Introductory FLUENT Training: Workshop 2 Transonic Flow Over A NACA 0012 Airfoil

Workshop 2
Transonic Flow over a

NACA 0012 Airfoil
Introductory
FLUENT Training
ANSYS, Inc. Proprietary April 28, 2009

© 2009 ANSYS, Inc. All rights reserved. WS2-1 Inventory #002601
WS2: Transonic Flow Over a NACA 0012 Airfoil
Goals Workshop Supplement
• The purpose of this tutorial is to introduce the user to good

techniques for modelling flow in high speed external aerodynamic
applications.
• Transonic flow will be modelled over a NACA 0012 airfoil for which
experimental data has been published, so that a comparison can be
made.
• The flow to be considered is compressible and turbulent.
• The solver used is the density based implicit solver, which gives
good results for high speed compressible flows.
• The tutorial is carried out using FLUENT and CFD Post from within
Workbench, but it could also be completed in standalone mode.

Start a workbench project Workshop Supplement
• Launch Workbench and save the new project as naca0012 in your

working directory.
• Double-click or drag a FLUENT module from the component systems.
• Add a results module – double click or drag.
• Drag the mouse from cell A3 (Solution) to B2 (Results) to couple the
modules.

Import a mesh that was generated in Gambit Workshop Supplement
• Import the FLUENT mesh file (naca0012.msh).

– Right click on cell A2 (setup) and select ‘import FLUENT case file’
– Change the ‘Files of type’ to “FLUENT mesh file”
– Select the mesh file naca0012.msh
• The FLUENT launcher will start.
– Keep the default options.
– Note that ‘2D’ has automatically been selected

Mesh Workshop Supplement
• FLUENT will launch in a new window.

• The mesh will read in and display, and the zones will be written out
for the Workbench project.

• The mesh needs scaling, since it was created with lengths in mm.
• Select General > Scale and observe the current domain extents.
– Select ‘Mesh was created in mm’.
– Press ‘Scale’
– Check that the domain extents are as expected.
– Close the scale panel and select General > Check
• Review the text window and check there are no errors.
• Finally use ‘Report Quality’ to print out cell quality statistics.

• Zoom in and examine the mesh.

• The maximum aspect ratio in this mesh is quite high (around 7000)
• This is acceptable because these cells are close to the airfoil wall
surfaces.
• This is needed for the turbulence model being used, since it ensures the
first grid point is in the viscous sublayer.

Solver Workshop Supplement
• Select the steady-state density-based solver:

– From ‘General’ in the tree select Type: Density-Based
– Check time is steady
• Turn on the energy equation.

– This is needed because the flow is compressible and we will be using the
ideal gas equation.
– From ‘Models’ in the tree, select ‘Energy’ > Edit > and check box
• Select the turbulence model to be used:

• From ‘Models’ in the tree, select ‘Viscous’ and Edit
• Choose the one-equation Spalart-Allmaras model.
• Select strain/vorticity based production, then OK
• This is a relatively simple turbulence model that has been shown to give
good results for boundary layers subjected to adverse pressure
gradients, particularly where there is no or only mild separation.

Materials Workshop Supplement
• The properties to be used for the material ‘air’ need to be set.

– Select ‘Materials’ from the model tree
– Highlight ‘Air’ then Create/Edit
– For Density, select ‘Ideal Gas’
– For Viscosity, select “Sutherland”, and accept the default settings for the
3 Coefficient method.
• The Sutherland law for viscosity is well suited for high-speed compressible
flow. For simplicity, we will leave Cp and Thermal Conductivity as constants.
Ideally, in high speed compressible flow modeling, these should be
temperature dependent as well.
– Select Change/Create
• Assign the material ‘air’ to the grid cells:

– Select ‘Cell Zone Conditions’
– Highlight ‘fluid’ then ‘Edit’
– Observe ‘air’ is already selected.

Operating Conditions Workshop Supplement
• Set the Operating Pressure to Zero:

– Absolute pressure is the gauge pressure plus the operating pressure.
– Setting zero operating pressure means that all pressures set in FLUENT
will be absolute. This is the most common practice for compressible
flows.
– Select ‘Cell Zone Conditions > Operating Conditions
– Set the Operating Pressure to Zero, then ‘OK’

Boundary Conditions Workshop Supplement
• Set the upstream boundary conditions:

