VOL. 12, NO. 8 J.

AIRCRAFT AUGUST 1975

Aerodynamic Design of High-Perf ormance Biplane Wings

Robert B.Addoms*
University of California, Los Angeles, Calif.

and

Frank W.Spaidj
McDonnell Douglas Corporation, St. Louis, Mo.
Theme with increasing camber, up to an ideal C( of approximately
0.5. The usual sense of induced camber in a biplane is to
A LTHOUGH biplanes were quite popular during
the early days of aviation, they had virtually
disappeared from service by the mid-1930's. The biplane
reduce the effective camber of the airfoil, thus increasing the
severity of the nose suction peak at a given Ct and reducing
seemed to be plagued by inherently high drag and low
c- fmax*
v /

The flow about an airfoil may be divided into the mean

maximum lift coefficient. The purpose of the present study
velocity vector, Q (including the mean value of the induced
Downloaded by YORK UNIVERSITY on March 3, 2015 | http://arc.aiaa.org | DOI: 10.2514/3.59846

was to re-examine the reasons for the biplane's decline. It was

flow parallel to the chord), and the perturbation velocity com-
found that for any practical biplane configuration, it is
ponents u and v, parallel and normal to the chord line. The
possible to design a wing system which has a much lower
condition of tangential flow at the airfoil boundary may be
weight per unit area and which has essentially the same
approximated as
maximum useable lift coefficient as that of a comparable,
well-designed, monoplane wing, in either clean or flapped
configurations. Improvements in the design of fairings permit (1)
large drag reductions, relative to the earlier designs.
where y(x) is the mean camber line, a. is the local angle of at-
Because of these improvements, it is possible to design a
tack, k refers to velocity components induced by the vortex
biplane whose performance is superior, for some ap-
distribution of the given wing at the given spanwise station,
plications, to that of a well-designed monoplane. These ap- and / refers to the induced flowfield. The induced downwash
plications are those for which excellent low-speed
has the same effect as an increase in the local camber line
maneuverability, good short-field performance, good load-
slope.
carrying ability, low cost, and rugged construction are of An analysis presented by Millikan2 has been used to
primary importance.! calculate effects of induced camber. Millikan uses a horseshoe
Content vortex and a vortex pair to represent each wing. Some results
in the form of chordwise and spanwise average values are
If a biplane wing system is to develop a high C Lmax , both
presented in Fig. 2, where (AC L ) C is the increment in CL
wings must stall nearly together. A method presented by caused by induced camber, and CLi is the lift coefficient of the
Fuchs was used to identify those configurations for which a inducing wing. These results show that induced camber is
good stall match could be obtained. Each wing is idealized in almost exclusively a two-dimensional effect, since it is quite
this method as a single horseshoe vortex, with the bound part insensitive to aspect ratio. It is largest near (3 = 0 and decreases
of each vortex located at the center of pressure. Satisfactory as 1/31 or gap/chord are increased. Effects of gap and stagger
stall match was defined as a ratio of leading wing to trailing on measured values of CLmax for biplanes are in reasonable
wing Coequal to unity at an assumed CLmax for the com- agreement with these results, if (AC L ) c is interpreted as an
bination. Some results of this analysis are given in Fig. 1, and estimate of the change in CLmax.
some nomenclature is defined in the sketch. The airfoil chord
lines are parallel; results of this study showed no advantage to
be gained from the use of decalage. The geometric stagger
angle a for best stall match is quite insensitive to the Normal to
gap/chord ratio, and it decreases as the aspect ratio increases.
The best stall match always corresponds to small values of 0,
OQ \— - >— No
Normal to
the aerodynamic stagger angle. \—-4 i/' chord line
The chordwise variation in vertical velocity induced by one
airfoil upon the other corresponds to a local curvature of the
flow. Since within the framework of thin airfoil theory, this
effect is identical to the effect of a change in the mean camber
line, it will be referred to as induced camber. 0.38
Data for NACA airfoils of the four- and five-digit series
show increases in the maximum section lift coefficient Cfmax
Received November 1, 1971; synoptic received September 12, 1974;
revision received January 24, 1975. Full paper available from the 1 °-34
I I
National Technical Information Service, Springfield, Va., 22151, as 2 0.30
N75-14749 at the standard price (available upon request). This re-
search was supported by the National Science Foundation and by the
University of California at Los Angeles. Aspect ratio = 14
Index category: Aircraft Aerodynamics (including Component
I I I I I I
Aerodynamics). 0.26
0.6 0.8 .1.0 1.2 1.4 1.6 1.8 2.0
* Graduate Student, presently with Maritime Dynamics, Inc.,
Geometric gap/chord
Tacoma, Washington. Associate Member AIAA.
tSenior Scientist, Research Laboratories. Associate Fellow AIAA. Fig. 1 Wing arrangements for good stall match.
 630 R.B. ADDOMS AND F.W. SPAID J. AIRCRAFT

