Advice on how to keep cells inside a closed contour
Hi Foamers,
I'm new to OF and so i'm struggling with a complex structure. My structure has a curve configuration and on top of that there's a bunch of openings (see picture). My problem is that after running SHM the contour of the structure+openings isn't deleted and only the cells between the openings and inside the contour are kept. The .stl was created using Salome and with Helyx i saw that it was ok. As you can see in the picture, the SHM is cutting the main contour (structure) but keeping the cells of the openings assuming that it isn't a contour. I wanna create "holes" in the mesh to let the fluid pass through. I tried to use a different .stl for the openings but the result was the same. After some research i thought that maybe if i use multiRegionFeatureSnap the problem could be solved. But it persisted. Can anyone, please, tell me if there's a way of doing this and give me some guidance? Best regards Cristina Openings: https://www.dropbox.com/s/ulcj371mt7...te181.bmp?dl=0 Problem: https://www.dropbox.com/s/pq1pl8aoyo...tura2.png?dl=0 SHM: Code:
/*--------------------------------*- C++ -*----------------------------------*\ |
if you post your entire case, then we can debug it. The issue isn't quite clear so we will need just a bit more information.
|
I'm so sorry Dan but i can't put here the entire case.
I know it's hard to help without further information but if you could only point me a direction of what could be wrong, i would appreciate it. Best regards Cristina |
Cristina,
No worries, it is most likely that your location in mesh is not within your opening itself and during the castellation step, the cells are swept away in the undesired region. Make sure that you have a the proper location (i.e. material point) for what represents your fluid region. If you are doing a CHT type simulation then you can think about the multregion support in snappy. |
Hi Dan,
I'm sorry for the delay, i had to much work. My mesh is 10x0.34x0.8 and the openings are ~ at 9.5m so the point, (0 0.02 0), in mesh is further away from the openings. I'm not doing a CHT simulation so i had removed the multiregional option. Right now i'm on a battle against time, but in a week or so i will try again. Thank you so much for your help, i really appreciate it. Best regards Cristina |
Hello Cristina,
Regarding the stl surface file, have you included the curve part in your the stl surface file? Is your stl file a close surface ? Also, make sure that your initial mesh (generated with blockMesh I assume) englobes the entire stl surface before generating a mesh with snappy. Regards, Daniel |
Hi Daniel,
First of all thank you for your reply. Yes, i did and without the openings everything work just fine. The problem appears after the consideration of the openings. I double check everything and couldn't find any problem regarding the .stl. The base mesh was done using blockMesh, using a domain 10x0.34x0.8 that englobes all .stl surface. Best Regards Cristina |
Hello Cristina,
Then the problem seems to be coming from the intersection (for the openings) of the stl file and the base mesh. Before the Castellated mesh is generated, snappyHexMesh will compute all the edges of the base mesh that intersect the stl surface. The tolerance of the mesh (or resolution) plays an important part in finding out all the edges that intersect the stl surface. I suggest that your blockMesh be a cube with perfect cubic elements (let's say 11 X 11 X11 with a subdivion in the xyz direction of 10). This will help in the Snap mesh step.Make sure that you have a big enough offset between the stl surface and the initial mesh surfaces. Daniel |
All times are GMT -4. The time now is 16:23. |