Wind-Tunnel Investigation of a Full-Scale Canard-Configured General Aviation Airplan

82 Pages·2010·35.42 MB·English

https://ntrs.nasa.gov/search.jsp?R=19850011615 2019-03-27T23:06:37+00:00Z
FL2827
30SPW/XPOT TECHNICAL LIBRARY r Z, R F Ri
BLDG. 7015 U 1_4;7! ?'t.1 Ir ^' N I i 15 !
NASA 806 13th ST., SUITE A ,
VANDENBERG AFB, CA 93437-5223 VA.""'.Z-J' im
Technical
Paper
2382
March 1985
Wind-Tunnel Investigation of
a Full-Scale Canard-Configured
General Aviation Airplane
Long P. Yip
NASA
Technical
Paper
2382
1985
Wind-Tunnel Investigation of
a Full-Scale Canard-Configured
General Aviation Airplane
Long P. Yip
Langley Research Center
Hampton, Virginia
NASA
National Aeronautics
and Space Administration
Scientific and Technical
Information Branch
Summary Introduction
As part of the aeronautics program in the area
An investigation was conducted in the Langley 30-
of stall/spin research at the Langley Research Center,
by 60-Foot Tunnel to determine the aerodynamic char-
wind-tunnel tests were conducted to assess and docu-
acteristics of a powered, full-scale model of a general
ment the aerodynamic characteristics of a canard con-
aviation airplane employing a canard. Although pri-
figuration designed for general aviation use. In the mid-
mary emphasis of the investigation was placed on eval-
1970's, a new homebuilt airplane design, the VariEze
uating the aerodynamic performance and the stabil-
(ref. 1), made a significant impact on the general avia-
ity and control characteristics of the basic configura-
tion community because of its canard design and other
tion, tests were also conducted to study the follow-
advanced features. These advanced features included
ing effects of varying the basic configuration: effect of
use of composite construction for lighter weight and for
Reynolds number; effect of canard; effect of outboard
smoother surface contours to improve aerodynamic per-
wing leading-edge droop; effect of center-of-gravity lo-
formance, use of winglets on the main wing for direc-
cation; effect of elevator trim; effect of landing gear;
tional stability and, at the same time, for reducing drag,
effect of lateral-directional controls; effect of power; ef-
and use of a canard surface to increase pitch stability
fect of fixed transition; effect of water spray; effects of
near stall so that the maximum trim angle of attack was
canard incidence, canard airfoil section, and canard po-
less than wing stall angle of attack.
sition; and effects of winglets and upper winglet size.
This report presents results of a full-scale research
Additional aspects of the study were to determine the
model of the VariEze design tested in the Langley 30- by
boundary-layer transition characteristics of the airfoil
60-Foot Tunnel for which preliminary results were re-
surfaces and the effect of fixing the boundary layer to
ported in reference 2. Test data obtained included mea-
be turbulent by means of a transition strip near the
surements of aerodynamic forces and moments of the to-
leading edge. The tests were conducted at Reynolds
tal configuration, isolated loads on the canard, pressure
numbers from 0.60 x 106 to 2.25 x 106, based on the
distributions, propeller torque-thrust loads, and flow vi-
wing mean aerodynamic chord, at angles of attack from
sualization using tufts and sublimating chemicals. Also
—4.5° to 41.5°, and at angles of sideslip from —15° to
included in the study were effect of Reynolds number;
15°.
effect of canard; effect of outboard wing leading-edge
The investigation indicated that employing the ca-
droop; effect of center-of-gravity location; effect of el-
nard on this configuration was effective in providing in-
evator trim; effect of landing gear; effect of lateral-
creased stall departure resistance because the canard
directional controls; effect of power; effect of fixed tran-
stalled before the wing stalled. Influence of the canard
sition; effect of water spray; effects of canard incidence,
flow field on the wing decreased the inboard loading
canard airfoil section, and canard position; and effects
of the wing as the outboard loading of the wing in-
of winglets and upper winglet size.
creased. The increased outboard loading and spanwise
flow development on the wing caused wing tip stall.