change in lift forward manner to separate optimization of upper and lower

coefficient due
to induced camber
wings, or to spanwise camber-line tailoring.
lift coefficient of The method of Blackwell was chosen to calculate the in-
inducing wing duced flow. This method, in which the lifting surface is
0.02 represented by a distribution of horseshoe vortices, has been
shown to give satisfactory predictions of lift curve slope and
induced drag factor for elliptic and rectangular wings of
moderate aspect ratio. l
Changes in the mean camber line of the parent airfoil will,
in general, change its drag characteristics. The method for
computing airf oil drag used in this study is similar to that
described by Spence and Beasly. Their method was
programed as an iterative procedure in which the potential
flow and the boundary-layer development were matched, and
has been shown to give accurate results for cruise of con-
ventional airfoils. l
Calculations have been made for two configurations, based
Downloaded by YORK UNIVERSITY on March 3, 2015 | http://arc.aiaa.org | DOI: 10.2514/3.59846

on the NACA 23012 airfoil section. One wing arrangement

consisted of equal, rectangular wings of aspect ratio =
7.11, geometric gap/chord = 1.1 38, and geometric stagger/
chord =0.539. The value of C Lmax , 1.30 in this case,
was defined as the CL for which C( first reached C,max(1.51 at
/?ec = 3.5x!0 6 ) at the most heavily loaded station of either
wing. The design procedure was applied only at this station.
The biplane airfoil has significantly more camber than the 230
camber line from which it was derived; see Fig. 3. The com-
-0.16
0 10 20 30 40 50 60 puted increase in Cd is 0.0001, indicating that no drag penalty
Aerodynamic stagger angle (deg) is incurred by the increased camber.
Fig. 2 Influence of wing arrangement on induced camber. Similar calculations were made for a configuration having
aerodynamic gap/chord = 1 .0 and stagger/chord = 0, a con-
figuration for which the unfavorable induced camber is quite
large. Structural analyses indicated that there are no ad-
vantages to be gained by designs which employ smaller
gap/chord ratios. There was no significant increase in Cd at
cruise C f . Apparently the gap/chord and stagger can be
chosen to satisfy stall match and structural requirements, and
the airfoil section can then be designed to compensate for the
resulting induced camber.
The preceding ideas have been extended to the case of
biplanes having simple flap designs. Using conservative
assumptions, a CLmax is predicted to be essentially the same as
that for a monoplane wing having the same type of flap.
A design study was carried out, using results of the previous
analysis, in which a biplane and a monoplane were compared
20 40 60 80 100 under the ground rules of equal gross weight (8000 lb), the
Chordwise station, % chord
same engine, and the same equivalent parasite area for com-
Fig. 3 Comparison of biplane mean camber line with original 230 puting the drag of the fuselage, tail surfaces, and landing
mean camber line. gear. It was concluded that: a) the biplane wing weight, in-
cluding bracing, should be less than 60% of that for a com-
parable monoplane, b) The biplane bracing system can be
A design procedure is proposed in which a known airfoil streamlined and faired so that the resulting drag penalty is
section is selected. The gap/chord ratio and stagger angle are small; 2) A biplane can have superior slow-speed
chosen to provide a good stall match and structural efficiency. maneuverability and short field performance, d) The biplane
The induced flowfield is then calculated for a CL for which has a greater payload, which offsets its slightly lower cruise
the C,, of the most heavily loaded section equals C(^ax. The speed, giving it greater cargo carrying capacity.
camber line of this section is chosen to have the same
chordwise vortex distribution as that of the two-dimensional
airfoil, i.e., References
s, R. B., "Aerodynamics and Structural Design Con-
= (dy/dx)2_D - ( -v-JQ\ (2) siderations for High Lift Biplane Wing Systems," Ph.D. thesis, Nov.
1971 , Department of Mechanics and Structures, University of Califor-
This camber line is used for the full span of both wings. The nia, Los Angeles, Calif.
2
process does not require iteration, since the spanwise lift Milikan, C. B., "An Extended Theory of Thin Airfoils and its Ap-
distribution is not changed. It can be extended in a straight- plication to the Biplane Problem," Rept. 362, 1930, NACA.