Difference in near wall treatment

~Akshay350~ · March 8, 2017, 10:27

Hi cfd members,
I wanna know the basic difference in standard wall functions, enhanced wall functions and scaleable wall functions.
Tnx in advance!

LuckyTran · March 9, 2017, 01:39

Seriously this topic has been discussed to death already on this forum. Here's my short, spur of the moment response. If you google it enough you might find some contradictions, but I hope it gets you started.

First you need to understand what any wall function does and then what the standard wall function does. It's part of getting boundary conditions when your cell centers are very far from the boundary, and you can't solve for all the physics in between. So you start to assume that the flow does certain things as you approach the wall. And you use this assumed behavior to get your boundary conditions.

The scalable wall function is like the standard wall function, except that it applies a limiter to the calculation of y+ (actually y*) that prevents the wall function from going into the low-y+ regime. The scalable wall functions allows you to use the log behavior anyway in the low-y+ region where you normally wouldn't apply them. There are pros and cons of this. The pros are that your solution doesn't blow up as you make the mesh arbitrarily small and keep trying to use wall functions. The cons is that you are still using wall functions in an area where you would rather brute force the turbulence model (like in the k-omega model).

The enhanced wall treatment is the monster. The difference is that next to walls, instead of solving the full k-epsilon equations, you would rather solve the one equation model by Wolfshtein. So Fluent does both and then blends the two together. The blending isn't really important because it's just to make the model user friendly and make the numerics happy. The reason you want to solve Wolfshtein's model is because the epsilon equation doesn't work near walls (because the epsilon equation doesn't have two length scales).

March 8, 2017, 10:27	Difference in near wall treatment	#1
~Akshay350~ New Member Join Date: Mar 2017 Posts: 6 Rep Power: 9	Hi cfd members, I wanna know the basic difference in standard wall functions, enhanced wall functions and scaleable wall functions. Tnx in advance!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
decomposePar problem: Cell 0contains face labels out of range	vaina74	OpenFOAM Pre-Processing	37	July 20, 2020 05:38
Natural convection in a closed domain STILL NEEDING help!	Yr0gErG	FLUENT	4	December 2, 2019 00:04
Near wall treatment in k-omega SST	Arnoldinho	OpenFOAM Running, Solving & CFD	38	March 8, 2017 13:48
wall roughness in enhanced wall treatment	lobstar	FLUENT	7	June 29, 2012 04:43
RSM with enhanced wall treatment	zwdi	FLUENT	1	April 24, 2005 20:36

March 9, 2017, 01:39		#2
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,673 Rep Power: 65	Seriously this topic has been discussed to death already on this forum. Here's my short, spur of the moment response. If you google it enough you might find some contradictions, but I hope it gets you started. First you need to understand what any wall function does and then what the standard wall function does. It's part of getting boundary conditions when your cell centers are very far from the boundary, and you can't solve for all the physics in between. So you start to assume that the flow does certain things as you approach the wall. And you use this assumed behavior to get your boundary conditions. The scalable wall function is like the standard wall function, except that it applies a limiter to the calculation of y+ (actually y*) that prevents the wall function from going into the low-y+ regime. The scalable wall functions allows you to use the log behavior anyway in the low-y+ region where you normally wouldn't apply them. There are pros and cons of this. The pros are that your solution doesn't blow up as you make the mesh arbitrarily small and keep trying to use wall functions. The cons is that you are still using wall functions in an area where you would rather brute force the turbulence model (like in the k-omega model). The enhanced wall treatment is the monster. The difference is that next to walls, instead of solving the full k-epsilon equations, you would rather solve the one equation model by Wolfshtein. So Fluent does both and then blends the two together. The blending isn't really important because it's just to make the model user friendly and make the numerics happy. The reason you want to solve Wolfshtein's model is because the epsilon equation doesn't work near walls (because the epsilon equation doesn't have two length scales).