See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/257652123

CFD CALCULATIONS OF THE FLOW OVER A NACA 0012 AIRFOIL

Conference Paper · July 2010

CITATIONS READS

3 2,535

3 authors, including:

Eleni Douvi Dionissios P. Margaris

University of Patras University of Patras
21 PUBLICATIONS   107 CITATIONS    153 PUBLICATIONS   726 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

On the modelling of deepsea oil-gas discharges View project

Multiphase flow separation techniques using a new mixture separator View project

All content following this page was uploaded by Dionissios P. Margaris on 18 July 2014.

The user has requested enhancement of the downloaded file.

 4th International Conference from Scientific Computing to Computational Engineering
4th IC-SCCE
Athens, 7-10 July, 2010
© IC-SCCE

CFD CALCULATIONS OF THE FLOW OVER A NACA0012 AIRFOIL

Eleni C. Douvi, Athanasios I. Tsavalos, Dionissios P. Margaris

Fluid Mechanics Laboratory (FML)
University of Patras, Dept. of Mechanical Engineering and Aeronautics
26500 Patras, Greece
e-mail: douvi@mech.upatras.gr, margaris@mech.upatras.gr

Keywords: CFD, airfoil, aerodynamic coefficients, lift, drag, turbulence models

Abstract. Steady – state, two dimensional CFD calculations for the subsonic flow over a NACA 0012 airfoil at
various angles of attack and operating at a Reynolds number of 3×10 6 are presented. The aim of the work is to
show the behavior of the airfoil at these conditions. Looking for the most appropriate turbulence model for this
simulation, a number of different turbulence models were tested. For the validation of this simulation,
comparisons between computational results and existing experimental data from reliable sources are made. This
work highlights two areas in CFD that require further investigation: transition point prediction and turbulence
modeling. The laminar to turbulent transition point must be modeled correctly in order to get accurate results
for the drag coefficient. In addition, calculations show that the turbulence models used in commercial CFD
codes doesn’t give accurate results at high angles of attack.

INTRODUCTION
The past decade has been computational fluid dynamics (CFD) become the method of choice in the
design of many aerospace, automotive, and industrial components and processes in which fluid or gas flow play
a major role. In the fluid dynamics, there are many commercial computational fluid dynamics packages available
for modeling flow in or around objects. The computer simulations show features and details that are difficult,
expensive, or impossible to measure or visualize experimentally.
The first step in modeling a problem involves the creation of the geometry and the meshes with a
preprocessor. After the creation of the grid, a solver is able to solve the governing equations of the problem. The
basic procedural steps for the solution of the problem are the following. First, the modelling goals have to be
defined and the model geometry and grid are created. Then, the solver and the physical models are stepped up in
order to compute and monitor the solution. Afterwards, the results are examined and saved and if it is necessary
we consider revisions to the numerical or physical model parameters.
In this project, curves for the lift and drag characteristics of the NACA0012 airfoil are developed.
Dependence of C D and C L on the angle of attack is determined using three different turbulence models. The
aim of this project is to find the most appropriate turbulence model for this simulation. In fluid dynamics,
turbulence or turbulent flow is a fluid regime characterized by chaotic, stochastic property changes. This
includes low momentum diffusion, high momentum convection, and rapid variation of pressure and velocity in
space and time. In this analysis the Spalart - Allmaras model, the realizable k   model and the SST k  
model are compared[1].
The Spalart - Allmaras[2]model is a relatively simple one-equation model that solves a modelled
transport equation for the kinematic eddy (turbulent) viscosity. It was designed specifically for aerospace
applications involving wall-bounded flows and has been shown to give good results for boundary layers
subjected to adverse pressure gradients. It is also gaining popularity for turbo machinery applications. In its
original form, it is effectively a low-Reynolds-number model, requiring the viscous-affected region of the
boundary layer to be properly resolved. The near-wall gradients of the transported variable in the model are
much smaller than the gradients of the transported variables in the k   or k   models.
The simplest "complete models'' of turbulence are two-equation models in which the solution of two
separate transport equations allows the turbulent velocity and length scales to be independently determined. The
standard k   model falls within this class of turbulence model and has become the workhorse of practical
engineering flow calculations in the time since it was proposed by Launder and Spalding[3]. Robustness,
economy, and reasonable accuracy for a wide range of turbulent flows explain its popularity in industrial flow
and heat transfer simulations. It is a semi-empirical model, and the derivation of the model equations relies on
phenomenological considerations and empiricism. As the strengths and weaknesses of the standard k   model
 Eleni C. Douvi, Athanasios I. Tsavalos, Dionissios P. Margaris

have become known, improvements have been made to the model to improve its performance. Two of these
variants are available: the RNG k   model and the realizable k   model[4].
The standard k   model is based on the Wilcox k   model[5], which incorporates modifications for
low-Reynolds-number effects, compressibility, and shear flow spreading. The Wilcox model predicts free shear
flow spreading rates that are in close agreement with measurements for far wakes, mixing layers, and plane,
round, and radial jets, and is thus applicable to wall-bounded flows and free shear flows. A variation of the
standard k   model called the SST k   model is also available. The shear-stress transport (SST) k  
model was developed by Menter[6] to effectively blend the robust and accurate formulation of the k   model
in the near-wall region with the free-stream independence of the k   model in the far field. To achieve this, the
k   model is converted into a k   formulation. The SST k   model is similar to the standard k  
model, but includes some refinements: These features make the SST k   model more accurate and reliable for
a wider class of flows (e.g., adverse pressure gradient flows, airfoils, transonic shock waves) than the standard
k   model. Finally, existing experimental data from reliable sources are performed to validate the
computational results[7].

