How to decide mesh resolution for Large Eddy Simulation (LES) and for SAS-SST model

juzer_700 · May 7, 2015, 10:03

I am performing Scale-resolving Simulation of a Francis turbine. Therefore, need Inputs in how fine my mesh should be for a resolved flow.

ghorrocks · May 7, 2015, 19:15

First of all - post your question here on the forum in future. If you want to have the discussion on linkedin then don't post here as well.

Have you read the documentation on meshing and advanced turbulence models? It gives a lot of information on this topic.

juzer_700 · May 7, 2015, 19:56

Quote:

Originally Posted by ghorrocks

First of all - post your question here on the forum in future. If you want to have the discussion on linkedin then don't post here as well.

Have you read the documentation on meshing and advanced turbulence models? It gives a lot of information on this topic.

I am just trying to gain different viewpoint on this topic from different people. Therefore I used other platforms also to post my queries.

Coming back to the topic, yes I have read about index of quality resolution required for LES. But still unable to get a work around regarding y+ values for LES. Are y+ values greater than say 30, still able to resolve my boundaries? My francis turbine goemetry is too large and therefore lower y+ values would certainly increase computational mesh.

And which documentation are you mentioning? Am I missing something?

ghorrocks · May 8, 2015, 04:02

I have no problem with you posting on other forums. But the discussion on each forum should be self contained.

Wall conditions for LES are problematic. There are many approaches.
* If you are lucky enough that walls do not affect the flow much then you are probably OK with coarse wall resolution (eg turbulent jets)
* There is a lot of literature on wall functions and modellign for LES. It is a very complex area.
* You probably need to use a mesh of the order of y+=1 for most applications.
* The variations on LES available in CFX such as DES and SAS have some very useful ways of assisting with these sort of problems. I recommend you investigate some of these approaches.

The Modelling guide contains lots of useful information on meshing. Also if you are considering LES you really need to understand the special requirements LES imposes - look for a CFD textbook or the general literature for this.

May 7, 2015, 10:03	How to decide mesh resolution for Large Eddy Simulation (LES) and for SAS-SST model	#1
juzer_700 Member Join Date: Apr 2012 Location: Mainz, Germany Posts: 60 Rep Power: 14	I am performing Scale-resolving Simulation of a Francis turbine. Therefore, need Inputs in how fine my mesh should be for a resolved flow.

May 7, 2015, 19:15		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,871 Rep Power: 144	First of all - post your question here on the forum in future. If you want to have the discussion on linkedin then don't post here as well. Have you read the documentation on meshing and advanced turbulence models? It gives a lot of information on this topic.

May 8, 2015, 04:02		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,871 Rep Power: 144	I have no problem with you posting on other forums. But the discussion on each forum should be self contained. Wall conditions for LES are problematic. There are many approaches. * If you are lucky enough that walls do not affect the flow much then you are probably OK with coarse wall resolution (eg turbulent jets) * There is a lot of literature on wall functions and modellign for LES. It is a very complex area. * You probably need to use a mesh of the order of y+=1 for most applications. * The variations on LES available in CFX such as DES and SAS have some very useful ways of assisting with these sort of problems. I recommend you investigate some of these approaches. The Modelling guide contains lots of useful information on meshing. Also if you are considering LES you really need to understand the special requirements LES imposes - look for a CFD textbook or the general literature for this.