Flight Testing Method

of Determining Elevator Hinge Moment Parameters Associated with

Elevator Deflection, Trim Tab Deflection and Angle of Attack.

Princeton University Rep. 113, 1947.

7. Silverstein, Abe, and Katzoff, S.: Design Charts for Predicting

Downwash Angles and Wake Characteristics Behind Plain and Flapped

Wings. NACA Rep. 648, 1939.

8. Sears, Richard I.: Wind Tunnel Data on the Aerodynamic Charac-

teristics of Airplane Control Surfaces. NACA Wartime Rep. L-663,

1943.
 50

9. Pinkerton. Robert M. : The Variation with Reynolds Number of

Pressure Distribution Over an Airfoil Section. NACA Rep. 6 13,

1938.

Advisory Group for Aeronautical Research and Development Flight

Test Manual, Volume n, Stability and Control, Edited by Perkins,

CourtlandD., Princeton University, 1954.

Advisory Group for Aeronautical Research and Development Flight

Test Manual, Volume IV, Instrumentation Systems, Edited by

Durbin, Enoch J., Princeton University, 1957.

Duncan, W. J.: The Principles of the Control and Stability of

Aircraft, 1952, Cambridge University Press.

 )

51

TABLE I

PHYSICAL CHARACTERISTICS OF NAVION

A. WING

1. Total Wing Area, S w (includes flaps, 184.34 ft

2
ailerons and 19.87 ft 2 covered by the
fuselage)
2. Span, b-w 33.38 ft
3. MAC 68.35 in.
4. Angle of Incidence:
Root, i r +2°
Tip, it -1°
5. Twist:
Aerodynamic 2° 31'
Geometric 3° 00'
6. Airfoil Section:
Root NACA4415R
Tip NACA 641 OR
7. Aspect Ratio, AR W 6.044
8. Taper Ratio, X w 0.526
9. Dihedral 7.5o
10. Root Chord 7.2 ft
11. Tip Chord 3.92 ft
12. Zero lift line angle of attack -1.7°

B. HORIZONTAL TAIL
1. Total Area (includes 2.368 ft covered 43.05 ft"
by fuselage
2 . Span, b n 13.17 ft

3. Airfoil Section NACA 0012

4. Taper Ratio, At .667
5. MAC, ct 3.34 ft
6. Aspect Ratio 4.03
7. Root Chord 48 in.
8C Tip Chord 30.89 in.

B-l. HORIZONTAL STABILIZER

28.95 ft
2
1. Area, S s
FRP -3°
2. Tail incidence with respect to
 ) )

52

TABLE I (Cont.)

PHYSICAL CHARACTERISTICS OF NAVION

B-2. ELEVATORS
(No Trailing Edge Extensions) Smooth skin
Flat sided
No trim bungee
Balance spring
1. Area (aft of hinge line)
Both elevators 2
14. 098 ft
2. Span (physical dimension of half ( 6. 132 ft
elevator (73. 582 in.
3. Deflection, 6
e
30° Up; 20° Dn
4. Root Chord (aft of hinge line) 16. 8 in.
5. Tip Chord (aft of hinge line) 10. 81 in.
6. Elevator MAC 1. 28
7. Root Mean Square Chord i. 15
3. Trim Tabs (two tabs, 6 x 32.5 in.)
9. Elevator Gearing Ratio 1.
10. (trailing edge included angle) 13. 5°

C. VERTICAL TAIL
1. Total Area, S v ( includes 2. 577 ft 2
blanketed by fuselage and excluding
1.483 ft 2 of dorsal fin) 12.925 ft*
2. Airfoil Section:
Root NACA 0013.2 Mod.
Tip NACA 0012-64 Mod.

C-l. VERTICAL STABILIZER

c
1. Area, S v „ (includes 0.1427 ft
8
blanketed by fuselage 6.873 ft 2
2. tt ( with reference to fuselage Gl) 2° Nose Left
 53

TABLE I (Cont.)

PHYSICAL CHARACTERISTICS OF NAVION

C-3. RUDDER
(1.4 x 16.0 in. trailing edge extension) Smooth skin
Rigged 3 Rt to Fin G
Fixed bend tab
1. Area, S r ( ait of hinge line ) 6.052 ft 2
2. Rudder Deflection, 5 r 17° L; 23° R
3. Rudder Pedal Throw 5.75 in.
4. Trim Tab Fixed bend tab
5. Rudder Gearing Ratio 1.38
6. at top ofrudder 8.8°
7. at bottom of rudder 10.3°
8. Average Rudder Chord 15.2 in.

