Tutorial Six

Turbulence – Steady State

4th edition, Jan. 2018

This offering is not approved or endorsed by ESI® Group, ESI-OpenCFD® or the OpenFOAM®
Foundation, the producer of the OpenFOAM® software and owner of the OpenFOAM® trademark.
 OpenFOAM® Basic Training
Tutorial Six

Editorial board: Contributors:

 Bahram Haddadi  Bahram Haddadi
 Christian Jordan  Clemens Gößnitzer
 Michael Harasek  Jozsef Nagy
 Vikram Natarajan
Compatibility:  Sylvia Zibuschka
 OpenFOAM® 5.0  Yitong Chen
 OpenFOAM® v1712

Cover picture from:

 Bahram Haddadi

Attribution-NonCommercial-ShareAlike 3.0 Unported (CC BY-NC-SA 3.0)

This is a human-readable summary of the Legal Code (the full license).
Disclaimer
You are free:
- to Share — to copy, distribute and transmit the work
- to Remix — to adapt the work
Under the following conditions:
- Attribution — You must attribute the work in the manner specified by the author or
licensor (but not in any way that suggests that they endorse you or your use of the
work).
- Noncommercial — You may not use this work for commercial purposes.
- Share Alike — If you alter, transform, or build upon this work, you may distribute the
resulting work only under the same or similar license to this one.
With the understanding that:
- Waiver — Any of the above conditions can be waived if you get permission from the
copyright holder.
- Public Domain — Where the work or any of its elements is in the public domain under
applicable law, that status is in no way affected by the license.
- Other Rights — In no way are any of the following rights affected by the license:
- Your fair dealing or fair use rights, or other applicable copyright exceptions and
limitations;
- The author's moral rights;
- Rights other persons may have either in the work itself or in how the work is used,
such as publicity or privacy rights.
- Notice — For any reuse or distribution, you must make clear to others the license
terms of this work. The best way to do this is with a link to this web page.

For more tutorials visit: www.cfd.at

 OpenFOAM® Basic Training
Tutorial Six

Background
1. Why turbulence modeling?

Many engineering applications are turbulent. Turbulence is a highly transient

phenomenon, characterized by a wide range of eddy sizes. One can solve these eddies
numerically and obtain a full profile of the turbulent flow field. However, this is not
possible as it requires a huge amount of computational effort. Hence we require a
turbulence model.

An important feature in turbulence modeling is averaging, which simplifies the

solution of the governing equations of turbulence. As calculating resources are
limited, it is usually not possible to model the phenomena with desired grid and time
resolution, so to represent scales of the flow that are not resolved by the grid models
need to be applied.

There are different types of turbulence models:

 RANS-based models:
 Linear eddy-viscosity models
 Algebraic models
 One and two equation models
 Non-linear eddy viscosity models and algebraic stress models
 Reynolds stress transport models
 Large eddy simulations
 Detached eddy simulations and other hybrid models

In this tutorial, RANS-based model is explained in detail. In the next tutorial, large
eddy simulations (LES) and Smagorinsky-Lilly model will be covered.

2. RANS-based models

The governing equations for a Newtonian fluid are:

o Conservation of mass

( ̃)

o Conservation of momentum (Navier-Stokes equation)

( ̃) ̃
( ̃ ̃) ( ̃) ̃

o Conservation of passive scalars (given a scalar ̃ )

( ̃)
( ̃ ̃) ( ̃) ̃
 OpenFOAM® Basic Training
Tutorial Six

Note: suffix notation is used in the conservation of momentum equation for

simplicity, with corresponding to the x-direction, the y-direction and
the z-direction.

