Aircraft Design 3 (2000) 167}174

Determination of the Oswald e$ciency factor at the

aeroplane design preliminary stage
O. Samoylovitch
q
, D. Strelets*
Faculty of Aircraft Engineering, Moscow Aviation Institute, 4, Volokolanskoye Shosse 125871 Moscow, Russia
Abstract
A new engineering approach to the determination of Oswald's span-e$ciency factor results in good
convergence of calculated and experimental data. The proposed method allows to de"ne its value more
reasonably and to analyse possible ways of improving it. 2000 Published by Elsevier Science Ltd.
All rights reserved.
Oswald's span-e$ciency factor is a generalised parameter connected with an aircraft's aerody-
namic e$ciency. Speci"cally, for a parabolic drag polar, there exists a dependency:

D
`
"
1
2
eA
C
'"
. (1)
The Oswald e$ciency factor e re#ects the aeroplane lifting properties deterioration caused by the
distortion of an elliptical lift distribution and accounts for the non-ellipticity of the lift distribution,
the increase of pro"le drag of the wing, fuselage, tailplane, nacelles and various interference e!ects
with angle of attack [1].
At the initial stage of the design the Oswald e$ciency factor is rather di$cult to compute.
Nevertheless, several authors have tried to derive its analytical relation to the aeroplane geometri-
cal parameters. However, the results are still not accurate enough.
The research covers only a steady #ight regime area with linear dependency of lifting force on the
angle of attack. It includes only aeroplanes of classical layout (not biplanes or other extraordinary
schemes) and does not take into account the lifting force of tailplanes or canards.
q
Deceased.
* Corresponding author. Tel.: #7-095-158-4105; fax: #7-095-158-2771.
E-mail address: [email protected] (D. Strelets).
1369-8869/00/$- see front matter 2000 Published by Elsevier Science Ltd. All rights reserved.
PII: S 1 3 6 9 - 8 8 6 9 ( 0 0 ) 0 0 0 1 2 - 4
Nomenclature
taper ratio (c
''
/c
''
)

''
leading edge sweepback angle
radius of leading edge cross section
A wing aspect ratio ("b
2
/S
''
)
b wing span
c wing chord
C
'
drag coe$cient
C
'"
zero lift drag coe$cient
C
''
induced drag coe$cient
C
'
wing lift curve slope for incompressible #ow
d fuselage diameter at the wing attachment zone
D drag
e Oswald e$ciency factor
k
$
fuselage cross section shape factor
lift
(L/D)
`
maximum lift-to-drag ratio
S

leading edge suction force, % of theoretical

S
''
reference wing area
t thickness of an airfoil
=M
'
relative fuel capacity
Real values of Oswald's e$ciency factor for several aeroplanes are given in Table 1. As can be
seen from the data Oswald e$ciency factor values are within the limits from 0.6 to 0.85. It means
that taking e"0.85 instead of actual e"0.6 gives an error in the lift-to-drag ratio equal to 1.5}2
that is not acceptable even at the preliminary design stage.
The Oswald e$ciency factor is in#uenced by
E the wing planform (its aspect and taper ratio, leading edge sweepback angle);
E presence of a fuselage, nacelles and other components.
This in#uence is illustrated in Fig. 1 for three aeroplanes (F-104, F-18A and Falcon-900) with
di!erent aspect ratio but with approximately equal leading edge sweepback angle and taper ratio.
It follows from this "gure that with the aspect ratio decreasing at constant leading edge sweepback
angle and taper ratio, the Oswald e$ciency factor for a wing alone increases and tends to unity
while for the aeroplane as a whole the converse is true. This paradox can be explained by the fact
that the fuselage causes a loss in lift, leading to an irregular spanwise lift distribution. This example
clearly shows that the fuselage is essential and its size should be taken into account even at the early
stage of design.
Oswald's e$ciency factor is in#uenced also by the wing leading edge cross-sectional shape which
in the case of a correct pro"le can enforce the suction at the leading edge zone and decrease drag
168 O. Samoylovitch, D. Strelets / Aircraft Design 3 (2000) 167}174
Table 1
Oswald e$ciency factor values for some airplanes
Aircraft C
'"
A e (L/D)
`
Boeing 247D 0.0212 6.550 0.75 13.5
Douglas DC-3 0.0249 9.140 0.75 14.7
Piper J-3 `Cub'' 0.0373 5.810 0.75 9.6
Beechcraft D17S 0.0348 6.840 0.76 10.8
Cessna `Cardinal'' RG 0.0223 7.660 0.63 13
AntonovAn-12 0.0322 11.85 0.64 15.3
Ilyushin 11-18 0.0240 10.00 0.80 16.3
Yakovlev Yak-40 0.0240 9.000 0.82 15.5
Martin B-26F 0.0314 7.660 0.75 12
McDonnell F-4 `Phantom'' 0.0217 2.820 0.70 8.8
Lockheed-Martin F-22 `Raptor'' 0.0150 2.370 0.82 10.1
Sukhoi Su-27 0.0185 3.480 0.71 11.6
Mikoyan-Gurevich MIG-29 0.0225 3.430 0.85 10.1
Mikoyan-Gurevich MIG-AT 0.0238 5.350 0.61 12.6
Fig. 1. In#uence of the fuselage on the airplane Oswald e$ciency factor.
due to lift. This leading edge suction force e!ect on a relative fuel capacity is illustrated in Fig. 2. It
can be seen that for the hypothetical tailless layout with a small wing aspect ratio (A"2) the
increase of the leading edge suction force from 0 to 0.9 is followed by the increase in Oswald
e$ciency factor up to 0.83 and the corresponding max lift-to-drag ratio rise, that produces a bene"t
O. Samoylovitch, D. Strelets / Aircraft Design 3 (2000) 167}174 169
Fig. 2. E!ect of the leading edge suction force on a relative fuel capacity.
in fuel capacity at a given range of #ight up to &13%. So, the correct shape of the leading edge
allows to increase distinctly the fuel e$ciency.
It should be noted that the wing deformation is also important for Oswald's e$ciency factor but
it can be considered at the later design stages.
Let us express the Oswald e$ciency factor for an aeroplane as
e"e
`
k
$
, (2)
where
e
`
"
e
`1
e
`1 "
S