Symbols
The addition of a wing outboard leading-edge droop in-
creased stall angle of attack and increased pitch stability All longitudinal forces and moments are referred to
at low to moderate angles of attack. From tests using a the wind axis system, and all lateral-directional forces
chemical sublimation technique, the natural boundary- and moments are referred to the body axis system.
layer transition was found to be at 55 percent chord Unless otherwise noted, total-airplane and canard mo-
of the canard. Fixing transition near the leading edge ments are presented with respect to a center-of-gravity
of the canard resulted in a significant reduction of lift location at fuselage station 99, which was 0.71c ahead of
due to flow separation near the trailing edge of the ca- the leading edge of the wing mean aerodynamic chord c,
nard and, subsequently, a nose-down trim change and and at a vertical location on waterline 16. Also, unless
loss of elevator effectiveness. Variations in the canard otherwise noted, total-airplane and canard aerodynamic
airfoil showed that the canard airfoil-section character- coefficients were reduced by using a wing reference area
istics can strongly affect the airplane stall and poststall based on the trapezoidal planform of the wing projected
characteristics. Moving the canard to a lower position to the fuselage centerline.
had little effect on the static longitudinal and lateral-
directional aerodynamic characteristics of this configu- wing span, 22.17 ft
ration. The lateral-directional stability was generally
b,,, upper winglet span, ft
satisfactory, but the directional stability became weak
at high angles of attack. Larger upper winglets pro- CD total airplane drag coefficient, Drag
qS
vided significant increases in directional stability of the
configuration. CD,c canard drag coefficient, Canard balance drag
qs
CD,f skin-friction drag coefficient, Skin-friction drag S' exposed canard area, ft2
qS
CL total-airplane lift coefficient Lift V free-stream velocity, ft/sec
, qS
V/nd propeller advance ratio, V/(Propeller rotation
CL,, canard lift coefficient based on wing reference
speed x Propeller diameter)
area Canard balance lift (CL, in computer-
X chordwise distance from leading edge, ft
generated figures)
CL canard lift coefficient based on canard plan- (x/c)T boundary-layer transition location
^
form area, Canard balance lift Y spanwise distance from plane of symmetry, ft
q S'
CL,o lift coefficient at zero angle of attack Y' distance along winglet span, ft
lift-curve slope, per degree a angle of attack relative to WL, deg
CL^
a angle of sideslip, deg
CI rolling-moment coefficient, Rolling moment
qSb
ACD incremental drag coefficient
rolling moment due to sideslip, per degree
Cl,, ACI incremental rolling-moment coefficient
C,,, total-airplane pitching-moment coefficient,
AC,,, incremental yawing-moment coefficient
Pitching moment
qSc ACY incremental side-force coefficient
C,,,,,, canard pitching-moment coefficient relative to
ba aileron deflection based on a setting of equal
airplane c.g. Canard balance pitching moment
qSc and opposite deflection, positive when right
aileron is down, deg
C,r,,o pitching-moment coefficient at zero angle of
attack 8e elevator deflection, positive trailing edge
down, deg
C,no. slope of pitchimg-moment curve with respect
to angle of attack, per degree
b, rudder deflection based on setting one rudder
Yawing moment in an outward deflection for directional
Cn yawing-moment coefficient,
qS6 control, positive left rudder deflected, deg
yawing moment due to sideslip, per degree
Cn,, 7? propeller efficiency
Cp pressure coefficient, n—9p°° Subscripts:
CT thrust coeffic ient, ThrSust
qs c canard
Cy total-airplane side-force coefficient, side force I lower surface
qs
Cy,, side force due to sideslip, per degree max maximum
U upper surface
c local chord, ft
W winglet
c reference wing mean aerodynamic chord,
2.58 ft Abbreviations:
cn section normal-force coefficient obtained from BL butt line, in.
integration of pressure measurements
C. g. center of gravity
ic incidence angle of canard relative to WL,
FS fuselage station, in.
positive trailing edge down, deg
L.E. leading edge
LID lift-drag ratio
WL waterline, in.
P local static pressure, lb/ft2
Poo free-stream static pressure, lb/ft2
Model Description and Test Apparatus
q free-stream dynamic pressure, lb/ft2
The configuration used in the study was a powered
R Reynolds number based on c full-scale model of an airplane intended for the home-
built market (ref. 1). The model was constructed of
S reference wing area, 53.60 ft2
2
foam covered with fiberglass and epoxy. Body putty the engine inlet and exit areas were sealed and faired
was applied to the wing and canard to attain the de- for a no-flow-through condition. No attempt was made
sired airfoil-section contours. Geometric characteristics to simulate the internal duct flow due to a reciprocating
of the model are given in table I and shown in figure 1. internal combustion engine.