One of the solutions to the problem is to reduce the number of scales (from infinity to
1 or 2) by using the Reynolds decomposition. Any property (whether a vector or a
scalar) can be written as the sum of an average and a fluctuation, i.e. ̃ = Φ + φ where
the capital letter denotes the average and the lower case letter denotes the fluctuation
of the property. Using the Reynolds decomposition in the Navier-Stokes equations we
obtain RANS or Reynolds Averaged Navier Stokes Equations.

o Average conservation of mass

( )

o Average conservation of momentum

( )
( ) ( )

( ̅̅̅̅̅) ( ̅̅̅̅̅) ( ̅̅̅̅̅)

( )

o Average conservation of passive scalars (given a scalar ̃ )

( )
( ) ( )

( ̅̅̅) ( ̅̅̅) ( ̅̅̅̅)

( )

Note: a special property of the Reynolds decomposition is that the average of the
fluctuating component is identically zero, a fact that is used in the derivation of the
above equations.

However, by using the Reynolds decomposition, there are new unknowns that were
introduced such as the turbulent stresses ( ̅̅̅̅, ̅̅̅̅, ̅̅̅̅, ̅̅̅̅, ̅̅̅, ̅̅̅̅, ̅̅̅̅,
̅̅̅̅, ̅̅̅̅̅) and turbulent fluxes ( ̅̅̅, ̅̅̅, ̅̅̅̅) and therefore, the RANS equations
describe an open set of equations (where the over bar denotes an average). The need
for additional equations to model the new unknowns is called Turbulence Modeling.

We now have 9 additional unknowns (6 Reynolds stresses and 3 turbulent fluxes). In

total, for the simplest turbulent flow (including the transport of a scalar passive scalar,
e.g. temperature when heat transfer is involved) there are 14 unknowns (include u, v,
w, p, T)!

One possible approach to model the additional unknowns is to use the PDEs for the
turbulent stresses and fluxes as a guide to modeling. The turbulent models are as
follows, in order of increasing complexity:

 Algebraic (zero equation) models: mixing length (first order model)

 OpenFOAM® Basic Training
Tutorial Six

 One equation models: k‐model, μt‐model (first order model)

 Two equation models: k‐ε, k‐kl, k‐ω, low Re k‐ε (first order model)

 Algebraic stress models: ASM (second order model)

 Reynolds stress models: RSM (second order model)

 Zero‐Equation Models

In OpenFOAM®, there are two simulation types for turbulence flow, RAS and LES.
As the name suggest, the RAS simulation is based on the RANS-based models
covered above and will be the sole focus of this tutorial. In the next tutorial, we will
move on to LES modeling and compare the results generated from these two
modeling types.
 OpenFOAM® Basic Training
Tutorial Six

simpleFoam – pitzDaily

Simulation
Use simpleFoam solver, run a steady state simulation with following turbulence
models:

 kEpsilon (RAS)

 kOmega (RAS)

 LRR (RAS)

Objectives
 Understanding turbulence modeling

 Understanding steady state simulation

Data processing
Show the results of U and the turbulent viscosity in two separate contour plots.
 OpenFOAM® Basic Training
Tutorial Six

1. Pre-processing

1.1. Copy tutorial

$FOAM_TUTORIALS/incompressible/simpleFoam/pitzDaily

1.2. 0 directory
When a turbulent model is chosen, the value of its constants and its boundary values
should be set in the appropriate files. For example in kEpsilon model the k and
epsilon files should be edited. See below for the epsilon file:
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* * * * * *//

dimensions [0 2 -3 0 0 0 0];

internalField uniform 14.855;

boundaryField
{
inlet
{
type fixedValue;
value uniform 14.855;
}
outlet
{
type zeroGradient;
}
upperWall
{
type epsilonWallFunction;
value uniform 14.855;
}
lowerWall
{
type epsilonWallFunction;
value uniform 14.855;
}
frontAndBack
{
type empty;
}
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* * * * * *//

Note: Here is a list of files which should be available at 0 directory and need to be
modified for each turbulence model:

 laminar: no file

 kEpsilon (RAS): k and epsilon

 kOmega (RAS): k and omega

 LRR (RAS): k, epsilon and R

 Smagorinsky (LES): nuSgs

 kEqn (LES): k and nuSgs – This model is called „oneEqEddy‟ in V2.3.0

 OpenFOAM® Basic Training
Tutorial Six

 SpalartAllmaras (LES): nuSgs and nuTilda

Some files are available, e.g. epsilon, k and nuTilda, some files should be created by
the user, e.g. R, nuSgs. Templates for these files can be also found in the examples of
older versions of OpenFOAM®, e.g. 1.7.1.

