Introduction to stationary turbulence modeling (RAS) - Part 2

"Introduction to stationary turbulence modeling (RAS)" - Part 2

This material is published under the creative commons license CC BY-NC-SA (Attribution-NonCommercial-ShareAlike). If you plan to use it, please acknowledge it.

Download the blockMeshDict file here:

or

This video contains auxiliary material for students at the Johannes Kepler University in Linz, Austria. Students from other universities are welcome to use it for their learning purposes.

Thanks to Gavin Tabor from the University of Exeter for his input.

This series is based upon CFD tutorials created at the Vienna University of Technology in a cooparation with Bahram Haddadi, Christian Jordan and Michael Harasek and further improved at the Johannes Kepler University in Linz, Austria.

The used OpenFOAM version was precompiled by Andras Horvath from Rheologic GmbH, you can download it here:

[АзатГарифуллин-вх]

Thank you a lot! Your video lessons is really helpfull! Wish you good luck!

[ishaqquadri]

Hello. Thank you so much for the amazing lecture. I am in the learning phase of OpenFoam, I have a question, i do not know its relevant or not. So, I followed all the tutorial, but unfortunately, I can not see options of P, U, and T in the paraview.

[nikasarabi]

hello, thank you for the video,
is there any limit for k, when i calculate mine is very small, 0.08? can k be this small?

[汪洋-nw]

Hello Mr. Jozesf


When I used the default settings, the k blows up when iterations run about 20 times. but If I changed the relataxtionFactors the same to your settings on the video. It works. I am using OpenFOAM+V1712.


Thank you!

[garioncharles]

Hi, I am trying to model a T joint using this solver but my inlets are at two different velocities how would I calculate k and epsilon at my walls given the fact that my inlets will give me two different pairs of values ?

[nareshyethur]

First of all thanks for a great tutorial. I have a couple of questions.
1. is y+ should always necessarily be in the range of 5 to 30?
2. in the first part of this video u said simple foam can be also used for laminar flow. can u give a short description about which are the changes to be made . basically the turbulence model is  specified as laminar which other changes are neccessary.it would be very helpful. thx again.

[oscar]

Hi József,

I am adapting this RANS simulation you are showing for modelling the transitional flow over a low Reynolds number airfoil. I have a C-Mesh with significant refinement near the airfoil and along the line extending from the TE into wake. My question is, what boundary conditions should I use? For instance I have seen some people use 

inlet 
{
    freestream;
    freestreamValue uniform (0 0 0);
}

for all boundary fields other than the airfoil which is modelled with no slip. I'm not sure what boundary conditions are customary for these kinds of simulations? Any input would be much appreciated!

[parshvaparekh]

Also In this tutorial when I open the paraview why my cell arrays are not enabled? I am unabke to enable the cell arrays section why it is like that?

[sacramentofwilderness]

Do I correctly understand, that the turbulenceProperties and RAS properties have migrated to momentumTransport diсt?

[sebastianmurillo]

how could i put other backward step at this geometry for simulate an labyrinth seal?

[vinnyakhil]

Is there any 2d pipe geometry model that u hv created?...

[parshvaparekh]

Hello sir, I do not have file for RASproperties in constant folder? Where can I find that?

[ing.nicola]

I have tried kepsilon with another geometry. It works but my yplus il too elevated : 16000 . I have to make mesh more fine ? More course? Layers?

[abhishek]

Hello Mr. Jozesf,


My simulations is converging at time 134 s and not going to 2000s . Same setup, can you tell me what I am doing wrong.

[MrMangkokoo]

helo Dr. Nagy why did you assume turbulence intensity of 5%? did you make a rough estimation? if so what is the more apropiate way to calculate the intensity? thank you very much

[MLight]

Hi Mr.Najy,

If you didnt change the default values of k and epsillon, will this affect the results in a direct way? or it is just help the simulation to converge faster?

Thanks

[JoseLorenzoABarba]

Hello, I have some troubles when trying to run the simulation as directed in the tutorial. Specifically, the problem is when I change the discretization schemes to linearUpwind in the fvSchemes directory, after 22 iterations the program stops with the next error message:

smoothSolver: Solving for Ux, Initial residual = 0.590391, Final residual = 1.00201e+34, No Iterations 1000
smoothSolver: Solving for Uy, Initial residual = 0.606482, Final residual = 2.05892e+34, No Iterations 1000
#0 at ??:?
#1 Foam::sigFpe::sigHandler(int) at ??:?
#2 ? in
#3 double const&, Foam::UList<double> const&) at ??:?
#4 Foam::PCG::solve(Foam::Field<double>&, Foam::Field<double> const&, unsigned char) const at ??:?
#5 Foam::GAMGSolver::solveCoarsestLevel(Foam::Field<double>&, Foam::Field<double> const&) const at ??:?
#6 const&, Foam::Field<double>&, Foam::Field<double> const&, Foam::Field<double>&, Foam::Field<double>&, Foam::Field<double>&, Foam::Field<double>&, Foam::Field<double>&, >&, >&, unsigned char) const at ??:?
#7 Foam::GAMGSolver::solve(Foam::Field<double>&, Foam::Field<double> const&, unsigned char) const at ??:?
#8 const&) at ??:?
#9 const&) in
#10 in
#11 ? in
#12 __libc_start_main in
#13 ? in
Floating point exception (core dumped)


Does it has to do with the OpenFOAM version? (I am working with OpenFOAM 5)
I have to say that if I keep the discretization model unchanged for k and epsilon, the simulation runs properly.

I will be very thankful of any advice that somebody can offer.

[sumitzanje]

can you make a video on different wallfunctions and its impact?

[atzoulis]

I have 2 questions:
The first refes to turbulence in stationary modelling. If there is no time dependency, how is in this case turbulence defined? In the first tutorial of turbulence you implied only oscillations in time (vt).
The second question refers to the courant number in stationary simulations. If there is no time dependency, how is it calculated? If it applies to no physical time then it should be the timestep given in 'controlDict'. However there you could give a smaller endtime and a smaller deltaT in order for the quotient to be the same. Would that result in the same results?
Thank you

[fitnessmythwalakaka]

can I ask why you deleted the residual control for pressure and speed? and K?