– Select ‘Boundary Conditions’ > pressure-far-field-1 > edit
– The pressure-far-field boundary type is applicable only when the density is
calculated using the ideal-gas law. It is important to place the far-field boundary
far enough from the object of interest. For example, in lifting airfoil calculations, it
is not uncommon for the far-field boundary to be a circle with a radius of 20 chord
lengths.
– On the ‘Momentum’ tab, set the gauge static 
pressure to 73048 Pa po     1  2   1
– We need to input static pressure for a far-field  1   M 
p   2  
boundary. We can calculate this from the total
pressure, which was atmospheric at 101325 Pa
for the wind-tunnel test. where
– In the case of a real external aerodynamic simulation, po  total pressure  101325 Pa
rather than a wind tunnel, the static pressure (at a p  static pressure
given altitude) would actually be the same as the total   1.4 for air
pressure in the far field, because the air in the far field
would be stationary. M  Mach No.  0.7
– We have already set the operating pressure to zero, so  po  1.3871
we are now working in absolute pressure values. Hence p
the gauge static pressure input will be equal to the p  73048 Pa
absolute static pressure value, which we will calculate
to be 73048 Pa.
• Set the Mach Number to 0.7 and flow direction components as shown.
• The angle of attack (α) in this numerical case is 1.53 deg. The x-component of
the flow is cos α and the y-component is sin α.
– It is common practice to adjust the numerical α from the experimental α in order to match the lift obtained in
the wind tunnel, and then to determine the drag associated with this lift. This adjustment of α is carried out to
counter the effects of the wind tunnel enclosure.
• Set a reasonable boundary condition for the

far field turbulence:
• In reality the far-field air would be stationary. Wind
tunnels attempt to replicate this by using filters and
grids to obtain a low turbulence intensity at the
inlet.
• Select ‘Intensity and Length Scale’
• Set an intensity of 0.01%
• Choose a length scale proportional to
the boundary layer thickness. Based on
an estimated maximum boundary layer
thickness of 50mm*, a suitable length
scale is 0.4 x 0.05m = 0.02m
* taken from a previous simulation

• Select the thermal tab.

– The wind tunnel operating conditions for validation test data give the
total temperature as T0 = 311 K
– We can therefore calculate the static temperature to be 283.24 K
T0    1 2
 1  M
T  2 
where
T0  total temperatur e  311K
T  static temperatur e
  1.4 for air
M  Mach number  0.7
T
 0  1.098 and so T  283.24 K
T

• For both walls representing the airfoil, leave the default settings
which correspond to a no-slip condition for momentum and adiabatic
for thermal.

Reference Values Workshop Supplement
• Set the reference values: These are not used in the actual solution,
but are used for reporting coefficients, such as Cp.
– Use the freestream conditions as a reference, so choose ‘compute from’
then select ‘pressure-far-field-1’ in the drop down list.
Note the reference values for density,

enthalpy, pressure, temperature,
etc. will update from the freestream
values you specified in the
pressure-far-field-1 boundary.
– Set the reference length (which is not

updated from the far field boundary
values). In this 2D case, we will use
the airfoil chord length, of 1m.

Solution Methods Workshop Supplement
• The CFD computation is now defined. However the solver settings need to
be modified. These dictate how fast, stable and accurate (within the mesh
and BC constraints) the solution will be.
• Select Solution Methods in the LHS tree.
• Keep the default settings for the implicit formulation and Roe-FDS flux type.
– The explicit formulation is only normally used for cases where the characteristic
time scale is of the same order as the acoustic time scale, for example the
propagation of high Mach number shock waves.
– The implicit formulation is more stable and can be driven much harder to reach a
converged solution in less time.
Solution Methods Workshop Supplement
• Change the gradient method to Green-Gauss Node Based.

– This is slightly more computationally expensive than the other methods
but is more accurate.
• Select Second Order Upwind for flow and turbulence discretization.
– To accurately predict drag, select the ‘Second Order Upwind’ schemes.

Solution Controls Workshop Supplement
• The Courant number (CFL) determines the internal time step and
affects the solution speed and stability.
– The default CFL for the density-based implicit formulation is 5.0. It is
often possible to increase the CFL to 10, 20, 100, or even higher,
depending on the complexity of your problem. You may find that a lower
CFL is required during startup (when changes in the solution are highly
nonlinear), but it can be increased as the solution progresses.
As we will be using automatic

‘solution steering’, the choice of CFL
at this stage is not important for this
case.
Keep the default under-relaxation

factors (URFs) for the uncoupled
parameters.

Solution Monitors Workshop Supplement
• Set up residual monitors so the convergence can be monitored

– Monitors > Residuals > Edit
– Make sure ‘plot’ is on
– Turn off convergence checks by setting the criterion to ‘none’. This
means that the calculation will not stop based on the residual plots
convergence, but you can still see their progress.

Solution Monitors Workshop Supplement
• Set up a monitor for the drag coefficient on the airfoil.

