[technoB]

Hi,
I'm attempting to simulate turbulent flow over a rectangular shape and i'm having trouble fixing the meshing around the corners. I used an automated mesh(2D) with a few volumetric controls for mesh refinement. My prism layer stretching is 1.265, prism layer total Thickness is 0.15m. Attached is an image of what my meshing is and what i'm trying to achieve. In the attachment Image A is what i got. I am trying to get something similar to image B or anything better.

Thanks in advance

TechnoB

[Nikpap]

Have you tried using Advancing Layer Mesher instead of Prism Layer Mesher?

[technoB]

Thank you for the suggestion. I am unfamiliar with the Advanced layer mesher , i have attempted it and it looks better than before. If at all you have any tips on improving the meshing that will be great.

[Nikpap]

I am glad it helped. Unfortunately I cannot think of any other tips at the moment besides the classic "play around with your number of layers, layer thickness" etc.

Maybe it would help to check the documentation of the Advancing Layer Mesher. Nonetheless, to me it seems you have quite a good starting point now.

[technoB]

Hi,

Is there a way to fix the corner cells? I feel this may introduce a slight inaccuracy in my results. The attached image is the meshing at the corner.

[technoB]

Hi,

I just realised that using the advanced layer mesher brought up a complication i'm not sure how to fix. The total prism layer thickness does not match the measured height. i.e., my total prism layer height is 0.353m in the properties , but after executing the mesh it reads 0.07m when I check with the ruler

How would you advise I correct this?
Thanks in advance

[Nikpap]

Are you using any custom controls in your mesh that might be interfering with you global settings?

Also when using the ruler it is easy to missclick further inside or outside your domain. Are you sure about the value it reads? Also you should make sure you are not measuring close to a corner.

[technoB]

Hi,

Thanks for taking the time to respond. I have uploaded a snip of what I mean. I added a volumteric control to refine areas around the body and refine the wake region, i also customized the surface remesher. I also removed prism layers at the far field boundaries.

I changed the distribution mode from stretch factor to wall thickness, played around with the number of prism layers, there is still no change. I'm not sure what other setting could drastically have such an effect on the height.

However, my previous prism layer mesher had the correct prim layer total height but the mesh was not excellent as the prism layers collapsed at the corners as seen previously in the thread.

Thanks again in advance

[Nikpap]

In your caption you have the total prism layer thickness as 3.53 m not 0.353 m. Could it be that the units are wrong?

If that is not the case have you tried setting an absolute value instead for the thickness instead of a percentage of the base?

[technoB]

Hi,

The previous image was an attempt at increasing the height to see any change from before. I have also tried using an absolute height but it did not change. I have not tried adjusting the prism layers using a surface control yet, not sure if that would work.