A total of 322 pressure orifices were installed in the Overall aerodynamic forces and moments acting on
wing, canard, and winglet. The pressure orifice loca- the model were measured on the external scale balance
tions are given in table II. Photographs showing the system of the Langley 30- by 60-Foot Tunnel. (See
model installed in the Langley 30- by 60-Foot Tunnel ref. 6.) In addition, the model was instrumented with
are presented in figures 2 and 3. internal strain-gauge balances to measure isolated loads
The basic model configuration is defined as follows: on the canard and the propeller and with scannivalve
transducers to measure the surface pressures. Small
Outboard wing leading-edge
cotton tufts were used in conjunction with fluorescent
droop off
photography to provide flow visualization of the model.
Center of gravity located at FS 99
(See ref. 7.) Tufts were used in flow visualization studies
Nose gear removed
to examine areas of flow separation and other surface
Main wheel pants off
flow conditions at angles of attack up to complete wing
Propeller removed, spinner on
stall. Initially, tufts were installed on the upper surfaces
Inlet faired, exit area sealed
of the wing, upper winglet, and canard. However, the
High canard position with
tufts on the canard resulted in premature transition of
i, = 00
the boundary layer; thus, there was a large decrement in
Canard with GU 25-5(11)8
the lift performance of the canard. Therefore, canard
airfoil section (ref. 3)
tufts were not installed in later tests because of their
Small upper and lower winglets
adverse effect on the flow patterns of the canard. A
Variations to the basic configuration include the chemical sublimation technique (ref. 8) was used to
following: provide information on the extent of laminar flow on
the canard, wing, and winglet.
Adding a discontinuous outboard wing
leading-edge droop
Removing canard Test Conditions and Corrections
Moving center of gravity to forward and
Test conditions included a range of a from —4.5° to
aft locations
41.5° and a range of 0 from —15° to 15°. Aerodynamic
Varying landing-gear arrangements
data were obtained at free-stream tunnel velocities of
Adding power effects
26, 68, and 94 mph that correspond to Reynolds num-
Varying canard incidence
bers based on e of 0.60 x 106, 1.60 x 106, and 2.25 x 106,
Changing canard airfoil section
respectively. Most of the tests, however, were conducted
Changing from high canard position to
at a nominal free-stream velocity of 68 mph.
low canard position
The model was tested upright and inverted to eval-
Removing winglets
uate the flow angularity and strut tare corrections. An
Increasing upper winglet size
extensive wind-tunnel calibration was made prior to
Range of control settings tested were be = —20° to model installation to determine the horizontal buoy-
24°, 8a, _ —20° to 20°, and 6, = —40° to 40°. Pitch ancy correction, and flow-field surveys ahead of the
control was obtained with elevator deflections at a fixed model were made in the manner of reference 9 to de-
canard incidence setting. Canard incidences of —4°, 0°, termine the flow-blockage correction. These corrections
and 4° were tested. A low canard position (fig. 1(a)) have been applied to the data. Jet-boundary correc-
was also tested because of interest in improving pilot tions were made in accordance with the method of ref-
visibility. Since earlier studies (refs. 4 and 5) indicated erence 10. Since an electric motor, rather than a recip-
that the droop was effective in delaying tip stall, tests rocating engine, was used to power the model and no
were conducted with the leading-edge droop installed. attempt was made to simulate the internal duct flow,
(See fig. 1(d).) Upper winglets with 50 percent more no corrections were made for cooling drag due to a re-
area (figs. 1(b) and 1(c)) were also tested. ciprocating engine.
Powered tests were conducted with a 200-HP elec-
tric motor to turn a fixed-pitch, 4.83-ft-diameter, two-
Presentation of Results
bladed propeller. The propeller is a Hendrickson
H58G64 propeller designed for climb. The majority of The test results are presented in figures 4 to 44,
the tests was conducted with the propeller removed, and which are grouped in the order of discussion as follows:
3
Figure Longitudinal characteristics . . . . . . 39
Effect of Reynolds number . . . . . . . . . . 4 Lateral-directional characteristics . . . . . . 40
Pressure distributions . . . . . . . . . . . . 5 Effect of winglets:
Drag characteristics . . . . . . . . . . . 41
Section normal-force distributions . . . . . . 6
Lateral-directional stability . . . . . 42 and 43
Effect of the outboard leading-edge droop: Pressure distributions at angles of sideslip . . 44
Flow visualization with tufts . . . . . . 7 and 8
Longitudinal characteristics . . . . . . . . 9 Discussion of Results
Elevator trim requirements . . . . . . . . 10
Drag characteristics . . . . . . . . . . . 11 Effect of Reynolds Number
Lateral-directional stability characteristics 12
In order to assess the sensitivity of the configura-
Lift and pitching-moment characteristics: tion to Reynolds number effects, data were compared at
Effect of canard . . . . . . . . . . . . . 13 Reynolds numbers based on c of 0.60 x 106, 1.60 x 106,
Elevator control deflections . . . . . . . . 14 and 2.25 x 106. These data are shown in figure 4. The
Effect of center-of-gravity location on lift and pitching-moment characteristics of the basic
elevator trim requirements . . . . . . . . 15 configuration and, also, the isolated lift characteristics
of the canard obtained from the canard balance indicate
Drag characteristics:
that at the low Reynolds number this configuration ex-
Effect of elevator deflection . . . . . . . . 16
hibited significantly different lift and pitching-moment
Trimmed lift-drag ratio . . . . . . . . . . 17
characteristics from those at higher Reynolds numbers.