e
`1 "
!(1!S

)e
`1
, (3)
denotes the Oswald e$ciency factor re#ecting the di!erence between the actual wing circulation
distribution and an elliptical one, and the in#uence of the leading edge suction force, and k
'
is
a correction factor to incorporate the in#uence of a fuselage cross section shape on the induced
drag.
Thus the following tasks should be solved for calculating the aeroplane is Oswald e$ciency
factor:
E wing Oswald e$ciency factor calculation with full implementation of the leading edge suction
force e
`1
"1;
170 O. Samoylovitch, D. Strelets / Aircraft Design 3 (2000) 167}174
Table 2
Coe$cients for Eq. (5)
a

a
`
a
`
a
"
a
`
1.0 0.5269 0.1230 0.0441 !0.0057 0.0032
0.5 0.4919 0.1413 0.0157 0.0054 0.0061
0.2 0.5160 0.1176 0.0156 !0.0023 0.0071
0.0 0.5694 0.1202 0.0083 !0.0028 0.0081
Fig. 3. Cross section shape factor for the elliptical fuselages.
E wing Oswald e$ciency factor calculation at zero leading edge suction force e
`1 "
;
E calculation of the relative leading edge suction force S

;
E calculation of the fuselage cross section shape factor k
'
.
Solutions of all mentioned tasks are considered below in brief.
Wing Oswald e$ciency factor calculation with full implementation of the leading edge suction
force is based on a vortex model of a simple shape wing #ow.
e
`1
"
C
'
A
y

, (4)
where y

"2y C

/b is the relative distance between the trailing vortex centres of gravity in the
Tre!etz plane; and "(1!M` the compressibility correction.
Empirical dependencies with variable coe$cients were obtained for calculation of the distance
between the vortices:
y

"a

#a
`
A#Atan
''
(a
`
#a
"
A#a
`
Atan
''
) (5)
Values of the coe$cients for di!erent taper ratios are given in the Table 2.
O. Samoylovitch, D. Strelets / Aircraft Design 3 (2000) 167}174 171
Table 3
Airplane geometrical parameters
Aeroplane Geometrical parameters
A
''
(deg.) t/c (m) S (m
2
) dM
MIG-AT 5.66 0.357 5 0.13 0.076 17.67 0.221
MIG-29 3.43 0.323 42 0.05 0.019 38 0.298
Su-27 3.48 0.294 42 0.05 0.014 62 0.3
F-22 2.37 0.263 42 0.05 0.034 78 0.32
An-12 11.85 0.357 9 0.16 0.097 121.7 0.108
Table 4
Considered #ight regimes
Aeroplane Flight conditions
M H (m)
MIG-AT 0.6 5000
MIG-29 0.85 5000
Su-27 0.85 5000
F-22 0.85 5000
An-12 0.4 4000
Wing Oswald e$ciency factor calculation at zero leading edge suction force is de"ned with
a formula
e
`1 "
"
C
'
A
. (6)
Precise calculation of the leading edge suction force is di$cult at the stage of preliminary design.
That is why a special algorithm was developed based on the simpli"ed empirical dependencies
corresponding to the data obtained by Schemensky [2]. Comparison of the calculation results with
empirical data for #at wings with symmetrical pro"les have shown good convergence in the
researched range of characteristics. It allows to recommend the following approximate expression
(applicable at relative leading-edge curvatures "/t"0.9!2.5%) for the S

initial evaluation
S

"0.974!0.0976 exp

!0.456

A
cos
''

