Learning Forum

[melalisa96]

Hi

I'm not hitting the expected range of y+ values for my simulation. Is there a way in Fluent that will allow me to input my desired y+ values to obtain the first layer cell size?

[Rob]

No, what is happening is the flow (and probably first cell height) is not uniform and so y+ will drift. You could take that result and then try to remesh based on the y+ to find the cell height at each location: that's not a good idea.   I'd look at using adaption to refine the mesh where it's needed. 

[melalisa96]

Hi rwoolhou

How do I go about using adaption to refine the mesh where it's required?

[Keyur Kanade]

Can you please insert images of your mesh? Also let us know the boundary conditions and models you are using.

What y+ value you are targeting? Are you using and prism layer? 

 

[melalisa96]

Hi Kkanade

I am simulating a rotating propeller and am using K-epsilon standard wall function. The range of y+ values which I'm targeting are between 30 to 300. 

[Keyur Kanade]

Mesh looks ok to me. What y+ values you are getting after solving?

Can you please give more details about your set up?

[melalisa96]

Hi Kkanade

My y+ values are 8-175 for the default adaptive meshing. I tried to refine the mesh by diving the size of each mesh by 2 and got between 10-212. However, in doing so, my values drifted further from the wind tunnel experiment result which I found online as a means to validate my methodology. The link can be found here: https://m-selig.ae.illinois.edu/props/volume-1/propDB-volume-1.html 

Basically, I am simulating an apc 8x3.8 slow flyer propeller. As you can see from the images above, I have enclosed the propeller in a rotating region followed by the body of influence and them the control volume domain. I am using k-epsilon standard wall function which requires me to target a range of 30-300 for the y+ values. It appears I'm not reliably catching the flow and I'm guessing it's due to the mesh. How do I go about implementing the inflation layers using the adaptive meshing approach?

[Keyur Kanade]

When you use k-e with standard wall function, you will need y+ 30 to 300. 

Please check contours - turbulence - y+. Please identify the regions where y+ is less than 30 and then coarsen the mesh. 

If you use adaptive meshing to refine, it will further refine mesh and will further reduce y+