Effect of landing gear . . . . . . . . . . 18
The canard data for low Reynolds number exhib-
Configuration effects on lift-drag ratio . . . 19
ited significantly lower lift than the lift obtained at the
Lateral-directional characteristics: higher Reynolds numbers and were a primary factor in
Stability characteristics . . . . . . . . . . 20 the lower lift level of the total airplane. Also, lift-curve
Aileron control . . . . . . . . . . . . . . 21 slope of the canard for the low value of R was lower
Rudder control . . . . . . . . . . . . . . 22 for angles of attack less than 6° and increased with
increasing angle of attack. The ineffectiveness of the
Power effects:
canard to generate lift is probably caused by laminar
Propeller efficiency . . . . . . . . . . . . 23
separation of the boundary layer due to the effect of
Effect on longitudinal aerodynamic
low Reynolds number, whereas the increase in the lift-
characteristics . . . . . . . . . . . . 24
curve slope is probably caused by turbulent reattach-
Boundary-layer study: ment at the higher angles of attack. Since the canard
Extent of natural laminar flow . . . . . . . 25 is located well ahead of the airplane center of gravity,
Effect of fixed transition . . . . . . . . . . 26 changes in the lift and lift-curve slope of the canard sig-
Effect of transition on canard drag . . . . . 27 nificantly affected C,,,, and C,,,,. of the total airplane,
Effect of transition on as shown in figure 4. At the low Reynolds number,
pressure distribution . . . . . . . . . . 28 the loss in canard lift shifted C,,,,o to a more negative
Effect of transition on elevator value, whereas near az = 6°, the increased canard lift-
trim requirements . . . . . . . . . . 29 curve slope caused C,, to be unstable. Although this
Sketch of water-spray boom . . . . . . . . 30 Reynolds number is low compared to flight conditions,
Effect of water spray on canard the data are presented here to illustrate the sensitivity
aerodynamics . . . . . . . . . . . . . 31 of total-airplane lift and pitching moments of canard
configurations to subcritical Reynolds number.
Effect of canard incidence:
The data at mid and high Reynolds numbers indi-
Longitudinal characteristics . . . . . . . . 32
cated much better agreement on the lift and pitching-
Elevator trim requirements . . . . . . . . 33
moment curves. The data at mid Reynolds number are
Effect of canard airfoil section: representative of landing approach speed of the airplane
Comparison of section contours . . . . . . 34 at higher angles of attack. The remaining analysis in
Longitudinal aerodynamic characteristics . . 35 this report is based on the data obtained at this mid
Canard balance lift data . . . . . . . . . 36 Reynolds number.
