CFD Online Discussion Forums

CFD Online Discussion Forums (
-   OpenFOAM (
-   -   Turbulence modelling SIG (

grtabor October 16, 2008 10:14

Dear All, I've thought for
Dear All,

I've thought for a while now that there might be a case for establishing a Special Interest Group on turbulence modeling, similar in format to the Turbomachinery SIG. I've had a few discussions about this with Fabian Brännström - is there a general interest in forming such a group?

Also; what should its remit be? I was thinking of the group covering LES, RANS/RS, DES, wall and inlet modeling, validation and application. Any other areas of relevance?

Finally, if there is enough interest - how do we go about setting up the SIG? Advice and recommendations (for instance from the Turbomachinery SIG) would be invaluable.


hani October 16, 2008 12:40

Hi Gavin, I think that ther
Hi Gavin,

I think that there is a need for a SIG on turbulence modelling. One important contribution would be to develop some best-practice guidelines for turbulence modelling with OpenFOAM. Of course, documentation, and development and implementation of new models are also very important. There should be a sufficent number of people out there already for you to be able to set up quite a substantial amount of work in a short time. I have seen people developing new turbulence models and new wall functions, but I'm not sure if these implementations have been made available. At least I don't think that they have been made available in an organized way so that it is easy for others to find them and use them.

I am too much involved in other activities around OpenFOAM, like SIG Turbomachinery, so I will not be able to take a leading role in this new formation. However, whenever you need my experience from the work in the Turbo group, I'm here for you.

The Turbo group mainly has the following playgrounds: The OpenFOAM Wiki, OpenFOAM-extend, and an e-mail list. If you go to our page at the OpenFOAM Wiki you should understand the structure of our work. you are free to copy our structure, or to develop your own. You find our Wiki at:

Good luck!

mbeaudoin October 16, 2008 15:09

Hi Gavin, As Håkan mentionn
Hi Gavin,

As Håkan mentionned in his message, the OpenFOAM Wiki, OpenFOAM-extend, and an e-mail list are 3 very usefull playgrounds for any SIG.

Just drop me a line if you are interested in using a svn space and a dedicated mailing list on OpenFOAM-extend for your SIG, just like we did for the Turbo SIG.

Of course, a dedicated discussion thread on the OpenFOAM Message Board is also a usefull mean of communication.



cedric_duprat October 17, 2008 09:24

Hi Gavin, According to me,
Hi Gavin,

According to me, that's a good idea to buid a turbulence SIG. However, I think turbulence SIG (included RANS method) is too large, in fact, "every one" is doing tubulent flow. Maybe we should focus on some specific points.
I'm more interrested in a LES, DNS and related techniques SIG. By related technique I mean wall model, inflow, sub-grid models, RANS/LES, ect ... which is enought large.
Anyway, that's just my point of view.

Second point, what will be the goal of such SIG, according to me :

1- generate documentation on what is already existing in OpenFOAM (techincal documents, tutorials)
2- "build&play" benchmark like for the turbomachinery SIG for classical LES (simple geometry) ?
3- new implementation ? I think this is the less easy part. I'm working for 1 years and half with OpenFOAM. I'm pH'd student working for a firm and ... I'm not sure I can give my implementation before the end on my pH'd ... (and I think I'm not the only one in such case)

Here was my points of view. So I'll be happy to be include (and particiate) in such group.

any comments ?


braennstroem October 18, 2008 11:12

Hi to all, around Stuttgart
Hi to all,

around Stuttgart we had the other day a first User Group meeting and did collect some interests:

Turbulencemodeling with hybrid RANS/LES and wall treatment was one of those. I think we should combine these activities with the TSIG.

According to the validation cases, I think, one should take a first look at some of those ercoftac cases and provide all information to run the case at home (there is good example ). This includes relevant papers and result-files of experiments and/or DNS data. In addtion too these 'objective' results one should include results of other commercial CFD tools and calculations with simple RANS models as well.

The validation of more complex geoemtries should be done in the corresponding SIGs and the turbulence group could link to those calculations. E.g. if there are some hybrid implementations in the turbomachinery group one should integrate the work. So I don't think, that the group is too large; it would be more a kind of collection for the more complex problems.

About the third point of Cedric, you are right, that this is not as easy, but on the other hand as long as your are not inventing complete new models, but just an implementation of existing and published ones, there should be no problem. And you could benefit from the knowledge of others!? In addition all of us (or better some of us) are doing benchmark test on simple geometries (even when you do your Phd in industry like me :-) ), which we could make publish without any problems.


braennstroem October 23, 2008 08:21

Hi, if you don't mind, I wi

if you don't mind, I will use a bit of the turbomachinery SIG wiki structure to
create some wiki pages to collect some interest.


hani October 24, 2008 03:28

Hi Fabian, I think that it
Hi Fabian,

I think that it is a good idea if you use what has already been done in the Turbo Wiki.

You can also add some svn space next to the Turbomachinery svn space at SourceForge: OSIG/
This link is for code related to OpenFOAM-1.4.1. We will soon move our code into 1.5.x and put it in: 1.5/OSIG/

If you want to get write-access to the svn (which you need to add stuff there) you can contact me, Martin Beaudoin, Bernhard Gchaider or Hrvoje Jasak.


braennstroem November 9, 2008 09:18

Hi, I created the first pag

I created the first pages for the Turbulence SIG ... so feel free to 'enhance' those :-)


braennstroem March 10, 2009 15:59

Hi, I would like to announc

I would like to announce, that Martin Beaudoin helped out with setting up a dedicated turbulence sig mailing list. You can join the list by following the link:


All times are GMT -4. The time now is 05:00.