CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   wall function (http://www.cfd-online.com/Forums/cfx/99023-wall-function.html)

Studentdrak March 24, 2012 01:37

wall function
 
"The basic idea behind the scalable wall-function approach is to limit the y* value used in the logarithmic formulation by a lower value of 11.06 is the intersection between the logarithmic and the linear near wall profile"

These are the statements from help files of cfx.

The lower limit of y* is restricted to 11.06,does this mean that the linear near wall profile is neglected?

"Do not use Standard Wall Functions unless required for backwards compatibility."

These are the statements from help files of cfx.

what is backward compatibility?

I am trying to simulate a heat transfer problem, how does scalable wall function affect the results?




my

ghorrocks March 25, 2012 07:48

Quote:

does this mean that the linear near wall profile is neglected?
No, just that the logarithmic approach is not used below y+=11.06, it transitions to the integrate to the wall approach.

Quote:

what is backward compatibility?
http://en.wikipedia.org/wiki/Backward_compatibility

Quote:

how does scalable wall function affect the results?
Scalable wall functions are your best chance of getting a wall function which works well regardless of having a fine or coarse mesh. Standard wall functions requires the user to mesh appropriately for the wall function approach you intend to use.

Studentdrak March 25, 2012 09:58

Thanks Glenn.

Would u pl. explain me how does doubling and quadrupling the number of nodes(for grid independent study) and scalable wall function work together?

ghorrocks March 25, 2012 18:59

Most CFD packages use standard wall functions, and these are only valid for y+>11 or so. So if you start off with a mesh with y+=30, then half the mesh element length you get y+=15 (so far so good) but the next refinement gives you y+=7.5. The standard wall function approach is not valid for this value of y+, so you will get a finer mesh solution but it will not be physically correct as you are applying inappropriate wall functions.

The scalable wall functions help here because as you refine to below y+=11 it transitions to integrating to the wall. This keeps the wall functions physically valid and means you can do mesh refinement more easily.

Studentdrak March 26, 2012 00:09

Thanks again Glenn.

As i understand, the Scalable wall functions are better to use for grid independent study also. is this correct?

ghorrocks March 26, 2012 00:12

Isn't that what I just said?

Studentdrak March 26, 2012 04:21

Thank u Glenn

Far March 26, 2012 05:53

Quote:

No, just that the logarithmic approach is not used below y+=11.06, it transitions to the integrate to the wall approach.
No. Below Y+ = 11.06 mesh points are ignored. If you check the mathematical formulation of scalable wall function you will not find any term for linear profile. Therefore scalable wall function are not built for resolving the viscous sub layer. For that you have to choose the automatic wall treatment which is only available in k-omega based models.

Transition to linear profile is property of hybrid wall functions (aka automatic wall treatment in CFX)

Scalable wall function are designed to avoid the problems

1. With mesh refinement when Y+ goes to 1. In this case standard wall function approach ceases to be valid.
2. At separation when velocity is zero and Y+ is again very low, in this case standard wall function ceases to be valid.

In other words scalable wall function is robust wall function as compared to standard wall function.

Standard wall functions are also valid up to Y+ = 11.06 and it was the default option in Fluent when Y+ is higher than 11.06. See the SA model section of Fluent User guide (I know this is CFX forum ;))

ghorrocks March 26, 2012 06:01

Thanks for the correction Far. I have just read the documentation to confirm this time :)

Scalable wall functions effectively limit the y+ to 11. When y+<11 it sets y+=11 and therefore assumes the first node is outside the log layer.

Automatic wall functions are what I was confused with, they transition from wall functions to integrating to the wall over the y+=11 threshold.

Far March 26, 2012 06:04

I have learned many things from you.

Studentdrak March 26, 2012 08:06

Thank u Glenn and Far,

Does this mean that for forced convection heat transfer problem K epsilon - scalable wall function are not correct to use?

ghorrocks March 26, 2012 19:00

It is rare that a turbulence model is "right" or "wrong". They are all different with different strengths and weaknesses.

But in general the SST turbulence model is the default choice. It allows easy extension for some advanced models like automatic wall functions, curvature correction, transition model and others. k-e I would only recommend if you are comparing to published data, it is largely superseded by SST in my opinion. RSM should be considered then anisotropic turbulence is significant, and LES style models (inc SAS, DES) for when you have large scale vortex shedding which you need to resolve.