Effect on canard position:
Pressure Distributions
Photograph of model with canard in low
position . . . . 37 Presented in figure 5 are the chordwise pressure dis-
Wing-surface flow patterns . . . . . . . . 38 tributions and section normal-force coefficients of the
4
wing, upper winglet, and canard. The data of figure 5 stall. The tuft patterns of the wing with the L.E. droop
are presented graphically with the pressure distribu- on show that the leading-edge droop reduced spanwise
tions on the configuration so that they could be related flow, kept the flow attached at the wing tip region, and
to the spanwise distribution of section normal-force co- thereby delayed wing tip stall to a higher angle of at-
efficients. The data are presented for an angle-of-attack tack. The significance of these flow patterns is indi-
range from —2.5° to 31.5° for the basic configuration. cated by the lift and pitching-moment data of figure 9,
The data of figure 5 indicate that strong favorable which indicate that the leading-edge droop increased
pressure gradients, conducive to boundary-layer stabil- CL,,,,ax and increased the pitch stability near a = 4°
ity for laminar flow, were obtained on the canard upper which made the pitching-moment curve more linear in
surface from the leading edge to about 50 percent chord the mid angle-of-attack range. The effect of leading-
at angles of attack up to about 10°. Favorable pressure edge droop on elevator deflection required for trim is
gradients were also found on the wing upper surface shown in figure 10 for forward, mid, and aft center-of-
at angles of attack less than 5.5° and on the upper sur- gravity locations. The leading-edge droop provided a
face of the upper winglet throughout the angle-of-attack larger stall margin between the maximum trimmed CL
range presented. and
CL,max
The chordwise pressure distributions were inte- The effect of leading-edge droop on the trimmed
grated to obtain section normal-force coefficients for the drag characteristics of the configuration is shown in
wing, upper winglet, and canard. The data of figure 5 figure 11. A drag penalty, OCD = 0.0040, at cruise
indicate that the canard operates at a higher section condition of CL = 0.25 was incurred due to the addition
loading than does the wing at angles of attack up to of the leading-edge droop. This drag penalty probably
13.5°. The higher canard section loading promotes ca- would not be as large on an airplane with leading-
nard stall before wing stall; thus, airplane stall resis- edge droop integrated into the construction of the wing
tance is provided. because the test model leading-edge droop was made
The section normal-force coefficients c,, of the wing removable and was not fastened to the wing surface
and upper winglet are shown in figure 6 for the canard- as smoothly as the original construction surface. At
on and canard-off conditions. Figure 6 illustrates the higher lift coefficients corresponding to climb, there was
influence of the canard downwash/upwash flow field on no significant drag penalty associated with the leading-
the wing and winglet. On the inboard part of the wing, edge droop modification.
lower levels of c, were caused by the downwash of the The lateral-directional stability derivatives CY.,
canard, while higher levels of c,, on the outboard part C,,p, and Cl,, were obtained from tests conducted at
of the wing were caused by the upwash outboard of 0 = —5° and 5° and are shown in figure 12. The addi-
the canard tip. From a design point of view, the effect tion of the leading-edge droop increased directional sta-
of downwash on longitudinal stability and the effect bility C,,,, at angles of attack up to wing stall. Rolling
of upwash on wing tip stall must be considered. The moments due to sideslip increased with angle of at-
impact of the canard downwash/upwash flow field on tack typical of configurations with wing sweep. The
the aerodynamics of this configuration is discussed in addition of the droop reduced the magnitude of dihe-
later sections. dral effect —CI. of the configuration for angles of at-
tack up to about 20°. At low angles of attack near 2°,
Effect of Outboard Leading-Edge Droop
this reduction in dihedral effect made the configuration
Based on the design philosophy of references 4 marginally stable in Cl,,. However, at higher angles of
and 5 on wing leading-edge droop design, an outboard attack, the reduction in dihedral effect may be benefi-
wing leading-edge droop was installed on the VariEze cial in reducing the amount of lateral control required to
airplane to increase stall resistance and reduce the wing trim the configuration in sideslip, such as in a crosswind
rock tendency of the configuration. As reported in ref- landing situation.
erences 4 and 5, the outboard leading-edge droop pro-
vided attached flow near the wing tip to a higher angle Lift and Pitching-Moment Characteristics
of attack and reduced the autorotative moments in the Canard configurations require that the center of
poststall region. Wing tip stall was more prevalent for gravity be located between the canard center of lift and
the present configuration because of higher wing load- wing center of lift for both positive stability and control.
ing outboard, due to the canard, and from the wing If the canard stalls before the wing stalls, longitudinal
sweep effect. Wing tuft patterns of the droop-off and stability and airplane stall resistance are increased.
droop-on configurations are shown in figures 7 and 8, However, as pointed out in reference 11, many factors
respectively. Without leading-edge droop, the tuft pat- must be considered in order to make the configuration
terns on the wing show the spread of spanwise flow near stable and controllable as well as stall resistant. These
the trailing edge of the wing and the development of tip factors, including airfoil-section characteristics, power
5
effects, and center-of-gravity location, are discussed in pitching-moment coefficients for three center-of-gravity
later sections of this report. locations is presented in figure 14 for the basic config-
Figure 13 presents data for the canard-on and uration with the L.E. droop on. As expected, the lift
canard-off conditions and incremental data obtained by data of figure 14 indicate that increasing the elevator
subtracting canard balance data from the total-airplane deflection in the positive direction (trailing-edge down)
data. Analysis of the data indicates that the wing lift increases the overall lift. Thus, this canard configura-
is influenced by the presence of the canard because of tion does provide a positive increment in trimmed lift
its downwash effect. This downwash effect caused the as opposed to a conventional tail arrangement which
wing to experience less lift than would be predicted by would normally provide a decrement in trimmed lift.
