Adding a turbulence model into my own CFD code

quarkz · November 21, 2013, 02:50

Hi,

I have a written a 3D FVM Immersed boundary code. It works fine but I am thinking of running higher Re simulation, which now requires the addition of a turbulence model.

It seems that the Menter SST is a good option. However, how difficult is it to add the turbulence model to my code?

Also, is there any good references on how to start, since my turbulence knowledge is very basic?

Thanks!

samurai_01 · November 21, 2013, 03:22

Hi,

You can start reading Turbulence by Davidson for understanding principles of turbulence, and to implement turbulence numerically, you can read Pope or Wilcox, both are good books for numerical modelling.

There are basically two types of turbulence modelling that you can see, one is called as Reynolds Averaged Navier Stoke (RANS) and Large eddy simulation.
Turbulence is a mixture of knowledge and magic so I would suggest you to go through the above books and get a feel of 2 eqn model (k-e) model theoretically, before implementing in Code.

Can you please explain me a little about immersed boundary code that you have written, because I have also been developing FVM solver for general purpose flow problems and I think this might help.

FMDenaro · November 21, 2013, 04:34

Just a suggestion, do not start thinking of the turbulence model.... Try to focus first on the type of problem you want to solve, it will address you about the formulation to use (DNS/LES/RANS/URANS) and then you can think about the turbulence model

quarkz · November 23, 2013, 23:53

Thank you samurai_01 and FMDenaro for your suggestions.

I will look up the references you gave.

I used the IBM by Yang and Balaras JCP 2006. There's also some other implementation of IBM, dep on your needs.

For my problem, which is low Re, ~10k to 30k, I read that the menter SST is a good option. Anyway, I think the references should give me a better clue.

Thanks!

November 21, 2013, 02:50	Adding a turbulence model into my own CFD code	#1
quarkz Senior Member TWB Join Date: Mar 2009 Posts: 414 Rep Power: 19	Hi, I have a written a 3D FVM Immersed boundary code. It works fine but I am thinking of running higher Re simulation, which now requires the addition of a turbulence model. It seems that the Menter SST is a good option. However, how difficult is it to add the turbulence model to my code? Also, is there any good references on how to start, since my turbulence knowledge is very basic? Thanks!

November 21, 2013, 04:34		#3
FMDenaro Senior Member Filippo Maria Denaro Join Date: Jul 2010 Posts: 6,896 Rep Power: 73	Just a suggestion, do not start thinking of the turbulence model.... Try to focus first on the type of problem you want to solve, it will address you about the formulation to use (DNS/LES/RANS/URANS) and then you can think about the turbulence model samurai_01 likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
gamma-ReTheta turbulence model for predicting transitional flows	FelixL	OpenFOAM Programming & Development	123	August 30, 2022 12:50
An error has occurred in cfx5solve:	volo87	CFX	5	June 14, 2013 18:44
What model of turbulence choose to study an external aerodynamics case	raffale	OpenFOAM	0	August 23, 2012 06:45
Understanding k-omega SST model source code	tmhonka	OpenFOAM Programming & Development	1	September 8, 2009 08:33

November 21, 2013, 03:22		#2
samurai_01 Member Join Date: Oct 2012 Posts: 46 Rep Power: 14	Hi, You can start reading Turbulence by Davidson for understanding principles of turbulence, and to implement turbulence numerically, you can read Pope or Wilcox, both are good books for numerical modelling. There are basically two types of turbulence modelling that you can see, one is called as Reynolds Averaged Navier Stoke (RANS) and Large eddy simulation. Turbulence is a mixture of knowledge and magic so I would suggest you to go through the above books and get a feel of 2 eqn model (k-e) model theoretically, before implementing in Code. Can you please explain me a little about immersed boundary code that you have written, because I have also been developing FVM solver for general purpose flow problems and I think this might help.

November 23, 2013, 23:53		#4
quarkz Senior Member TWB Join Date: Mar 2009 Posts: 414 Rep Power: 19	Thank you samurai_01 and FMDenaro for your suggestions. I will look up the references you gave. I used the IBM by Yang and Balaras JCP 2006. There's also some other implementation of IBM, dep on your needs. For my problem, which is low Re, ~10k to 30k, I read that the menter SST is a good option. Anyway, I think the references should give me a better clue. Thanks!