D. MISCELLANEOUS
1. Tail Length ( CG to AC of vertical tail),
1 16.88 ft
vt
2. Distance from Fuselage Center Line to
Wing Tip Parachute Mount 16.00 ft
3. Weight (including two pilots and
30 gal. fuel) 2620 lbs
4. Wing Tip Parachute (diameter) 2 00 ft
2
Area (effective) 1 .73 ft

Drag Coefficient (based on effective area) 1 ,7

5. Dorsal Fin Area 1, ,483 ft
Rudder Fixed Tab Area 155 2
6. . ft
 54

TABLE II

NAVION TAIL WIND TUNNEL TEST

Run I: Elevator Hinge Moment Coefficients at Constant Angle of Attack

Date: 13 April 1958

1 2 3 4 5
6 ch
a e force q
deg. deg. grams psf .000421 (col. 3)

- 21.4 24. 122 - 0.90

3.80 -126.0 - 5.31
8.25 -262.0 -10.05
12.25 -397.0 -16.70
16.00 -589.0 -24.00
19.90 -697.0 -29.35
- 4.00 + 72.0 + 3.03
- 8.00 225.0 9.47
- 8.00 225.0 9.47
-12.00 383.0 16.10
-16.00 497.0 20.90
-20.00 585.0 24.60
-24.25 710.0 29.90
 55

TABLE III

NAVION TAIL WIND TUNNEL TEST

Run II: Elevator Kinge Moment Coefficients at Constant Angle of Attack

with Wire at .07c of Horizontal Tail

Date: 13 April 1958

1 2 3 4 5

0! 6e force Ch
•
q
deg. deg. grams psf .000421 (col. 3)

20.0 24. 122 - .84

1.9 60.2 - 2.54
3.7 99.0 - 4.17

8.5 220.0 - 9.26

12.4 320.0 -13.48

16.0 427.0 -17.90
19.8 546.0 -23.00
- 8.4 225.0 + 9.48
- 4.0 55.1 2.32
- 8.0 148.5 6.25
-12.0 276.0 11.61
-15.8 382.0 16. 10
-20.0 489.0 20.60
-24.0 618.0 26.00
- 4.2 77.0 3.24
+ 8.5 198,0 - 8.34

16.0 416.0 -17.50

 56

TABLE IV

NAVION TAIL WIND TUNNEL TEST

Run III: Elevator Hinge Moment Coefficients at Constant Elevator

Deflection

Date: 13 April 1958

1 2 3 4 5

ch
deg.
a \
deg.
force
grams psf
q
.000421 (col. 3)

12,4 24. 122 .0052

+ 2 32.6 .0137
4 52.2 .0220
6 70.8 .0290
8 92.3 .0389
10 122.2 .0515
- 2 4.9 .0021
-4 12.4 .0052
- 2 4.1 .0017
21.6 .0091
+ 2 38.2 .0161
4 54.1 .0228
6 72.0 .0304
8 92.4 .0380
10 121.8 .0510
4 54.0 .0228
2 39.4 .0166
21.4 .0090
- 2 6.8 .0028
- 4 9.0 .0038
 57

TABLE V

NAVION TAIL WIND TUNNEL TEST

Run IV: Elevator Hinge Moment Coefficients at Constant Elevator

Deflection with Wire at .07c

Date: 13 Apr il 19 50

1 2 3 4 5

«e
a force q °h
deg. deg. grams psf .000421(col.3)

*
— — 24 122
.

2 27.0 .0114
4 39.4 .0166
6 . 87.8* .0370
8 175.9* .0741
10 265.0* .1116

Buffeting Tail

TABLE VI

NAVION TAIL WIND TUNNEL TEST

Run V: Elevator Hinge Moment Coefficients at Constant Elevator

Deflection with Wire at 07c .

Date: 13 April 1958

*e a force q
ch
deg c deg. grams psf .000549 (col. 3)

11.9 18.512 .0065

-2 3.6 .0020
-4 4.5 .0025
+2 19.8 .0109
+4 29.2 .0159
 58

TABLE VII

NAVION TAIL WIND TUNNEL TEST

Run VL Elc»vator Hinge Moment Coef ficients at Constant Elevato:

Deflection with Wire at .25c

Date: 13 April 1958

1 2 3 4 5

&e force q Ch
deg, i deg. grams pBf .000549 (col. 3)

