Membrane Structure - Direction of Pressure

Using SimScale
1

Hi @ggiraldo, I just saw your reply and it helped me a lot. I do have a follow up question though:

I’m running a numerical wind tunnel test of a thin membrane structure. The membrane doesn’t have a thickness (in real life it is 1-3 mm thick) so when I extract the result as you are suggesting in what direction is the pressure then acting? When the wind is hitting the surface it will be pressure on one side and suction on the other side and since I only get one pressure value for each point I assume that is the difference of pressure and suction. That is all good but I need to know in what direction it acts.

Thankful for help here.

Cheers,
David

2

Hi @sippanspojk, thanks for your question!

I think the direct way to know is to check surface normals. In Paraview you can extract surface normals and plot them as arrow glyphs. Positive pressure should be in the direction contrary to the normal, and negative pressure in the direction of the normal.

I am just curious about your case. How did you simulate flow around a membrane? Was this done in SimScale? Some other package? Does it include the membrane deformation? Can we have a look at your project? Is the membrane really zero thickness or just very thin? Because if it has top and bottom surfaces, you can apply the former advice and you will be fine.

PD @jousefm maybe fork this into a new thread?

3

I model the membrane as a surface in Rhino3D and then export it as .stl and then mesh it using snappyHexMesh in OpenFoam (I was looking for answers regarding paraView so this is the reason I ended up on this forum). I am not doing FSI so the membrane acts as a stiff geometry that the wind is flowing around.

This works if all cells are facing the same direction but I am not sure if that is guaranteed. Also I extract results in the points that belongs to more than one cell so if the neighboring cells face the opposite direction it is even trickier.

I will have a look in paraView as soon as I get the chance though and report back.

Thanks for your help so far!

//david

1 Like
4

I advice you to model your membrane with some thickness, even if it is very low (1~3 mm is not that little). Then make sure surface normals are pointing outward and your mesh refinement is low enough.

1 Like