• Select both wall zones and toggle on ‘Print’, ‘Plot’ and ‘Write’.
• Remember that α is 1.53° so we need to use the force vector as shown. -Lift and drag
are defined relative to the wind, not the airfoil.
You can specify

which window
FLUENT uses to
display plots. For
now, accept the
defaults.
• Press OK, then follow the same process to setup a monitor for Lift.
– The settings are identical except for the File Name (cl-history instead of cd-history) and the
Force Vectors defined as shown here:

Solution Initialization Workshop Supplement
• Initialize the flow field based on the far-field boundary:

– Select Solution Initialization from the model tree
– Compute from > pressure-far-field-1
– Press ‘Initialize’.
• Solution Steering enables the robust first order discretization in the

early-stages of the computation, then blends to the more accurate
second order schemes as the solution stabilizes.
– Select Run Calculation, and toggle on Solution Steering
– Change the flow type to transonic and keep default options
• Full-Multi-Grid Initialization will compute a quick, simplified solution based on
a number of coarse sub-grids. This will then be used as a starting point for the
main calculation. FMG can help to get a stable starting point.

Case Check Workshop Supplement
• Check the case file and make sure there are no reported issues.
– Use Run Calculation > Check Case
– Any potential problems with the case setup will be raised in the case
check panel if there are no problems this panel will not appear. In this
case there is a recommendation to check the reference values for the
force monitors. Since we have already set these we can ignore this
warning.
• Save the case file.

– File > Save Project (if running under workbench)

Run Calculation Workshop Supplement
• Although the calculation is ready to compute, It is good practice (but

not strictly necessary) to run the FMG and then check the coarse
FMG solution before starting the main calculation iterations.
• Set the number of requested iterations to zero, and press ‘Calculate’.

• Check the pressure and velocity contours to make sure that no

spurious values are predicted.
• Go to ‘Graphics and Animations in the LHS tree, choose ‘Contours’
and ‘Set Up’

• Choose Contours of Pressure > Static Pressure and ‘Filled’

• Display. If you need to autoscale the display press <control> A
• Zoom in as required.
• Examine the min and max reported values.
• Repeat for Contours of Velocity> Mach Number.

• There are no spurious results from the FMG, so proceed to the main
calculation.
• Return to ‘Run Calculation’ in the LHS tree.
• Change the number of windows to three (for the residual, drag and lift
monitors that we set up earlier).
• Request 900 iterations.
• ‘Calculate’

• After 900 iterations the calculation has fully converged.

– Note that the CFL has been updated during the calculation in a number of
stages, ramping up from 5 to 200 as we requested. This can be seen in
the CFL window and the effect on the residuals is also evident. By the
end of the calculation the residuals have converged well and are no
longer changing. The drag and lift monitors are also stable.

Post Processing [FLUENT] Workshop Supplement
• Select ‘Graphics and Animations’ in the LHS menu

• Examine the contours of static pressure.
– Turn off ‘Filled’ to just display the The contour will display in the active
contour lines. window (click a window to activate).
Alternatively, use the drop down menu to
– Adjust the Levels to increase the return the display to a single window as
number of contour lines. shown here

• Plot contours of Velocity > Mach Number and notice that the flow is
now locally supersonic.

• Select ‘Plots’ in the LHS menu.

• Plot Pressure Coefficient along the top and bottom airfoil surfaces.

• Compare experimental pressure coefficient plots which we can

import and plot here alongside the numerical prediction.
• Click on ‘Load File’ and browse for the files in your directory.

• Once loaded, plot the CFD and experimental Cp plots together.

• A good agreement can be seen.

• In order to obtain a good drag prediction, and for the turbulence

model to work effectively, we need to have a mesh that is well
resolved near to the wall, such that the first grid point is in the
viscous sub-layer. Ideally we want a Y+ value of 1 or less.
• Plot Turbulence > Y+, along both of the airfoil walls.
– Deselect the Pressure Coefficient File Data.
• We can see that this is achieved here, the max Y+ is 0.75

• Compare the predicted Cl and Cd against the experimental values.
• From Reference 1
– Cl = 0.241 and Cd = 0.0079
• From the console window, we have predicted

– Cl = 0.241 and Cd = 0.0083
– Again, good agreement can be seen.

Post Processing Workshop Supplement
• Save the project from the FLUENT file menu .

– Take the middle option ‘Continue after replacing settings file’)
• Close FLUENT (File > Close FLUENT)
• Additional post-processing will now be performed in CFD Post.
• Return to the Workbench Project window.
• Click on ‘Update Project’ and notice the Results panel update.
• Right click on cell B2 (Results) and select edit. This will launch CFD Post.

• Note that CFD Post works in 3D, so a unit thickness will be added to
the 2D airfoil, with symmetry side boundaries.

