Aerodynamic Characteristics of Airfoil with Single

Plain Flap for Light Airplane Wing

Michael TODOROV
Technical University of Sofia, Kliment Ohridski 8, Sofia 1000, Bulgaria, e-mail: michael.todorov@tu-sofia.bg

Abstract—A numerical study was performed on a NACA 23012 unstructured or hybrid. The structured mesh is identified by
airfoil with a single plain flap to examine the aerodynamic regular connectivity. The possible element choices are
coefficients at Reynolds number of 3×106, and to help for quadrilateral in 2D and hexahedral in 3D. This model is highly
identifying the forces acting on a light airplane wing. Besides, in space efficient, i.e. since the neighborhood relationships are
the paper the flow fields around the airfoil with single plain flap defined by storage arrangement. Some other advantages of
were shown. All calculations were made using a CFD code. For a structured mesh are better convergence and higher resolution.
turbulent model the Spalart-Allmaras method was chosen. The unstructured mesh is identified by irregular connectivity,
Conclusions were made about the aerodynamic efficiency of the [8], [13]. It cannot easily be expressed as a two-dimensional or
proposed configuration wing-single plain flap.
three-dimensional array. This allows a solution, which might
Keywords-airplane wing; single plain flap; aerodynamic
be used for any possible element. Compared to structured
characteristics; CFD; Fluent. meshes, this model can be highly space inefficient since it calls
for explicit storage of neighborhood relationships. These grids
typically employ triangles in 2D and tetrahedral in 3D. A
I. INTRODUCTION hybrid mesh contains of a mixture of structured portions and
Aircraft wing high-lift configuration design is an important unstructured portions. It integrates the structured meshes and
and challenging part of the whole aircraft aerodynamic the unstructured meshes in an efficient manner, [4].
configuration design, even dealing with a 2D high-lift
Another important step is the choice of a turbulent model.
configuration design task which is an essential step for the 3D
The turbulence is the most challenging area in fluid dynamics
high-lift configuration design [1], [2], [3], [4].
and the most limiting factor in accurate computer simulation of
During the take-off and landing of an aircraft, performance the flow. An overview of turbulence modeling is done in [14],
of high-lift devices has strong impact on the operating costs Fig. 1.
and environment around airports, such as improvements of
payload, fuel consumption, and noise emission. Take-off and
landing performance for very light airplanes are governed by
the requirements as EASA CS-VLA [5]. The take-off and
landing distances, and the important speeds as the stall speed
with flaps retracted – VS, the design maneuvering speed – VA,
the speed with flaps fully deflected – VF, and the stall speed
with flaps fully deflected – VSF, depend on aerodynamic
characteristics of the wing with a flaps deflected.
Nowadays, Computational Fluid Dynamics (CFD) is
widely used for the prediction of the aerodynamic performance
of the wing, at least in cruise flight. The computation of the
flow over a multi-element wing in high-lift configuration
remains, however, one of the most difficult problems
encountered in CFD [6]. The computations normally include a
comprehensive code, coupled to Euler or Navier-Stockes
solvers. The examples for a successful application of CFD are Figure 1. A schematic overview of turbulent modeling [14].
the codes FLUENT, OVERFLOW of NASA, FLOWer and
TAU of Deutshes Zentrum für Luft und Raumfahr [4], [6], The direct numerical simulation (DNS) of Navier-Stockes
elsA and WAVES of ONERA [7], CFD++ [8], Star-CCM+ [9], equation can be applied only for very low Reynolds numbers
[10], TAS of Takoku University and UPACS of Japan Institute and very simple and limited geometry, and practically no
of Space Technology and Aeronautics [11, 12]. numerical solution for flows of interest to engineers can be
obtained. Large eddy simulation (LES) solves the spatially
The high-lift configurations considerably complicate the averaged Navier-Stockes equations. Large eddies are directly
flow physics by boundary layer transition, separations and resolved, but eddies smaller than the mesh are modeled. LES is
reattachments. Therefore, it is very important to generate the less expensive than DNS, but the amount of computational
appropriate mesh around it. The mesh can be structured, resources and efforts are still too large for most practical
applications. Spalart-Allmaras is a low-cost Reynolds- - k-ε method
Averaged Navier-Stockes (RANS) model solving a transport
equation for a modified eddy viscosity. It is designed - k–ω SST method
specifically for aerospace applications involving wall-bounded - Scale-adaptive simulation (SAS) method.
flows. It embodies a relatively new class of one-equation
models where it is not necessary to calculate a length scale
related to the local shear layer thickness. The k-ε turbulent
models are the most widely-used engineering turbulence model
for industrial applications, which are a robust and a reasonably
accurate. However they generally perform poorly for flows
with strong separation, large streamline curvature, and large
pressure gradient. The k–ω turbulence models have gained
popularity mainly because the model equations do not contain
terms which are undefined at the wall, i.e. they can be
integrated to the wall without using wall functions, and they are
accurate and robust for a wide range of boundary layer flows
with pressure gradient. Reynolds stresses are solved directly
using transport equations, avoiding isotropic viscosity
assumption of other models. It uses for highly swirling flows.
The quadratic pressure-strain option improves performance for
many basic shear flows [15]. Figure 3. View of the mesh geometry of NACA 23012 airfoil.