Mina_Shahi June 28, 2012 10:45

Standard wall function and SAS model
 
Quote:

Originally Posted by ghorrocks (Post 351450)
Thanks for the correction Far. I have just read the documentation to confirm this time :)

Scalable wall functions effectively limit the y+ to 11. When y+<11 it sets y+=11 and therefore assumes the first node is outside the log layer.

Automatic wall functions are what I was confused with, they transition from wall functions to integrating to the wall over the y+=11 threshold.

Hi ghorrocks

You told When y+<11 it sets y+=11 and therefore assumes the first node is outside the log layer, you mean for y+< 11 it assumes u+=y+ (linear approach)?

in CFX when you select SAS-SST model it automaticly selects standard wall function , while from your discussion i learned that scalable wall function is a better choice. However i can't do anything for that because it will be selected automaticly. so in this case (standard wall function) i have to take care of the mesh and should refine it well close to the wall, am i right?

the y+ in my geometry (a combustor) is changing in the range of y+<40. so can you please tell me according to this wall function (standard) and this turbulence modell (sas-sst) which i used my mesh is good enough close to the wall?

Regards
Mina

Far June 28, 2012 12:24

Quote:

Originally Posted by Mina_Shahi (Post 368744)
Hi ghorrocks

You told When y+<11 it sets y+=11 and therefore assumes the first node is outside the log layer, you mean for y+< 11 it assumes u+=y+ (linear approach)?

in CFX when you select SAS-SST model it automaticly selects standard wall function , while from your discussion i learned that scalable wall function is a better choice. However i can't do anything for that because it will be selected automaticly. so in this case (standard wall function) i have to take care of the mesh and should refine it well close to the wall, am i right?

the y+ in my geometry (a combustor) is changing in the range of y+<40. so can you please tell me according to this wall function (standard) and this turbulence modell (sas-sst) which i used my mesh is good enough close to the wall?

Regards
Mina

It is the log law of the wall. There is option of choosing the standard wall function, but default is the scalable wall function for epsilon based models. For omega based models the default choice is automatic wall function.

Mina_Shahi June 28, 2012 16:21

Quote:

Originally Posted by Far (Post 368751)
It is the log law of the wall. There is option of choosing the standard wall function, but default is the scalable wall function for epsilon based models. For omega based models the default choice is automatic wall function.


Thanks Far!

1- in CFX the default for the sas model is standar model. But what is your idea about the range of y+ in my mesh? is it fine enough?
and how does the software take care of y+<11?

2- how can we know for a certain turbulence model how big y+ can be or how much it can vary?


Regards
Mina

ghorrocks June 28, 2012 19:29

I am not sure what wall boundary conditions SAS uses, you will have to look that up.

Not only do different turbulence models have different requirements but different flow regimes have different requirements. So the only way to be sure is to do a mesh refinement sensitivity study on your configuration.

Mina_Shahi June 29, 2012 07:03

Quote:

Originally Posted by ghorrocks (Post 368787)
I am not sure what wall boundary conditions SAS uses, you will have to look that up.

Not only do different turbulence models have different requirements but different flow regimes have different requirements. So the only way to be sure is to do a mesh refinement sensitivity study on your configuration.


Thanks

But can you explain what happens for y+<11 (in standard wall function)? I guess that turbulence equatuion will be integrated untill the wall in th eregion with y+<11, Is that true? or simply for the small y+ it assumes u+=y+ (linear approch and not logaritmic approach)? i am a bit confused with this layers :)

ghorrocks June 29, 2012 08:52

If you are using standard wall functions then the wall function equation will be used regardless of y+. Refer to the CFX documentation or a turbulence textbook like Wilcox for details on this. But the key point is that the wall function equation is only applicable from around y+>11. So if you apply standard wall functions with y+<11 then you are simply applying them in a regime where they are not accurate.

Far June 29, 2012 09:49

Actually there is change in definition of u+. so instead of u+, u* is used for the scalable wall function. see documentation.

ghorrocks June 29, 2012 09:54

I think the question was what happens for standard wall functions.


All times are GMT -4. The time now is 19:54.