COMPUTATIONAL METHOD
For all flows, the solver solves conservation equations for mass and momentum. Additional transport
equations are also solved when the flow is turbulent. The equation for conservation of mass, or continuity
equation, can be written as follow:

 
   u   S m (1)
t
Equation (1) is the general form of the mass conservation equation and is valid for incompressible as well as
compressible flows. The source S m is the mass added to the continuous phase from the dispersed second phase
(e.g., due to vaporization of liquid droplets) and any user-defined sources.
Conservation of momentum in an inertial (non-accelerating) reference frame is described by equation (2)

  
t

u     uu   p    τ  g  F (2)

 
where p is the static pressure, τ is the stress tensor (described below), and g and F are the gravitational

body force and external body forces (e.g., that arise from interaction with the dispersed phase), respectively. F
also contains other model-dependent source terms such as porous-media and user-defined sources.
The stress tensor τ is given by

 2
   u  u T     uI
  
(3)
 3 

where  is the molecular viscosity, I is the unit tensor, and the second term on the right hand side is the effect
of volume dilation.
For the 2-D, steady and incompressible flow the continuity equation is:

u v
 0 (4)
x y

Momentum equations for viscous flow in x and y direction are respectively:

Du p τ xx τ yx
ρ     ρf x (5)
Dt x x y

Dv p τ τ
ρ    xy  yy  ρf y (6)
Dt y x y

where due to characteristics of the 2-D flow in continuity equation the term w z and in momentum equation,
τ zx z and τ zy z drop out.
 Eleni C. Douvi, Athanasios I. Tsavalos, Dionissios P. Margaris

An airfoil from the 4-digit series of NACA airfoils is utilized, NACA 0012. The NACA 0012 airfoil is
symmetrical; the 00 indicates that it has no camber. The 12 indicates that the airfoil has a 12% thickness to chord
length ratio: it is 12% as thick as it is long. Reynolds number for the simulations and the experimental data is
Re=3x106. A segregated, implicit solver is utilized. Calculations are done for angles of attack ranging from -12
to 20 degrees.
The airfoil profile, boundary conditions and meshes are all created in the pre-processor. The pre-
processor is a program that can be employed to produce models in two and three dimensions, using structured or
unstructured meshes, which can consist of a variety of elements, such as quadrilateral, triangular or tetrahedral
elements. The resolution of the mesh is greater in regions where greater computational accuracy is needed, such
as the region close to the airfoil. Present simulations used a C-type grid topology with 80000 quadrilateral cells
that is presented in figure 1.

Figure 1. Mesh around NACA0012 airfoil (left) and detail closed to the airfoil (right).

RESULTS
Simulations for various angles of attack are done in order to be able to compare the results from the
different turbulence models and then validate them with existing experimental data. To do so, the model is
solved with a range of different angles of attack from -12o to 20o.
On an airfoil, the resultants of the forces are usually resolved into two forces and one moment. The
component of the net force acting normal to the incoming flow stream is known as the lift force and the
component of the net force acting parallel to the incoming flow stream is known as the drag force. The
characteristics are described by the dimensionless quantities

2L
C  (7)
L
U 2S

2D
C  (8)
D
U 2S

where L and D are the lift and drag force respectively,  is the density of the fluid, U  is the upstream
velocity and S is the wing area. Below are the curves of the lift and the drag coefficient for various angles of
attack, computed with three turbulence models and compared with experimental data.
At Figure 2 it could be observed that at low angles of attack, the dimensionless lift coefficient increases
linearly with angle of attack. Flow is attached to the airfoil throughout this regime. At an angle of attack of
roughly 15o or 16o, the flow on the upper surface of the airfoil begins to separate and a condition known as stall
begins to develop. The realizable k   model and the SST k   model did not have a good agreement with the
experimental results. It is obvious that the Spalart - Allmaras turbulence model is the most appropriate for this
simulation.
Near stall, disagreement between the data is shown. The lift coefficient peaks and the drag coefficient
increases as stall increases. The predicted drag coefficients are higher than the experimental results. This
overprediction of drag was expected since the actual airfoil has laminar flow over the forward half. In order to
 Eleni C. Douvi, Athanasios I. Tsavalos, Dionissios P. Margaris

get more accurate results, the computational domain could be splitted into two different domains to run mixed
laminar and turbulent flow. The disadvantages of this approach are that the accuracy of simulations

2,0

1,5

Lift Coefficient, CL 1,0

0,5

0,0

-0,5

-1,0

-1,5
-20 -10 0 10 20 30
angle of attack, a

experimental Spalart-Allmaras
k-ω SST k-ε Realizable

Figure 2. Comparison between experimental data and three different turbulent models simulation results of the
lift coefficient curve for NACA 0012 airfoil.