8 110.0* 18.512 .0604

6 60.7 .0333
4 36.0 .0198
2 24.1 .0132
13.3 .0073
-2 3.6 .0020
-4 6.8 .0037

* Buffeting Tail
 59

TABLE VIII

NAVION TAIL WIND TUNNEL TEST

Run VII: Rudder Kinge Moment Coefficierts at Constant Yaw

Date: 12 April 1958

1 2 3 4 5

* *r force q ch
deg. deg. grams P6f .000764 (col. 3)

3.75 24. 122 .0029

- 2 17.00 .0130
- 4 34.00 .0260
- 6 58.00 .0443
- 8 79.00 .0603
-10 112.50 .0859
-10 110.00 .0840
-12 141.00 . 1075
-14 180.00 .1393
- 8 79.50 .0606
5.50 .0042
- 2 18.00 .0137
- 4 37.00 .0282
- 6 59.00 .0443
4.75 .0036
+ 2 - 12.50 -.0096
4 - 36.00 -.0275
6 - 59.00 -.0450
8 - 79.50 -.0606
10 -106.30 -.0810
12 -137.50 -.1051
14 -167.50 -.1275
15 -182.00 -. 1390

8 - 78.00 -.0595
8 - 82.50 -.0630
4 - 38.00 -.0290
4 - 38.50 -.0294
 60

TABLE DC

NAVION TAIL WIND TUNNEL TEST

Run VHI: Rudder Hinge Moment Coefficients at Constant Yaw

Date: 12 April 1958

1 2 3 4 5

* *r force q Ch
deg. deg. gram6 psf .00764(col.3)

5.0 24. 122 .0038

- 2 22.0 .0168
- 4 40,0 .0305
- 6 72.0 .0550
- 8 79.6 .0607
-10 115.0 .0878
-14 180.0 . 1375
- 6 60.0 .0458
- 4 38.0 .0290
4.0 .0031
+ 2 -15.2 -.0116
4 -39.2 -.0299
6 -59.0 -.0450
8 -80.0 -.0610
10 -105.0 -.0802
12 -136.0 -. 1040

14 -168.0 -. 1282
 61

TABLE X

NAVION TAIL WIND TUNNEL TEST

Run DC: Rudder Hinge Moment Coefficients at Constant Yaw with

Wire at .25c on Rudder Fin and Elevator

Date: 13 April 1958

*

1 2 3 4 5

* *r force q Ch
deg. deg. grams psf . 000995 (col. 3)

- 4.5 18 .512 -.0045

- 2 + 6.8 +.0068
- 4 20.4 .0203
- 6 32.9 .0327
- 8 45.4 .0450
-10 60.2 .0616
-12 77.1 .0767
-14 95.4 .0948
- 8 46.5 .0462
+ 2 - 15.9 -.0158
4 - 29.5 -.0294
6 - 44.2 -.0440
8 - 63.5 -.0631
10 - 81.6 -.0811
12 -100.0 -.0995
14 -124.8 -. 1232

8 - 65.8 18 .675 -.0650

 62

TABLE XI

NAVION TAIL WIND TUNNEL TEST

Run X: Rudder Hinge Moment Coefficients at Constant Yaw with No

Rudder Trim Tab and with Wire at .25c of Rudder and Elevator

Date: 13 April 1958

•%

1 2 3 4 5

* *r force q Ch
deg. deg. grams psf .00 102 (col. 3)

- 6.8 18.512 -.0069

- 2 + 4.5 +.0046
- 4 15.9 .0162
- 6 29.1 .0296
- 8 40.8 .0416
-10 56.3 .0573
-12 70.4 .0716
-14 87.4 .0890
- 8 43.2 .0440
2 - 15.4 -.0157
4 - 29.5 -.0300
6 - 44.2 -.0450
8 - 59.0 -.0600
10 - 77.2 -.0786
12 - 96.5 -.0984
14 -118.0 -. 1200
 63

TABLE XII

NAVION TAIL WIND TUNNEL TEST

Run XI: Ru dder Hinge Moment Coefficients at Constant Yaw with

Smooth Trim Tab

Date: 13 April 1958

*

1 2 3 4 5

* 6r force q Ch
deg. deg. grams psf .000955 (col. 3)

1.8 18.512 .0018

- 2 4.8 .0048
- 4 27.2 .0270
- 6 39.8 .0396
- 8 55.6 .0553
-10 71.5 .0711
-12 88.5 .0880
-14 113.5 .1128
-14 113.5 . 1128
- 2 17.1 .0170
1.8 .0018
2 - 7.9 -.1079
4 - 25.0 -.0248
6 - 43.2 -.0430
8 - 60.0 -.0596
10 - 84.0 -.0835
12 -109.0 -. 1085