adding the interference-free contributions of wing and However, analysis of the canard balance lift data in-
canard individually. A beneficial effect of canard down- dicates that the canard lift is not directly additive to
wash is that it delays the stall of the wing; thus, the the total lift because of the increasing downwash due to
angle of attack margin between canard stall and wing elevator deflection of the canard on the wing. This in-
stall is increased. The data of figure 13(a) indicate that creasing downwash caused a destabilizing effect on the
canard stall occurred at about a = 13°, whereas wing total airplane pitching moment at angles of attack be-
stall occurred at about 21°. With the canard off, the low canard stall.
wing stall occurred at about 19° which is 20 less than Since changes in the center-of-gravity location would
with the canard on. not alter the lift curves of figure 14(a), only pitching-
Examination of the pitching-moment data of fig- moment data are presented in figures 14(b) and 14(c)
ure 13(b) showed three significant changes in pitch sta- for the forward and aft center-of-gravity locations. The
bility throughout the test angle-of-attack range. The data of figure 14 indicate that canard stall reduced the
first change occurred at about a = 4° where there was effectiveness of the elevator at high angles of attack;
a decrease in the lift-curve slope of the wing. This de- thus, the maximum trim angle of attack was limited.
crease in CL. caused a reduction in total-airplane pitch The maximum trim angle of attack for the aft center-of-
stability since the aerodynamic center of the wing is lo- gravity configuration was obtained with an elevator set-
cated aft of the center of gravity. The decrease in CL. ting of 8e = 15°. A plot of elevator deflection required
of the wing is caused by the development of spanwise for trim, shown in figure 15, indicates that increasing el-
flow due to wing sweep. This development of spanwise evator deflection beyond 15° actually trimmed the con-
flow is shown by the tuft photographs of figure 7. This figuration at slightly lower values of lift coefficients. In
decrease in pitch stability at a = 4° can also be found all test conditions, the trimmed lift coefficient was less
in the canard-off configuration. The second change in than the maximum lift for the configuration; thus, stall
pitch stability occurred at about a = 130 where canard resistance to the configuration was provided.
stall resulted in a nose-down pitching-moment incre-
Drag Characteristics
ment to provide a large increase in pitch stability. This
increase in pitch stability would require more elevator The effect of elevator deflection on drag character-
deflection to trim the configuration. The third change istics of the basic configuration with L.E. droop on is
in pitch stability occurred at about a = 21° where wing shown in figure 16. The drag of this canard configu-
stall occurred. Because the wing is located aft of the ration increases with increasing elevator deflection for
center of gravity, a destabilizing pitching-moment in- a given lift coefficient which indicates a drag penalty
crement occurred when the wing stalled. This pitch-up associated with trim. A trimmed lift-drag polar for a
tendency would not normally be encountered in flight mid center-of-gravity location is indicated by the dashed
as long as the canard provides enough stall resistance line of figure 16. Values of the trimmed lift-drag ra-
to limit the airplane angle of attack to that below the tio are plotted in figure 17, and a maximum value of
wing stall angle of attack. Although the canard does 12.6 was obtained for this configuration. Incremental
provide beneficial increment to stall resistance of this values of drag for the nose gear, main landing gear,
configuration, several factors, including elevator con- and wheel pants of the main gear are presented in fig-
trol authority, center-of-gravity location, airfoil section, ure 18. As shown in figure 19, these increments were
power effects, and surface roughness could adversely af- incorporated into the drag curve of the basic configu-
fect the configuration stability and cause the configura- ration to obtain new values of trimmed lift-drag val-
tion to trim at angles of attack higher than wing stall. ues. The maximum lift-drag ratio of this configura-
Effects of elevator control and center-of-gravity loca- tion was improved by moving the center of gravity aft
tion are examined in the following discussion, and the by 0.106 ((L/D)max = 13.1), by adding wheel pants
effects of airfoil section, power, and surface roughness ((L/D)max = 14.1), by removing leading-edge droop
are discussed in a subsequent section. ((L/D)max = 15.4), and by removing the main landing
The effect of canard elevator deflection on lift and gear ((L/D)max = 17.1)•
6
Lateral-Directional Characteristics conditions a nose-down increment of pitching moment is
associated with increasing power setting. Although the
The lateral-directional stability derivatives Cy,,,
thrust line of the propeller is slightly above the center of
Cnp, and C1p were obtained from tests conducted at
gravity, the moments produced by this offset do not ac-
0 = —5° and 5° and are shown in figure 20 for the count for the nose-down increment of pitching moment
basic configuration with L.E. droop on. This configu-
due to power. This increment probably comes from the
ration was directionally stable at low angles of attack;
power-induced flow cleanup of the wing trailing edge
however, the directional stability decreased to zero at
and from increased suction pressures acting on the base
about a = 19°. This configuration exhibited stable
area of the cowling. The data of figure 24 also indicate
dihedral effect that increased with angle of attack up
that there is a slight increase in pitch stability due to the
to wing stall. The effect of deflecting the elevator on
propeller except for CT = 0.11 at a < 6°. This stabiliz-
lateral-directional stability of this configuration is also
ing effect is probably due to the rotating propeller disk
shown in figure Over the test angle-of-attack range
20. developing a propeller normal force (ref. 11), which on
up to the stall, deflecting the elevator caused the direc-
a pusher configuration produces a nose-down moment
tional stability to decrease and the lateral stability to
because the propeller is located behind the center of
become more stable. The decrease in directional sta-
gravity. Conversely, a propeller located ahead of the
bility is probably due to the increase in canard drag
center of gravity would have a destabilizing effect, es-
ahead of the center of gravity. The increase in the di-
pecially if the propeller slipstream immerses the canard
hedral effect is probably due to the asymmetric canard
and provides increased lift.