The following study aims to identify the aerodynamic
characteristics of an airfoil with single plain flap, as outlined in
Fig. 2. This configuration is very easy for manufacturing from
a technological point of view. A NACA 23012 airfoil with a
1.00 m chord has been used in all the CFD simulations. The
single plain flap with a 0.32 m chord, corresponding to 32 %
chord, was constructed in such a way as to match the geometry
of the baseline airfoil.
Figure 2. NACA 23012 airfoil with single plain flap.
II. AERODYNAMIC CHARACTERISTICS OF NACA 23012
AIRFOIL. CHOICE OF AN ADEQUATE TURBULENCE MODEL.
COMPARISON WITH EXPERIMENTAL WIND TUNEL DATA
The calculations are made for Reynolds number of
Re = 3 × 10 6 (respectively V = 43.81 m/s) at the sea level in
accordance to the experimental data [16]. The turbulent
intensity and turbulent viscosity are 2.48 % and 10
respectively. The results are shown in Fig. 4 and Fig. 5.
The Fig. 4 shows that the Spalart-Allmaras method gives
closest results to the experimental data. Besides this method is
the fastest with relatively good convergence. The SAS method
also gives good results, but after an angle of attack the lift
coefficient is still growing. The k–ω SST method follows the
experimental data, but for high angles of attack it does not have
a good convergence. The k-ε method after angle of attack of 8º
does not yield results because of a divergence. Fig. 5 shows
that all methods give higher values for the drag coefficient, but
it can be seen in results in similar CFD applications. The
reason is that in the calculations the boundary layer is fully
turbulent without a laminar region as in the real flows.
2

First, the aerodynamic characteristics of NACA 23012

airfoil will be obtained by commercial CFD code FLUENT 1.5
14.0. The obtained numerical results will be compared with
experimental wind tunnel data [16] to establish adequacy of the
modeling. The mesh density and the selection of a turbulence 1
model are critical to the accuracy of the CFD solution.
CL

The NACA 23102 airfoil is generated by 298 points from 0.5

[17].
Wind tunel experiment
The structured mesh is generated, as it is shown in Fig. 3, Spalart-Allmaras method
0
using ANSYS Workbench 14. The circle has a 10c (c-airfoil K-ε method
chord) radius. Around the airfoil and downstream is provided a SAS method
high refinement. Thus the mesh has 180704 nodes and 180125 -0.5
K-ω method

elements. -5 0 5 10 15 20 25
α , deg
The following turbulent methods are used in the modeling:
Figure 4. Lift coefficient CL plot over the range of angles of attack of the
- Spalart-Allmaras method numerical results and experimental data.
 2
Wind tunel experiment
0.055
Spalart-Allmaras method
0.05 K-ε method
SAS method 1.5
0.045 K-ω method
0.04
1
0.035
CD

