Learning Forum

[JanEllmenreich]

Hello there!

I am performing a "simple" simulation with ANSYS CFX to consider the different heights of roughness in a pipe flow.
I´ve read the user guide regarding the mathematical formulations,(https://www.sharcnet.ca/Software/Ansys/16.2.3/en-us/help/cfx_thry/cfxTurbModeMath.html#i1303614) but I still have some questions about:

• The SST-model uses a automatic rough wall treatment, with no scalable approach. Does my y+ of my first adjacent cell has to be higher than the h+ (dimensionless sand grain roughness)? Or is there a delimiter like in the k-e formulation?


• It is stated that for the k-e, when roughness is treated, you use a correction term for the log-law.
  Hence the scalable Wall function uses this delimiter=
  If my adjacent cell has a lower y+ than the delimiter, will the cell not be considered by the solver?

 

[AhmedCfx]

Dear JanEllmenereich

How can I get variable roughness 13-30 mm in fluent?

[JanEllmenreich]

Hi Ahmed!

I think you mean micrometer right? mm would be a hell of a roughness.

Anyways you want to have a variable roughness over the length of a body I guess?
I have never thought about that. Honestly I don't know if you could implement this behaviour in Ansys.

First I would perform a simulation with both limits, to look where the bounds the K+_s are, in order to charachterize if i am in a fully smooth, transitional or rough enivornment (https://www.afs.enea.it/project/neptunius/docs/fluent/html/ug/node250.htm). 

If both limits are in one region, which will likely be for your roughness, then i would just take the mean value and not focus to much on implementing a variable roughness distribution.

 

[Rob]

Click on "Help" and follow the links. There will be an explanation on calculating roughness etc.