Note: A missing R file can be created by OpenFOAM®. Open the

turbulenceProperties file in the constant directory, set the simulationType to RAS,
and RASModel to kEpsilon. Run the command „simpleFoam –postProcess –fun R‟
from terminal, it will create the turbulenceProperties:R file in the 0 directory. Then
simply rename the file to „R‟.

1.3. constant directory

In the turbulenceProperties file, the simulationType can be set as either RAS, LES
or laminar. Then the corresponding sub-dictionary of the chosen simulation type
needs to be defined. In this case, the sub-dictionary for RAS contains information
about the chosen RAS model (kEpsilon), and the status of turbulence and
printCoeffs are turned to on.

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* * * * * *//

simulationType RAS;

RAS
{
RASModel kEpsilon;
turbulence on;
printCoeffs on;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* * * * * *//

Note: For the laminar model, set turbulence and printCoeffs to off.

1.4. system directory

Note: Since it is a steady state simulation, endTime in controlDict shows the number
of iterations instead of time and deltaT should be 1, because it is the amount of
increase in the iteration number.

For the LRR model, discretization model for the new variable R needs to be specified.
It is done through the fvSchemes file,

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* * * * * *//

ddtSchemes
{
default steadyState;
}

gradSchemes
{
default Gauss linear;
}

divSchemes
{
 OpenFOAM® Basic Training
Tutorial Six

default none;
div(phi,U) bounded Gauss linearUpwind gradf(U);
div(phi,k) bounded Gauss limitedLinear 1;
div(phi,epsilon) bounded Gauss limitedLinear 1;
div(phi,omega) bounded Gauss limitedLinear 1;
div(phi,v2) bounded Gauss limitedLinear 1;
div(phi,R) bounded Gauss limitedLinear 1;
div(R) Gauss linear;
div((nuEff*dev2(T(grad(U))))) Gauss linear;
div(nonlinearStress) Gauss linear;
}

laplacianSchemes
{
default Gauss linear corrected;
}

interpolationSchemes
{
default linear
}

snGradSchemes
{
default corrected
}

wallDist
{
Method meshWave;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* * * * * *//

Furthermore, fvSolution needs to be changed due to the new R parameter. The solver
type for R is defined, in this case the solver used will be the same as the one for other
variables (U, k, epsilon, omega).

2. Running simulation

>blockMesh
>simpleFoam

Note: When the solution converges, “SIMPLE solution converged in …

iterations” message will be displayed in the Shell window. If nothing happens and
you do not see a message after a while (this is not the case in here, it converges after
a short time), then you should check the residuals which are displayed in the Shell
window manually (you should check initial residual values, it shows the
difference between this iteration and the last one), if all of the Initial residual
(see below) values are close to amounts you have set in the fvSolution then you can
stop simulation (ctrl+c).
Time = 298

smoothSolver: Solving for Ux, Initial residual = 0.00013831, Final residual =

9.28001e-06, No Iterations 6
smoothSolver: Solving for Uy, Initial residual = 0.000977894, Final residual =
6.73868e-05, No Iterations 6
GAMG: Solving for p, Initial residual = 0.00192871, Final residual =
0.000174838, No Iterations 7
time step continuity errors : sum local = 0.000840075, global = 6.13868e-05,
cumulative = -0.193739
 OpenFOAM® Basic Training
Tutorial Six

smoothSolver: Solving for epsilon, Initial residual = 0.000175322, Final

residual = 1.138e-05, No Iterations 2
smoothSolver: Solving for k, Initial residual = 0.000404928, Final residual =
2.99083e-05, No Iterations 2
ExecutionTime = 56.7 s ClockTime = 57 s

SIMPLE solution converged in 298 iterations

3. Post-processing

The simulation results are as follows (all simulations scaled to the same range):
RAS
Velocity magnitude Turbulent viscosity
model

kEpsilon

kOmega

LRR

Velocity magnitude and turbulent viscosity for different RAS models