CL
0.03

0.025 0.5
0.02

0.015
0
0.01
NACA23012 Airfoil Section
0.005 NACA23012 Airfoil Section with Single Plain Flap
-5 0 5 10 15
α , deg -0.5
-5 0 5 10 15 20 25
α , deg
Figure 5. Drag coefficient CD plot over the range of angles of attack of the
numerical results and experimental data. Figure 7. Lift coefficient CL plot over the range of angles of attack of the
numerical results for a NACA 23012 airfoil and NACA 23012 airfoil with a
On the basis of the obtained results the Spalart-Allmaras single plain flap.
method is used in the further calculations.
0.6
III. AERODYNAMIC CHARACTERISTICS OF NACA 23012 NACA23012 Airfoil Section
NACA23012 Airfoil Section with Single Plain Flap
AIRFOIL WTH SINGLE PLAIN FLAP
0.5
To calculate the aerodynamic characteristics of NACA
23012 airfoil with single flap, the multi-object hybrid mesh is
0.4
generated, as it is shown in Fig. 6. The circle has also a 10c (c-
airfoil chord) radius. Around the airfoil, flap and downstream
are provided with a high refinement. Thus the mesh has
CD

0.3
201063 nodes and 200176 elements.
The calculations are made for Reynolds number of 0.2

Re = 3 × 10 6 (respectively V = 43.81 m/s) at the sea level. The

turbulent intensity and turbulent viscosity are 2.48 % and 10 0.1

respectively. The flap deflection angle, δF, is 20º.

0
Fig. 7 and Fig. 8 show CL-α and CD- α data of the numerical -5 0 5 10 15 20 25
results for a NACA 23012 airfoil and NACA 23012 airfoil with α , deg

single plain flap.

Figure 8. Drag coefficient CD plot over the range of angles of attack of the
Figs. 9-13 show pressure and velocity fields around NACA numerical results for a NACA 23012 airfoil and NACA 23012 airfoil with a
23012 airfoil with a single plain flap in the range from 0º to 20º single plain flap.
angles of attack.
IV. DISCUSSION
The deflection of the flap results in an increase in the lift
coefficient by about 30 %, in the range of angles of attacks
from 0 to 12º (see Fig. 7). The maximum angle of attack in the
case with deflected flap is at a 12º angle of attack, while in the
case with the baseline airfoil it is 18º (see Fig. 7). This is
because after α = 12º the flow is fully detached on the upper
surface, as it is shown in Figs. 12 and 13. Furthermore, the
deflection of the flap increases the drag coefficient by about
82 % at small angles of attack (see Fig. 8), as at the high angles
of attack that increase is much higher. This is a result of the
detachment of flow on the upper surface of the wing section,
and the drag by the loss of pressure due to the vertical cutting
of the wing airfoil before the flap.
Figure 6. View of the mesh geometry of NACA 23012 airfoil with single
plain flap.
Figure 9. Pressure and velocity fields around NACA 23012 airfoil with a single plain flap at α = 4º.

Figure 10. Pressure and velocity fields around NACA 23012 airfoil with a single plain flap at α = 8º.