0,025

0,020
Drag Coefficient, CD

0,015

0,010

0,005

0,000
-20 -10 0 10 20
angle of attack, a

experimental Spalart-Allmaras
k-ω SST k-ε Realizable

Figure 3. Comparison between experimental data and three different turbulent models simulation results of the
drag coefficient curve for NACA 0012 airfoil.

depends on the ability to accurately guess the transition location, and a new grid must be generated if the
transition point has to change[8].
 Eleni C. Douvi, Athanasios I. Tsavalos, Dionissios P. Margaris

Near stall, disagreement between the data is shown. The lift coefficient peaks and the drag coefficient
increases as stall increases. The predicted drag coefficients are higher than the experimental results. This
overprediction of drag was expected since the actual airfoil has laminar flow over the forward half. In order to
get more accurate results, the computational domain could be splitted into two different domains to run mixed
laminar and turbulent flow. The disadvantages of this approach are that the accuracy of simulations depends on
the ability to accurately guess the transition location, and a new grid must be generated if the transition point has
to change[3].
At Figure 4 and Figure 5 are the simulation outcomes of static pressure at angles of attack 3o and 9o
with the Spalart – Allmaras turbulence model which is the most appropriate for this problem.

Figure 4. Contours of static pressure at 3o angle of attack.

Figure 5. Contours of static pressure at 9o angle of attack.

Figure 4 and figure 5 demonstrate the pressure distribution over the airfoil. The pressure on the lower
surface of the airfoil is greater than that of the incoming flow stream and as a result it effectively “pushes” the
airfoil upward, normal to the incoming flow stream. On the other hand, the components of the pressure
distribution parallel to the incoming flow stream tend to slow the velocity of the incoming flow relative to the
airfoil, as do the viscous stresses.
Contours of velocity components at angles of attack 3 o, 9o and 16o are also presented. The trailing edge
stagnation point moves slightly forward on the airfoil at low angles of attack and it jumps rapidly to leading edge
at stall angle. A stagnation point is a point in a flow field where the local velocity of the fluid is zero.
 Eleni C. Douvi, Athanasios I. Tsavalos, Dionissios P. Margaris

Figure 6. Contours of velocity magnitude at 3o angle of attack.

Figure 7. Contours of velocity magnitude at 9o angle of attack.

Figure 8. Contours of velocity magnitude at 16o angle of attack.

The upper surface of the airfoil experiences a higher velocity compared to the lower surface. That was
expected from the pressure distribution. As the angle of attack increases the upper surface velocity is much
 Eleni C. Douvi, Athanasios I. Tsavalos, Dionissios P. Margaris

higher than the velocity of the lower surface. Figure 9 demonstrates the stagnation point for various angles of
attack.

position on chord length, x

0,06
0,05
0,04
0,03
0,02
0,01
0,00
-20 -10 0 10 20
angle of attack, a

Figure 9. Stagnation point for various angles of attack.

CONCLUSION
This paper shows the behaviour of the 4-digital symmetric airfoil NACA0012 at various angles of
attack. The most appropriate turbulence model for these simulations is the Spalart - Allmaras one-equation
model. We have to highlight two areas in CFD that require further investigation and development in order to
enable accurate numerical simulations of flow around airfoils at high angles of attack: transition prediction and
turbulence modelling. Transition from laminar to turbulent regime has to be predicted and then the
computational grid to split in two regions, a laminar and a turbulent region. Turbulence modelling is also very
important for the accuracy of the results.

REFERENCES
[1] Fluent Inc.(2006), FLUENT 6.3 User’s Guide
[2] Spalart, P. R. and Allmaras, S. R. (1992), “A One-Equation Turbulence Model for Aerodynamic Flows”,
AIAA Paper, 92-0439.
[3] Launder, B.E., Spalding, D.B. (1974), “The numerical computation of turbulent flows”, Computer Methods
in Applied Mechanics and Engineering, Vol. 3, pp.269-89.
[4] Shih T.H., Liou W. W., Shabbir A. and Zhu J. (1995), “A new k-ε eddy – viscosity model for high Reynolds
number turbulent flows – model development and validation”, Computers Fluids, Vol. 24, pp 227-238.
[5] Wilcox, D.C. (1988), “Re-assessment of the scale-determining equation for advanced turbulence models”,
AIAA Journal, Vol. 26, pp. 1414-1421.
[6]F. R. Menter., (1994), “Two-Equation Eddy-Viscosity Turbulence Models for Engineering Applications”
AIAA Journal, Vol. 32, pp. 1598-1605
[7] Abbott I. H., Von Doenhoff A.E.(1959), Theory of Wing Sections, Dover Publishing, New York.
[8] Silisteanu P.D., Botez R.M, (2010), “Transition flow occurrence estimation new method”, 48th AIAA
Aerospace Science Meeting including The New Horizons Forum and Aerospace Exposition, Orlando, Florida,
Etats - Unis, 7-10 janvier

View publication stats