(7)
Calculation of the fuselage cross section shape factor is carried out with the help of a graphical
dependency in the form k
'
"k
'
(dM, a ), where dM"d/b is the relative fuselage diameter at the zone of
wing attachment, and a "a/d the relative fuselage cross section size.
An example of such a dependency for fuselages with elliptic cross section is presented in Fig. 3.
172 O. Samoylovitch, D. Strelets / Aircraft Design 3 (2000) 167}174
Table 5
Calculation results and experimental data
Aeroplane Parameter e calculated by Eq. (2) Experimental data
MIG-AT e 0.82 0.61
C
'"
0.022 0.0238
(L/D)
`
12.8 12.6
MIG-29 e 0.87 0.85
C
'"
0.0225 0.0225
(L/D)
`
10.13 10.1
Su-27 e 0.86 0.71
C
'"
0.0185 0.0185
(L/D)
`
11.3 11.6
F-22 e 0.87 0.82
C
'"
0.015 0.015
(L/D)
`
10.34 10.1
An-12 e 0.73 0.64
C
'"
0.028 0.0322
(L/D)
`
15.5 15.3
The proposed method for the Oswald e$ciency factor calculation was tested on several known
aircraft samples. The results presented in the supplement allow to conclude that this engineering
method is completely applicable to the preliminary design problems.
Supplement: The Oswald e$ciency factor calculation method proposed by the authors has been
tested on several aeroplanes: MIG-AT, MIG-29, Su-27, F-22 and An-12 with geometrical para-
meters given in Table 3.
Calculations have been carried out only for one steady #ight regime for each aeroplane with
parameters listed in Table 4.
Calculation results and experimental data are presented in Table 5.
References
[1] Torenbeek E. Synthesis of subsonic airplane design. Delft: Delft University Press, 1982. p. 148.
[2] Schemensky RT. Empirical methods, vol. 1, AFFDL-TR-73-144.
Obituary
O.S. Samoylovich
The "rst author of this article sadly died before the completion of the "nal version. The following
summary of his career characterizes him as a distinguished aircraft designer and academic.
Oleg Sergeevich Samoylovich (1926}1999) was a Deserving Person of Science and Engineering
of the Russian Federation, Doctor of Technical Sciences and Professor. He graduated in 1957 with
O. Samoylovitch, D. Strelets / Aircraft Design 3 (2000) 167}174 173
the highest marks from the Moscow Aviation Institute (MAI) on the specialization `Airplane
Designa. During his main industrial career (1957}1985) O.S. Samoylovich was associated with the
design bureau installed by P.O. Sukhoy, where he started as an engineer and subsequently became
Deputy Chief Designer and Chief Designer. As Deputy General Designer he executed direct
management over the conceptual design of the T-4, Su-24, Su-25 and Su-27 aircraft. He was in
possession of 26 USSR invention certi"cates and "ve foreign patents (USA, United Kingdom,
France and Korea).
O.S. Samoylovich was a recognized leader in the development and introduction into industrial
practice of computer-aided design methods. He worked as the USSR Aviation Ministry Chief
Designer on airplane CAD systems, he was a member of an interdepartmental commission on
design automation and of the USSR Academy of Science Council on the methodology of arti"cial
intelligence. Since O.S. Samoylovich combined his industrial work with academic activities as
a professor in the Central Institute for the USSR Aviation Ministry (aimed at improving the
quali"cation of workers at the ministry) and in the Moscow Aviation Institute, he published 16
academic and scienti"c articles. He was also member of two Expert Councils of the Russian
Federation Higher Attestation Committee supervising the preparation and defence of dissertations
for the doctoral degree.
O.S. Samoylovich delivered invited lectures on aircraft design at the Massachusetts Institute of
Technology (1991), the University of Michigan (1992), the University of Tokyo (1994) and the
Technical University of Berlin (1994). He was the recipient of "ve USSR decorations, and the
following distinctions were awarded to him: the USSR Lenin Prize Laureate (1976) and USSR
Council of Ministers Premium Laureate (1981).
E. Torenbeek
Faculty of Aerospace, Delft University of Technology,
Kluyverweg 1, 2629 HS Delft, Netherlands
174 O. Samoylovitch, D. Strelets / Aircraft Design 3 (2000) 167}174