downwash on the wing with sideslip angle, which causes
an incremental increase in rolling-moment contribution Boundary-Layer Transition Study
due to sideslip.
Several pilots have reported that while flying their
The aileron and rudder control authorities of the
homebuilt version of this configuration in rain condi-
basic configuration are shown in figures and re-
21 22,
tions, the airplane exhibited a pitch trim change. This
spectively. Both positive and negative control inputs
pitch trim characteristic seemed to indicate an effect
were tested; results were averaged to reduce effects of
caused by changes in the boundary-layer properties
model asymmetries and tunnel flow angularity; data are
of the canard or wing. In order to investigate the
presented for a right roll input (ba, < 0) and right yaw
boundary-layer characteristics, tests were conducted to
input (b, < 0). The data of these figures indicate that
determine the extent to which laminar flow existed on
both aileron and rudder control authorities decreased
the configuration and to determine the effect of a loss in
at higher angles of attack. Also, aileron deflections pro-
laminar flow which might occur when the airfoil surfaces
duced favorable yawing moments in the normal opera-
become contaminated by insect or rain-drop accumula-
tional angle-of-attack range (a = 2° to 18°).
tions on the leading edges.
A chemical sublimation technique (ref. 8) was used
Power Effects
to locate the boundary-layer transition on the canard,
Propeller thrust and torque were measured by wing, and winglets. The technique involved spraying
means of a balance mounted between the motor and the a coat of chemical film on the model surface, starting
propeller. The data shown in figure 23 indicate that the wind-tunnel airflow, and observing the sublimating
a maximum propeller efficiency of 0.75 was obtained. process of the chemicals. Since the surface chemicals
This value of propeller efficiency is relatively low com- sublimate at a higher rate in a turbulent boundary layer
pared with that for more optimized arrangements. The than in a laminar boundary layer, a definite pattern
propeller used in the tests was of low pitch for maxi- of chemical residue is observed on the wing which
mum climb performance and was therefore not properly denotes transition. Tests were conducted on the upper
matched for cruise conditions. The low value of effec- surface only at an angle of attack of and at a
1.5°
tiveness in the tests is consistent with that obtained in Reynolds number based on c of 1.60 x 106. Test results,
reference 12 for low propeller blade angle settings. The shown by the photographs of figure 25, indicate that
low value of efficiency may also be associated to some transition was located at 55 percent chord of the canard,
degree with the pusher arrangement at the rear of the 65 percent chord of the wing, and 60 percent chord of
fuselage. Improved shaping of the aft fuselage and en- the winglet. These transition results were confirmed by
gine nacelle and careful matching of the propeller geom- flight tests as reported in reference 8. The large amount
etry with the fuselage flow field could provide increased of laminar flow on the configuration can be attributed to
propeller efficiency. the composite construction of the aircraft which allowed
Effects of propeller thrust on the longitudinal aero- for smooth airfoil contour.