14 -136.5 -. 1351
6 - 43.6 -.0433
 64

TABLE Xin

NAVION TAIL WIND TUNNEL TEST

Run XIL Rudder Hinge Moment Coefficients at Constant Rudder

Deflection

Date: 12 April 1958 1

1 2 3 4 5

* *r force q ch
deg. deg. grams psf .0184/q (col. 3)

2.0 24.122 .0015

- 2.0
- 4.0 - 10.3 .0079
- 6.0 - 18.2 .0139
- 8.0 - 26.3 .0207
-10.0 - 51.3 .0391
-12.0 - 67.8 .0517
-13.0 - 77.0 .0587
+ 1.0 .0008
+ 2.0 7.0 .0053
4.0 17.0 .0130
6.0 30.0 .0229
8.1 49.6 .0378
10.0 62.5 .0477
12.5 100.0 .0763
 65

TABLE XIV

NAVION TAIL WIND TUNNEL TEST

Run XD1: Rudder Hinge Moment Coefficients at Constant Rudder

Deflection

Date 12 April 1958

1 2 3 4 5

* »r force q ch
deg. deg. grains p8f .0184/q(col.3)

- 2.0 24. 122 .0015

- 2.0 - 5.0 .0038
- 4.0 - 16.0 .0122
- 6.0 - 27.0 .0206
- 8.0 - 39.5 .0301
- 8.0 - 39.0 23.836 .0301
-14.2 -106.0 .0818
+ 2.0 + 5.0 .0039
4.0 15.0 .0116
6.0 26.2 23.467 .0205
8.0 41.5 .0325
8.0 39.2 .0307
10.0 61.0 .0478
10.0 73.0 .0572
12.0 95.0 23.181 .0753
15.0 150.0 23.017 .1198
 66

TABLE XV

NAVION TAIL WIND TUNNEL TEST

Run XIV: Rudder Hinge Moment Coefficients at Constant Rudder

Deflection with a Smooth Tab

Date: 13 April 195 8

1 2 3 4 5

* *r force q Ch
deg. deg. grams psf .0184/q(col.3)

1.0 18.582 .0011

- 2 -3.2 .0032
- 6 -15.9 .0157
-10 -38.6 .0382
-14 -73.7 .0729
+ 2.3 .0022
+ 2 9.1 .0090
6 25.0 .0247
10 43.2 .0428
14 88.0 .0871
6 25.0 .0247
 3
67

TABLE XVI

NAVION TAIL WIND TUNNEL TEST

Run XV: Ru dder Hinge Moment Coeffi Lcients at Constant Rudder

De flection with Wire at .25c of Rudder and Elevator

Date : 1 April 1958

1 2 ,3 4 5

* 6r force q Ch
deg. deg. grams pef .0184/q(coL3)

- 2 -11.4 18.582 .0112

- 4 -21.4 .0212
- 6 -33.6 .0333
- 8 -44.2 .0437
-10* -59.0 .0584
-12* -76.2 .0754
-14* -87 o 8 .0869
- 8 -44.2 .0437
+ 2 - 1.1 .0011
4 + 7.9 .0079
6 19.3 .0191
8 28.6 .0283
10* 40.8 .0404
12* 51.7 .0512
14* 73.8 .0731

Buffeting Tail
 68

TABLE XVII

NAVION TAIL WIND TUNNEL TEST

Run XVI: Rudder Hinge Moment Coefficients at Constant Rudder

Deflection with No Rudder Trim Tab and Wire at .25c
of Rudder and Elevator

Date: 13 April 1958

1 2 3 4 5

* *r force q Ch
deg. deg. grams psf .01886/q(col.3)

- 6.8 18.582 .0069

- 2 -10.2 .0104
- 6 -21.6 .0219
-10 -48.8 .0496
-14 -75.0 .0761
+ 2 - 2.9 .0026
6 + 17.1 .0173
10 38.6 .0392
14* 65.9 .0670

Buffeting Tail
 HM

Elevator Convention
^
A

-HM
+ HiA

Rudder Convention

Fig. 1 Sign Conventions

 T-

Fig. 2 Diagram Showing Determination of Angle 6 Q

used in Analytical Development (Thin Airfoil Theory)

 CI jK>S
ri» s

a? -

.
ACA OB lo. 5BOS Tig «

./o I

< >

5?.-!
- -
— 1 > — _
<X3 ""*" -, l

Fig. 4

^A
JX - -6 1
> |l

«* _ r" i

_
i r- "

o+ -. t
,\' c*~ — -**. »-
|

*^
jS* >»