• Insert a new Contour and accept the default name Contour 1

– Top menu > Insert > Contour
• Choose the location as symmetry-1
• Choose the variable to be pressure and ‘Apply’ (zoom in)

• A useful feature in CFD Post is the ability to compare two different

sets of CFD data.
– Verify that the file NACA0012-mach-0.5-conv.dat.gz is in your working
directory.
– File > Load Results – Browse to your working directory.
– Under ‘Case options’ make sure ‘keep current cases loaded’ is checked.
– Open the File NACA0012-mach-0.5-conv.dat.gz.
• Click OK if an Information/Warning dialog box appears.
• We now have two data sets loaded and can do a case comparison.

• Make sure that two windows are open, and select the respective
cases in a different window.
• Lock the views so they are synchronised.

• Toggle on location ‘Symmetry 1’ in each case.

• Select ‘Contour 1’ and apply. We can compare the two pressure plots.

• Finally, we can plot the difference between the two.

• In the Outline view, double-click Case Comparison.
• The Case Comparison details view appears.
• Select Case Comparison Active and click Apply.
• A third viewport opens that displays the pressure difference between
the two cases.

Summary Workshop Supplement
• In this tutorial we have used FLUENT within a Workbench project to

compute the transonic, compressible flow over a naca0012 airfoil.
• We have imported a mesh that was generated in Gambit.
• We have used the density based solver with solution steering.
• We have compared the results to published experimental data and

seen good agreement.
• We have seen how FLUENT can be linked to CFD Post in a

Workbench project, and we have explored some of the features
within CFD Post.

References Workshop Supplement
• T.J. Coakley, “Numerical Simulation of Viscous Transonic Airfoil

Flows,” NASA Ames Research Center, AIAA-87-0416, 1987.
• C.D. Harris, “Two-Dimensional Aerodynamic Characteristics of the

NACA 0012 Airfoil in the Langley 8-foot Transonic Pressure Tunnel,”
NASA Ames Research Center, NASA TM 81927, 1981.


Introductory FLUENT Training: Workshop 2 Transonic Flow Over A NACA 0012 Airfoil

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Introductory FLUENT Training: Workshop 2 Transonic Flow Over A NACA 0012 Airfoil

Uploaded by

Copyright:

Available Formats

Workshop 2

Transonic Flow over a

ANSYS, Inc. Proprietary April 28, 2009

Goals Workshop Supplement

• The purpose of this tutorial is to introduce the user to good

• The flow to be considered is compressible and turbulent.

ANSYS, Inc. Proprietary April 28, 2009

Start a workbench project Workshop Supplement

• Launch Workbench and save the new project as naca0012 in your

ANSYS, Inc. Proprietary April 28, 2009

Import a mesh that was generated in Gambit Workshop Supplement

• Import the FLUENT mesh file (naca0012.msh).

ANSYS, Inc. Proprietary April 28, 2009

Mesh Workshop Supplement

• FLUENT will launch in a new window.

ANSYS, Inc. Proprietary April 28, 2009

Mesh Workshop Supplement

ANSYS, Inc. Proprietary April 28, 2009

Mesh Workshop Supplement

• Zoom in and examine the mesh.

ANSYS, Inc. Proprietary April 28, 2009

Solver Workshop Supplement

• Select the steady-state density-based solver:

• Turn on the energy equation.

• Select the turbulence model to be used:

ANSYS, Inc. Proprietary April 28, 2009

Materials Workshop Supplement

• The properties to be used for the material ‘air’ need to be set.

• Assign the material ‘air’ to the grid cells:

ANSYS, Inc. Proprietary April 28, 2009

Operating Conditions Workshop Supplement

• Set the Operating Pressure to Zero:

ANSYS, Inc. Proprietary April 28, 2009

Boundary Conditions Workshop Supplement

• Set the upstream boundary conditions:

Boundary Conditions Workshop Supplement

• Set a reasonable boundary condition for the

ANSYS, Inc. Proprietary April 28, 2009

Boundary Conditions Workshop Supplement

• Select the thermal tab.

ANSYS, Inc. Proprietary April 28, 2009

Boundary Conditions Workshop Supplement

ANSYS, Inc. Proprietary April 28, 2009

Reference Values Workshop Supplement

Note the reference values for density,

– Set the reference length (which is not

ANSYS, Inc. Proprietary April 28, 2009

Solution Methods Workshop Supplement

Solution Methods Workshop Supplement

• Change the gradient method to Green-Gauss Node Based.

ANSYS, Inc. Proprietary April 28, 2009

Solution Controls Workshop Supplement

As we will be using automatic

Keep the default under-relaxation

ANSYS, Inc. Proprietary April 28, 2009

Solution Monitors Workshop Supplement

• Set up residual monitors so the convergence can be monitored

ANSYS, Inc. Proprietary April 28, 2009

Solution Monitors Workshop Supplement

• Set up a monitor for the drag coefficient on the airfoil.

You can specify

ANSYS, Inc. Proprietary April 28, 2009

Solution Initialization Workshop Supplement