Figure 11. Pressure and velocity fields around NACA 23012 airfoil with a single plain flap at α = 12º.
 Figure 12. Pressure and velocity fields around NACA 23012 airfoil with a single plain flap at α = 16º.
Figure 13. Pressure and velocity fields around NACA 23012 airfoil with a single plain flap at α = 20º.
These results show that the chosen arrangement of wing-
single plain flap is not sufficiently effective from aerodynamic
point of view, although it is attractive with a simple design.
V. CONCLUSION
A numerical analysis was performed for a NACA 23012
airfoil. The comparison with experimental wind tunnel data
permits to select the appropriate turbulent method, when
calculating the aerodynamic characteristics of a NACA 23012
airfoil with a single plain flap.
The analysis aimed to identify the aerodynamic forces
acting on the proposed wing and flap at Reynolds number of
3×106.
The 2D CFD model was used to examine the major features
around the proposed configuration airfoil-flap.
The CFD results showed that lift coefficient increased by
30 %, but the maximum angle of attack decreased from 18º to
12º. On the other hand, the drag coefficient increased by more
than 82 %. These results are not good from the aerodynamic
point of view.
Therefore the further work is needed to investigate the
aerodynamic characteristics of other designed configuration
airfoil-flap.
ACKNOWLEDGMENT
The author thanks to D-r Ivan Dobrev from the Laboratory
of Fluid Mechanics at Arts et Metiers ParisTech in the
development of the paper.
REFERENCES
[1] C.P. van Dam, “The aerodynamic design of multi-element high-lift
systems for transport airplanes,” Progress on Aerospace Sciences 38
(2002), pp. 101-144.
[2] A. Prabhakarr and A. Ohri, “CFD analysis on MAV NACA 2412 wing
in high lift take-off configuration for enhanced lift generation,” J.
Aeronautics & Aerospace Engineering, volume 2, issue 5, 2013, pp. 1-8.
[3] R. G. F. Steed, “High lift CFD simulation with an SST-based predictive
laminar to turbulent transition model,” AIAA 2011-864.
[4] K. Schindler, D. Reckzeh and U. Scholz, “Aerodynamic design of high-
lift devices for civil transport aircraft using RANS CFD,” AIAA 2010-
4946.
[5] EASA, “Certification specifications for very light aeroplanes CS-VLA,”
Amendment 1, March 2009.
[6] J. Brezillon, R. P. Dwight and J. Wild, “Numerical aerodynamic
optimization of 3D high-lift configuration,” Proc. of 26th International
Congress of The Aeronautical Sciences, ICAS 2008.
[7] L. Cambier, S. Heib and S. Plot, “The ONERA CFD software: input
from research and feedback from industry,” Proc. Of 28th International
Congress of The Aeronautical Sciences, ICAS 2012.
[8] A. Khare, R. Baig, R. Ranjan, S. Shah, S. Pavithran, K. Nikam and A.
Moitra, “Computational simulation of flow over a high-lift trapezoidal
wing,” Proc. of ICEAE 2009.
[9] P. D. Vecchia and D. Ciliberty, “Numerical aerodynamic analysis on a
trapezoidal wing with high lift devices: a comparison with experimental
data,” XXII AIDAA Conference, September 9-12, 2013 Italy.
[10] P. Shankara and D. Snyder, Numerical simulation of high lift trap wing
using STAR-CCM+,” AIAA 2012-2920.
[11] M. Murayama, Y. Yokokawa and K. Yamamoto, “Validation study of
CFD analysis for high-lift systems,” Proc. Of 25th International Congress
of The Aeronautical Sciences, ICAS 2006.
[12] M. Murayama and K. Yamamoto, “Numerical simulation of high-lift
configurations using unstructured mesh method,” Proc. Of 24th
International Congress of The Aeronautical Sciences, ICAS 2004.
[13] J. A. de Oliveira Neto, D. Cavali, C. B. Junior, J.L.F. Azavedo and A. L.
F. de Lima e Silva, “Computational simulations of high-lift
configuration using unstructured grids,” Proceedings of the ENCIT
2006, ABCM, Curitiba – PR, Brazil – Paper CIT06-0269.
[14] B. Anderson, R. Anderson, L. Hakanson, L. Mortensen, R. Sudiyo and
B. van Wachem, “Computation fluid dynamics for engineers,”
Cambridge University Press, 2012.
[15] www.fluentusers.com
[16] I. H. Abbott and A. E. von Doenhoff, “Summary of airfoil data,” NACA
Report No. 824, 1945.
[17] www.ppart.de/aerodynamics/profiles/NACA5.html