.OJl - ~.£
—
S_ _V j______

M
O
~i ) eo so 40 5O

> I

f
C**
>

**
t \*
^
°< 1
>
j- ^ \

At 9.-1 ' 3o Jl

1 — -

I ATIOi AL AS rise* T 0O« MITT1 1

TOR AIROI AUTIC •

0O4
N k

•
1
BO SO 4
/so
s
V 1
1

'XAJ+ y 1

s*
**
^? 41 <t/~
^
_J
J t" •
^***
C <£*/•

-.ixx

-on.
-fVf
 H
T

U&
S
1f-':
|

-
 1

i* ',
\. i

*W(

f?3&

L*s.? jyvB/fi$

ttvjj* b./Sfcto *

f/& '
*J*j9B
•&
mMl
i
 *

To p

<,..-•• •
SectW
Sc <>! 1 in - 5 -ft.

4 4 + V + 4 ^ ft

^ci\\ t'flctl : \' Ce>\C

3 D m€ f i

•
Se c "f » f>v
Test s<r etion

DTf-f i -.-.
i it ,

Fit .
l » •
••"

p; f • •
i

FanS

Section f .

( i

/At

Turiv&blc Sfct'cn

. 9
 Fig. 10 Model installed in test section

Fig. 11 View beneath test section showing rudder

strain gage beam
TJ
O
B
<->
O
o
rH

CO
e
o>
si
o
00

(1)

0)

H
en

CVi

u.
 . _

_ —_.^ ..

i
i

i !
>N 1

J
<i

\\
\ \ CD v
\
i*

\ \\ \\ \ ^
\ \ to
\\ \\ \ \
v \ \\ \\
\ V
«. \ \ \
,,. . .
1
\ m '

\
\\ V \
^"
\ \ \ \ \ \ ,
* -

\
\\
'
v \ \ to 1

o>
u\ r -TV f-<\ S\ H yr-t \
0\ 7\ <i \ -

j
V ^ \ \ \ "W"
1

\
\ \
\ -S \
\ ^ \
\i
\
\
V" \
\ \ \ 1-4

\
•
\
\\ \ * >
v
\
\
\\ Y\
\ 1 \ \
\\ \^ \
\ \
1

A\
i

t [\

^
o
1 > <

V
v.
j» |

4
V \\
\
\^
i \ V |\ v \
\ \
1

\ \
\ \
, \ \
\ \ \ \
—
B \ \
> \ \

o \ \ y
"

•
\ \ V \

•
s to \
«

1 a \
\ ' '1

1 *a at
w*
1
\
\\
io •H
Ol
to
to to
\ •
—
t
H Gi. ^

tO > to
\
•

H * 1
o
1
n \
6H i
9 CO r H &4 <

•H CO u
i 4-> o<
cI
Q
^ I
p o
4»
f3
h
V
<*4
4
g 11 *
*rjvr* a
'

, . .
. - . -
 . . !

! i i

\ .

\
\ .

\ o
OP
\
\ \ \ <n •

\ \
\
\ \ \ \ i. ,

\ \ \ \

\ i\ \ ! \
\ \ \ \ 1

\ \ \ \
\ \
V |\ 9
\ '\ «o
\ v.
\ \
°c A CO
V
*V ^
Pnr-f^ r-A r-^ ft
—
• \ IF
\ \\
* \

\ \ \\
• \ \ \ *
V \ A to
,

\. \ V \ \ \
i\ \ \\ \l
•
w
\
Y \
\ \ \
\
\ \
^
\ \
\\
\ \
\ \
V v
\ \ \ —
X
i

j.. >. £> !*

c i ft 1 » V i \ i » V
\ \
i

g \ 1

\
l| \ \
V \
\ \ \ \
\ \ \
\ t
\ \
_l \
* \
\
c \
' •

V \
\ \ \.
\ ? >

£
M 1,
^
\ \ V
\ \ ^

— i
.
\ \
3> \
i ° }

^ i j V \ \ .

\A
-*

n
. .

s
! * w y \
6

bo
r-l
-> \
\
A\
\
tX, u •?
\ \
\
•>
(SI
i6 t 1 V

p \ \
s « im \ \
i 3
k
\
\
1 \
* >lr-t 8 (i

y >
\ .

r \ \
>
-*_ \

4|» 1 \
.

