SimScale CAE Forum

Meshing as a Begineer


#1

I am currently trying to generate a mesh around a NACA0012 airfoil using Simscale. I have an enclosure built around my airfoil with a refinement region and surface refinement. Additionally I had inflation set on the airfoil which I removed due to thinking it may have been causing the lengthy meshing process. I havent been able to produce a mesh for it - always stops at 50%. I have attached the link below of the project.
https://www.simscale.com/workbench/?pid=8602421153110782559&mi=spec%3Adfd5b828-cd61-42bf-b4e3-ebeba38592c8%2Cservice%3AMESHING%2Cstrategy%3A5&sh=4

Thank you in advance for the help.


#2

Immediately thought about @DaleKramer and @Retsam here :wink: Both have already performed incredible mesh studies and can probably help you out with that one! Andrzej, want to give it a spin with an unstructured mesh? :wink:

Cheers!

Jousef


#3

Hi @ptreacy: Currently your volume (Enclosure) for internal HEX mesh is wrong. You need that volume to have a hole (airfoil removed) across. Just left the air in the domain, airfoil removed.

So if your airfoil has 25 cm span, enclosure should be <= 25 cm span: you should be able to see that hole in the domain. If it is not the case, you would be unable to generate correct mesh.

Personały, I create my simulation domain in CAD program (it is OnShape in my case), and it does not need to be a hexagonal shape. Here is the example of very small domain, with NACA0012 AoA defined in geometry:



Tell me how it works when ready,

Cheers,

Retsam


#4

Hi Retsam,

Thank you for replying. I was using ANSYS meshing before this and usually use the boolean command to remove the airfoil from the domain, wasnt exactly sure how it worked with SIMSCALE. Itā€™s working now so thank you!

Pippa


#5

I have fixed the enclosure but I am still unable to mesh a simple mesh. Any more guidance as to what I may be doing wrong? Thanks


#6

When generating Mesh Clip you will discover mesh defects and they look nasty:



Perhaps you should try first to use ā€˜Automatic BLā€™ and do not define them manually, but I cannot find an explanation of your mesh distortion (swiss cheese). I would also start by using moderate or coarse mesh resolution first. Log files report that your own BL definition is useless (15 BLs intended):

patch           faces    layers   overall thickness
                                  [m]       [%]
-----           -----    ------   ---       ---
B2_TE5_B2_TE121 319198   0.155    0.000214  1.84    

For simulation itself, you need both big size of the domain to be ā€˜Slip Wallsā€™. In Simulation Control please use ā€˜Potential Foam initialisationā€™.


#7

Hi Retsam,

How did you know that B2_TE5_B2_TE121 corresponded to the inflation boundary layer I had assigned?

I have implemented what you suggested and am waiting for it to run.

Thank you for all the help,
Pippa


#8

I found it in the log fileā€¦ Near the log end you will have statistics about BLs.