dynamic characteristics of the configuration are shown In order to simulate conditions in which laminar flow
in figure 24. The data of figure 24 indicate that for most would be lost, such as in rain conditions or with insect
7
accumulations, a transition strip of No. 60 carborun- Effect of Canard Incidence
dum grit was applied at the 5-percent-chord location
In order to obtain an inherently stall-proof airplane
of the canard and wing in accordance with the method
that employs a canard, it is important that the canard
of reference 13. Results of the transition grit tests are
incidence be set so that it will stall at an angle of attack
shown in figure 26 for test conditions of transition free
below the wing stall angle of attack. Canard incidences
(no grit applied), transition grit at 5 percent chord of
of 0°, —4°, and 4° were tested to determine the effect
the canard, and transition grit at 5 percent chord of
of canard incidence on the longitudinal aerodynamic
both the canard and wing. The data of figure 26 in-
characteristics of this configuration. The data from
dicate that fixed transition at 5 percent chord of the
these tests, presented in figure 32, show the expected
canard caused a decrease in the lift-curve slope of the
changes in C,,,,,,, and stall angle of attack where positive
canard by about 30 percent. This decrease in the canard
incidence produced increased C,,,,,o and reduced angle of
lift-curve slope resulted in an increase in pitch stabil-
attack for canard stall. At an incidence angle of —4°,
ity and a large nose-down pitching-moment increment
the canard stall angle occurred at about 18° which is
at the higher angles of attack. With transition grit on
below the wing stall angle of 23°. At a canard incidence
the wing and canard, the data of figure 26 indicate that
angle of 4°, the pitching-moment curve was more linear
the configuration exhibited a slight increase in nose-up
in the mid angle-of-attack range (a = 4° to 101. This
pitching moments at angles of attack of 4° or below.
effect was probably caused by the combination of an
This effect of fixed transition at low angles of attack is
increased downwash on the wing which delayed the
probably due to the nature of the boundary layer on
angle of attack where the pitch stability changed near
the wing which is indicated by pressure distributions as
a = 4° and an increase in pitch stability caused by
highly laminar at low angles of attack but quickly be-
early canard stall. Elevator settings required to trim
come turbulent near the leading edge at angles of attack
the configuration with canard incidence settings of —4°,
above 4°. Thus, fixed transition would only affect the
0°, and 4° are presented in figure 33 through the trim-
laminar flow nature of the wing at low angles of attack.
lift-coefficient range. The data of figure 33 indicate that
Presented in figure 27 are canard balance lift-drag the effectiveness of the elevator with canard incidence
polars. The data of figure 27 indicate significant drag at i, = 4° decreased and was probably caused by the
increases due to fixed transition on the canard. An ex- elevator operating in a separated flow region above
amination of the chordwise pressure distribution on the a = 8°. (See the data for CL, of fig. 32.)
canard, shown in figure 28, indicates that the loss of
lift due to a fixed transition is a result of trailing-edge Effect of Canard Airfoil Section
separation which was probably caused by the thickened
The canard is an important factor in the configura-
turbulent boundary layer having to overcome a sharp
tion's trim capability and stall characteristics. To de-
pressure recovery near the trailing edge. This sepa-
termine the effects of the canard airfoil section on the
rated flow condition also resulted in a decreased elevator
configuration, an NACA 0012 airfoil section was tested
control authority as indicated by the data of figure 29.
on the configuration. The NACA 0012 section is typical
This decreased control authority could become signifi-
of airfoil sections used on conventional general aviation
cant when flying in rain where loss of laminar flow could
airplane horizontal tails. The basic canard airfoil sec-
require sudden changes in elevator settings to trim the
tion, GU 25-5(11)8, was designed for high lift and low
pitching-moment changes encountered.
drag at low speeds. (See ref. 3.) This airfoil was rel-
In other tests without transition grit, water spray atively thick and highly cambered, and a comparison
from a horizontal boom fixture in the wind tunnel was between it and the NACA 0012 airfoil is shown in fig-
used to study effects of surface water on transition. The ure 34. As discussed earlier, the basic airfoil section is
water-spray boom, shown schematically in figure 30, characterized by large amounts of laminar flow. Test
was located approximately 4 canard chord lengths in data comparing the effect of canard airfoil on the total-
front of the canard and covered only one side. The airplane lift and pitching moments are shown in fig-
spray rate was approximately 1 gal/min. Results from ure 35. The data of figure 35 show the change in C,,, ,o of
water-spray tests of the canard, shown in figure 31, are the total airplane which is primarily due to the change
similar to results of fixed transition on the canard, that in CL,, of the canard with uncambered NACA 0012 air-
is, a reduction in the canard lift-curve slope and an foil section. Changing to the NACA 0012 canard also
increase in drag. It should be noted that only about lowers CL,,,,,,x. The poststall lift characteristics, which
one-half of the canard was immersed in water spray for are significant to the total-airplane pitch stability, are
these tests. If the canard were fully immersed in water examined in more detail in figure 36. A comparison
spray, the data would be in closer agreement with the of the canard lift obtained from the canard balance is
data obtained with fixed transition. shown in figure 36 based on the canard area. The data