1
. ..
 -
/ 1 r /i-
/
/
' '
J
/
/
/
O A
i
;

f-t i
/ ^
J
f
^ '
i
1
'

rH
fc>

B //
O
i
p -4 . kCU . ?L_
7 _. . ..
o rH
/
ID
<D iH
1
I;
0) ^~) /
c OB
<-T> rH
i

lO •h a /
rf [J- rl
a '/,/
o Li O. J> ;

CO ;

;
:

ioi 4-»
/
'

cd
»
r > * /
> O /
(D
Cd 1 r
1
/
/ /
r
. .

..
-
,

b/
fj
1

Oil
//
J
.

1 >

-> u
D ^ M>
:
1

*3
t
V
. <3
i.
; t
o
g
<
R
4
s t
\ -
o
•
o 6
s 0J 1 ,

i 1 t I

//

tf 1

. A •^ w !
f
/r P
: 1
C
t
//
to
/ ~T 03
'

-4

/
fcrj 1
1

r\ /
;

/I CO o
H /
i

/ !

^-1
>
1 3 o
O-
r
,,

rH u
1

\ /
jr

/
I
*7*
'

/
A
ra
>
\
' <M

1
£
•H

- .

_^ "
a*
H i i

c
/ 1
1
^

/ &
7
1
O rH
to
o '- M 1-
/ I 1
^.
1 ex
/
• s £
/
U o 4 .'

OO . .

I
\ J I i
 —
"

1
1

jctior
1 •

'

|;~
H i
i
i

....

'•
s
:

rH
o.
J
t
-

!
» OJ > Aflft

£ ^ i

y
!

i
to

1
> .- o
1

cH "C
1 1

c
p
0)
"
rl 00 _ .
s T
g «
*
i

c ~tf}
* i

w
& "

i
.-.*.-.

w
"SP
/;
r
,

\
. _

^ *&
.

V
*
I

3
D >
u c <
ft c < 5 <: <

X: < 3 c 3 < 3 w < y < 3 C D ( 3 •

-4

o r
i

• 4 ~< TCffU
• t
i ' 1 1 r 1

n ^
c
a
C- C >

•^ ro
\r
TIT c5
8 8 \
>
1

> *H o 1
r '

| r\ /at* n w^- —4P

Hf ii rr r i

1 5 *\ // / i r.

V
«^ o.&
§ X *
/™ OD
99 1 cr c>
« _/ /
|

ffl^
9" /t\
'/* V
£E5 b
H
\x i
1
5 i

>i

k
• •

n
3 l»^
^*> J > "
1 . CVi "u
4->
r" C
c t 2 ; oi
i =
r* i 43
n <* 0) o 1
1

w u ^ f
n 1 J
«

4
1 1
1 I 3
.

1
:

•
o

'

1
'"
<
n. # L aC ai 6_
nj
$ ^ Q <>

. .
,
-- . .
 t . s .

—
! ;

1 ;
;?
1 1

. .

o -1
ftf)

5 L"
<
A— —«
v;^
A \\
I
M 1

a
a OJ / n #-v

R
H (fa
K/f

3
H
c
B
fc o
rH
Yd L „

O 5
>

t
1

1 g c, ? *
CO
1r
/
Ar 1
it
— ,

-
J!
An
i
H ___
<£> 1 1

'
fc
/
£
"
5
I
/

J 1 3
3
! j

•4 - I
1

S
¥f S~
"

J
XJ
oj
r
i D i
IX
=>
i
D . *
D
c«
<
3
//
/
f
N
3
!
c
K
&.
:
- j

s
1
P c
<
po
<

r
d <
r
4 ^ p

c 1
.
/ -A /rtl OJ '

»
'

o>
"
/ i to r-t
y
/ o
i « n o
o
t
*
/ r
1 .

•
-V W n IQ_

/ /
I
$ «

k .

+\ X«
i— .

^y
// fd
on
1

/ I xS
r^ ra~ •P XI
o
f^ o cd
'

3 4V f l
•**
r> «
rH O o
J

I o
06
to
rvj

o
....

ft OJ
rH a» <0
1 h u
:

1
r-i
M * i
a
.

"v PI
tj tf
>4 \z>
•H -tttJ
1
til
>
> X
-.
o X
fi 5
3-
1 |
T a <3>
?r
Fig. 19 General Three-view drawin
S
 Fig. 20 Test aircraft, Navion N91566

Fig. 21 Strain gage instrumented wheel for measuring

stick force
Fig. 22 Strain gage instrumented beam for measuring
rudder pedal force

Fig. 23 Rudder pedal strain gage beam installed

in a i re raft
Fig. 2k Elevator deflection indicating potentiometer

Fig. 25 Rudder deflection indicating potentiometer

 Fig. 26 Side slip chute

Fig. 27 Side slip angle indicating vane

Fig, 28 Drawing of sideslip chute installation
 -

;
— ,

_
o
PJ

|_
lO
cu

c > »
1 ^v
r

-
o
.

