[Vignesh2508]

Hello Guys,

I have a problem with fixing the refinement box dimension since it is changing automatically. I do not know why snappyhexmesh is showing this behaviour.

I was conducting a mesh independence study where i have to increase the number of cells and check the changes in the coefficient of drag and arrive at a better mesh. But when i tried to increase the number in the base mesh(i.e the blockmesh) the final mesh after the SHM is not what i expect. That is the dimension of the refinement box is changing after i mesh using SHM. I have attached the picture of the picture of the mesh i obtained. And i have also the snappyhexmeshdict file along for reference.

Please take a look at the picture and help me out in figuring out the reason for this strange behaviour.

Thanks in advance

Vicky

[Antimony]

Hi,

I don't quite understand what you mean by sHM changing the dimensions of your refinement region.

What I noticed was that your maxGlobalCells is quite low, only at 2e6. Are you sure with all the levels of refinement you have specified, 2e6 cells will be enough?

It is possible that refinement does not happen according to how you expect it because it hits this limit early.

Hope this helps.

Cheers,
Antimony

[Vignesh2508]

Yeah i figured that the max global cells was low and then changed it. Now its working properly.

Even then I was wondering that the SHM should stop and give me an error if it detects that max global cells is low. It used to do it when i was using openfoam 3.0.1 but in openfoam 4, if you can see the image you can notice that the refinement box was not what i mentioned in the SHM dict.

Am i wrong in my assumption that the process will be killed and a error will be displayed if i do not provide a good number of maxGlobcells.

Thanks

Vicky

[Antimony]

Hi,

sHM to the best of my knowledge has never terminated if the number of maxGlobalCells is low, as it assumes that the user knows what exactly he/she wants when specifying it in the sHMDict file.

In my opinion, that approach by sHM is logical. After all, there are a myriad number of ways in which you might want to run sHM and terminating when your maxGlobalCells is deemed too low, can be incorrect. As an extreme example, if I wanted to use sHM to simply remove cells intersected by surfaces and have a very fine blockMesh whose cell count exceeds sHM's maxGlobalCells, why should sHM terminate with an error?

As far as I understand it, it is best to ensure that when sHM works, it never reaches that maxGlobalCells limit. The minute it does, there is always going to be the chance that the mesh has been insufficiently refined. Remember that the final mesh count from sHM can be much lower than the maxGlobalCells, once it removes cells that have been intersected by the surface.

Hope this clarifies.

Cheers,
Antimony

[Vignesh2508]

Thank you.

Right now I am trying to add layers to my model in sHM. Do you know if there is any tool where I can determine the final layer thickness and expansion ratio for the layer addition process for a required y+. Or do I have do it in the traditional way of calculating them through the 'estimated wall distance'?

As you can see from my sHM dict file I have tried to add 10 layers and when I try to increase the layers to a number of 20, something happens and all the layers are collapsed and I can not see the layers after meshing. Do you know why sHM shows this behavior?

Thanks

Vicky

[KaLium]

Originally Posted by Antimony

Hi,

sHM to the best of my knowledge has never terminated if the number of maxGlobalCells is low, as it assumes that the user knows what exactly he/she wants when specifying it in the sHMDict file.

Agreed. If maxGlobalCells is too low, sHM skips some of the refiments.