[Zigainer]

Hi,

I am facing new problem with a new meshing tool.
When I run my simulation I get a wide range of y+ (0.5 - 90). My problem now is which wall function should I use. I am using realizable k-epsilon....

I know, that for standard wall function y+ should be in the range of 30.... 300. But can enhanced wall function handle my problem?

I can handle the y+ value at the area where I am most interested, but not for the whole mesh.

Hope you guys have any advice.

Thanks in advance

[yonchong]

Are you doing a heat transfer calculation?

Why not run both the standard and enhanced wall treatment and compare the result. At the same time try the k-omega (which has an enhanced wall treatment built in) and compare the results as well.

[Zigainer]

Yes I am running heat transfer.
Experience with realizable k-epsilon are quite good so far.

I just wanted to check I the results are bad anyway because I can not meet y+>30 for the standard wall function ... so it might not be reasonable to run the simulation at all!

What about k-w? Is there no limit for the y+ range and wall funtions? I don't think so ....

[yonchong]

My experince is that k-w seems to handle wider y+ better than k-e for heat transfer. I would still run it (considering how long it take to generate error free prism mesh) but would compare all three results as rerunning with different turbulence models that much.

[Jabba]

Originally Posted by yonchong

Our experince is that k-w seems to handle wider y+ better than k-e for heat transfer. I would still run it (considering how long it take to generate error free prism mesh) but would compare all three results as rerunning with different turbulence models that much.

Fluent's guide recommends to use the realizable k-epsilon turbulence model with the enhanced wall functions for meshes with y+ close to 1. The application of the standard wall function will not provide good results, I guess.

For the k-omega model, the default wall function is the same as the k-epsilon enhanced wall function.

I believe that would be interesting for you to run these both cases: k-epsilon with enchanced wall function and then using k-omega.

[Zigainer]

Originally Posted by Jabba

Fluent's guide recommends to use the realizable k-epsilon turbulence model with the enhanced wall functions for meshes with y+ close to 1. The application of the standard wall function will not provide good results, I guess.

Probelm is, that I can not get y+ values close to 1 at my traget surface ...

[Far]

use scalable wall function. It will discard the yplus values below 11.06.

[Zigainer]

Originally Posted by Far

use scalable wall function. It will discard the yplus values below 11.06.

And how to do this? User-defined wall function?

[Far]

through text user interface. It is only available for Epsilon based models.

However, in K-W model (at-least true for Fluent 14) it automatically changes to integration to wall approach to standard wall function according to following scheme:

Below 6 : integration to wall.

Between 6 and 30 : mix of both with blending factor.

above 30: standard wall function.

However for heat transfer calculation you should consider the appropriate range of thermal Y+ to accurately heat transfer.

[Zigainer]

Originally Posted by Far

through text user interface. It is only available for Epsilon based models.

I can't find it ... for k-e realizable/standard/rng

define/models/viscous/near-wall-treatment ???

or

define/models/viscous/turbulence-expert/low-re-ke ???

Thanks

[Far]

Quote from Fluent Guide

Scalable wall functions can be enabled only when the standard wall functions are used and only via the following text command:

define/models/viscous/near-wall-treatment/scalable-wall-functions?

https://www.sharcnet.ca/Software/Flu.../th/node99.htm


https://www.sharcnet.ca/Software/Flu...list/fltui.pdf

[Zigainer]

Thanks a lot!