> 00
.

>V| »

: ^>

»i>
rH

•
^V
,
' i

'

o "
s
3 o
M 3
!

I
e-i 1

<

i *3
N.
CM .

cc
o
—u* X
c<
i

"
|

«
i

M > to
M 1

CO
.

N^
. .

Q
CD
CI
«>
>
s*
}
:
c ^i
c*
<
Ci
i <
r
b
H
« P
H
«

r
r
4
c*
r
> C =* v CD «, . h < ^ o
I

.1
g C 1

VI "
T |

C
*
. ;

r . .
'

\
_'_j
 ,

r i
- —
! 1

. i

1 o—
^t
I

rH

... .

i
CO
'
1

, i

*
.

,
•
o
(_)

^^
•

s. i

'.i

s •

o
CO
J .

-' - 1
{;

^v
^^ V O
m
--
<»
'

>s
•,
i .

g ^>—

^S Si
to

P
o
<3i
^
,

"

(0 fi <r
1 1 o
>>N
"
iri i
1W t
. .

'

ft ,

£
. ...

-. ,
1 o ,

H .
00:

'

2
04
*- s
_
D
»-,
1 c t5 1 C 5 ^D Y
c

^
c
p i 8* I 34
4 C 4 c c c r t
, , ,
|
"*
rH i

19! 4«
± •

ir 1
i '
i . . i .
 — i

.
—n •
!

!"

s
\

;
to
OJ
•

! \
* ~
.

r--
o
W
P.
e
«<
t
•^

55 '
\
A
Ol ,

lO
H
V s
03;
\

\
.
1
\
g
z a \
o
!

0, 1 in
n 1
h a
o
l
r^
t H
<
b
1

• :
J f* 1

O.

c 1

p
3
(-h \ c itf
.

\
;

|\
•

"
n c 3 1
» uD I ) a 1!
f o^> c^ rH c > r- 1

4-»
to
•
.

•
. i

H o
£ l£ »•

;
Wb
 ^

—
ill! TT'TTT.I j I T I i I I .1
 <r

to
D
f-
j CJ
o
> OC
lO

o
UJ
o
DC

6 o
o
Ll.
to

h-

D
<
>
LlJ
 W5 u
>
o
O
LU
i

to
M Ll. to

LJ
\ G Ex.

cr
O
CO O
<
>
LU

LU

-A/WWV
u
o
..J

to

00
Ex.

u
,.L
L_J
Q
Q
ZD
CL
 u
• z
(

>
o Z
ki h-

> o
UJ
_J
00
to

U_ bO
Id
Q
CD < o

Q
Q
Z).
cr
L
-VWWW ~ > 1
 !_

o
a:
o
o
o
UJ
-J
Ll.
UJ to
Q
•H
UJ
z
<
>
CD O
9 9
a.

CO
uj.

9
-A/VWV CO
 .

1
|

. ._. ... ;• I

- , 1
U
o 1 ,

ft t

> 1
-J O
in
rH £
rH
M 5

o
o 4i
.
4> S 4
1

bO
V 1
-k to
Hi
b£ i

* in O
i

( 3 Oi s
4
o S D f!
P o
i 3 i3
-r «h

m 5;
ri
W r> r. '
s
I'D
\

'
\ -

V
O
CM
.

PC
ill
>
>l
•

2
1 .

,,

'
. . •

g
o

\
V
L
\ •
O
<->

N
H
\
So

4
\
.

. ..
\ .

^ j»
'

""H-
s

-
1
o
P
C>
4i
<>

1
fl 1

01

'

J i
 . .

~ -

. , ... . j — p_

. •
. 1 -

'
i 1

•a

.. .
r •* i o
as
1
-

.... _4 .....

:
H \
'

-. . I -H Q
09
> s a) rH

_ b
4?
Q>
8 i
:

s.
_ .
to
Gu
c H
W
O
—
o
to
rH
0) c "
X., .
1
•5
•H
o -

t
«3
ad
>v on 55

O .

H
.....
1

^V
...

.
O
3 >
o
1
x^ i

J O
1

o
\

. o
0»
Sv
.

Ar

\\ !

'
Q
33

.
00
„
» * J c3 1 o < < O -Q Oj
'
* < V
bJ* u* t c* "i " 1
• 4>
;j - c
F- i .

,. .
 '

; 1
''
[
- I

1 |

i
..
'

'
'

^
.

<-; :

T o O O
4 Si o d c
C O i*
4
IN P O- Q
8
a is 1

(X ' F ^3
^ en
H
CO
,i 1 H : 1
1

> |

?3. Hi V^ «** v^
Q>
o
rH
CJi
I

-H CJ CO
* t> <' <&
,
t •>
gl <*)
J
1

« 1
5 ^ 1
!
to J>
3 y ...
-to
o
1 CJ>
I

h V
ft
i j5
cS
3
A
.fej
i

oD J CI
cJ D <3 >
ft <
J c J cJ cJ r^ r rH ^<*&\r^ a )
c
3
o1
L

^-—
^^
^ -^
u'-«-
»^y»
X* J
-
*t
7
Xf
a-^
**^
F -cvr
-
t
<&* »

'
^-
^< r|1"
^ ' .y.
&^. ^- •
r
*

to
..
3$^ j .
& +

& *rr -
:

*-*c.

— ;
-**
r
*-=

"
> %&
^— -1
•-«
-* •-- UF Ik r
~-\
<£-T> f
^^' +
•*"^"^
^>1 T

^ik
--<1

^ a--*
v
""r1
-**-*.'
1
1

j
^ r
_ 1

>*
i !
'

-:j2^
^_r
>' '

1 P
^ jH>i*f
!

i>itir
J *r CD
'
* ^
v^u

t . .
 — — -

i
— —— — •—

1
.... .

'

V /

/
/ —
i

VX
* l

' y > rH'

Pi (
I 3
3
C4 C
p ^ 5^. if
>
>
pi'
rt
CD
rH
^
r
• di c > t> ^ >
1 o
01
+'

*r s -

.:<. * <*> -5* /

~s •
^ S* r-<:
+
#* '

# ,<>

^
1

>: ^<
' 1
.

<
'
'I /
-
4^ r
rf < /
cv;
+
>*'
r /
X* / <
.--
ji 1

•
/ *
/
H /
•

k '

/
;

»
/ / +
•r-f 3
/
r.
D o
t0) *-!
1

0)1 4->

•H
en rH
.

ro CQ
> f
7
B . 1

.2"
a> 5 §
o o
u< o 5
o 5
ft C o O a
cD r . 5 iO
•-4 ti 1
0) 03 s 11 1 t r i i. , „
id
03
> 1

u u
V/7J
i
1

a>
1 C >--. *
^ Ih
' 1
*' V
<*
' > l

T
31 1

OC
i .

.... <

'
 .

;
1
' 1

r .... ! . , 1 I
»

L_
:
'

i 1

a 1

C
c
T
.

,.. .

F
i
;
— A

«
i

i
o :
•

«
i

•
, .

> o .

5 H
ho
;

"
J3
i

e:
rv
B
s e o
J3 9
3 H •
.

3 »

gf
i
• A SB
H

• D

•
c :

cc

r 1

c
'*•'
>

nJ
c
§
1

4
I 1 c
I$ c cI
•

o
: I

1 1
•

!
j .V"l
<

fel
l

! . _
 —

'

. ' •

~ •

- 4
o
-*F -

i o
>
s «
3
-to.

>
-
a * s
o
"T~
'
1*. rH
. 1
jf 1

«fl

& 1

O -i

r-\ •
• '

r4

» 3
•>
o

o O
>
r4

a
.

• e
- O !

* o> •
'

r
o
CO

_ *

a 1 I

. > r i .
•
[']
-I 3
• • 1

4
I .>

: ...J —
 . : — r

— :
!
1 1 | 1 !'
1
1 1 j
1 : 1 ! • 1 1
'

| !

''
!

- .
©
\

o
r

.
i
1 *^ —
i

|
f

O
ac H l-

1 o
i

>

o
1

© C£
u
1 3*
:

< 1
o
o
'1

o
1

o © «

o
I

.
o
o

1 3 C 3 c 5 ( p
1 i ? c 3 »
H C D
3 ( c t \ C 3
• « • o
'
1 1 1 4
'
' 1 4

J ...

;
if

1 1
 . i

'

: I .i
!

-
—
—
i
o
_
BH

1
:"!
i

1.
tfc>

IQ 1

—
1* •

</D *»
<*
a 1

o i>

3
i
1

* »- -i
T
&4
o

..... i tf^
• ! r*\

'

•• K>1<
i

'

r
..:!;

r-\

1
-

1
i

— ti
- -v
rA
V el s

C
/-» /H
cv
,

> \>
i

C> c> O a I

r t

-
. . .... .....
i
-1

3 . .

'
i i .. _