Surface triangle mesh with strict constraint

arnom · October 4, 2012, 14:56

Dear all,

as I'm from the meshfree community my skills in meshing are practically non-existent. However, I have the following problem at hand which I can't seem to be able to find a solution for.

Let's assume I have a general shell geometry and I would want a surface mesh of this shell. First of all, I can only use triangles and secondly I want to impose a limit on the edge length of the triangles.

Now, I tried Icem, gmsh and Salome and although they let me impose a limit, it is not strictly enforced. That means that if I set my max edge length = 0.1, there would always be edges with length > 0.1. Unfortunately that renders the meshing useless as I really need strict enforcement of this constraint. Thus my simple question, is there a meshing software that can achieve this?

Bonus question: The actual constraint that I have is the distance from the triangles barycentre to the vertices. Again, I'd need strict enforcement but I guess there is no meshing software that actually even has such a constraint or am I wrong?

Thanks in advance for your help,
Arno

julien.decharentenay · October 4, 2012, 19:56

Hi Arno,

I do not know of a meshing software that would fit your requirements. But how hard is it to subdivide a triangle to meet the distance requirements?

Julien

arnom · October 5, 2012, 04:34

Hi Julien,

thanks for your reply. Well obviously it's not hard to just split a triangle into two. But for computational reasons I'd like to have as less triangles as possible. As you might now meshfree methods are already slow, so I'd like to avoid every additional load if I can.

Cheers,
Arno

ogloth · October 5, 2012, 05:47

Hi Arno,

we develop an open-source tool called enGrid which has its own surface mesher. Although the feature you require is not there, it would probably not be too difficult to implement it. If you want to try, I could point you in the right direction. Or, alternatively, if you have a small budget we could implement it for you.

One (current) requirement of enGrid, however, could be a show-stopper: As geometry you need to provide a water-tight STL geometry (or similar) -- you cannot import CAD geometries directly.

Cheers,
Oliver

arnom · October 5, 2012, 05:57

Hi Oliver,

I'm a PhD student, so funds for these kinds of things are not available unfortunately. Also, I probably won't have time to work on it during my day to day work, so it'd be a free time project. I'll send you a pm with my email so that we can discuss further things in private without spamming around here.

The STL requirement is not a problem for me. From your website I see that I could use STEP or IGES which is pretty much all I need.

Cheers,
Arno

October 4, 2012, 14:56	Surface triangle mesh with strict constraint	#1
arnom New Member Arno Mayrhofer Join Date: Jan 2010 Location: University of Manchester Posts: 4 Rep Power: 16	Dear all, as I'm from the meshfree community my skills in meshing are practically non-existent. However, I have the following problem at hand which I can't seem to be able to find a solution for. Let's assume I have a general shell geometry and I would want a surface mesh of this shell. First of all, I can only use triangles and secondly I want to impose a limit on the edge length of the triangles. Now, I tried Icem, gmsh and Salome and although they let me impose a limit, it is not strictly enforced. That means that if I set my max edge length = 0.1, there would always be edges with length > 0.1. Unfortunately that renders the meshing useless as I really need strict enforcement of this constraint. Thus my simple question, is there a meshing software that can achieve this? Bonus question: The actual constraint that I have is the distance from the triangles barycentre to the vertices. Again, I'd need strict enforcement but I guess there is no meshing software that actually even has such a constraint or am I wrong? Thanks in advance for your help, Arno

October 4, 2012, 19:56		#2
julien.decharentenay Senior Member Julien de Charentenay Join Date: Jun 2009 Location: Australia Posts: 231 Rep Power: 18	Hi Arno, I do not know of a meshing software that would fit your requirements. But how hard is it to subdivide a triangle to meet the distance requirements? Julien __________________ --- Julien de Charentenay

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
how to set periodic boundary conditions	Ganesh	FLUENT	15	November 18, 2020 07:09
[snappyHexMesh] SnappyHexMesh for internal Flow	vishwa	OpenFOAM Meshing & Mesh Conversion	24	June 27, 2016 09:54
[ICEM] surface mesh merging problem	everest	ANSYS Meshing & Geometry	44	April 14, 2016 07:41
[ICEM] Hexa mesh, curve mesh setup, bunching law	Anorky	ANSYS Meshing & Geometry	4	November 12, 2014 01:27
[snappyHexMesh] external flow with snappyHexMesh	chelvistero	OpenFOAM Meshing & Mesh Conversion	11	January 15, 2010 20:43

October 5, 2012, 04:34		#3
arnom New Member Arno Mayrhofer Join Date: Jan 2010 Location: University of Manchester Posts: 4 Rep Power: 16	Hi Julien, thanks for your reply. Well obviously it's not hard to just split a triangle into two. But for computational reasons I'd like to have as less triangles as possible. As you might now meshfree methods are already slow, so I'd like to avoid every additional load if I can. Cheers, Arno

October 5, 2012, 05:47		#4
ogloth Senior Member Oliver Gloth Join Date: Mar 2009 Location: Todtnau, Germany Posts: 121 Rep Power: 17	Hi Arno, we develop an open-source tool called enGrid which has its own surface mesher. Although the feature you require is not there, it would probably not be too difficult to implement it. If you want to try, I could point you in the right direction. Or, alternatively, if you have a small budget we could implement it for you. One (current) requirement of enGrid, however, could be a show-stopper: As geometry you need to provide a water-tight STL geometry (or similar) -- you cannot import CAD geometries directly. Cheers, Oliver

October 5, 2012, 05:57		#5
arnom New Member Arno Mayrhofer Join Date: Jan 2010 Location: University of Manchester Posts: 4 Rep Power: 16	Hi Oliver, I'm a PhD student, so funds for these kinds of things are not available unfortunately. Also, I probably won't have time to work on it during my day to day work, so it'd be a free time project. I'll send you a pm with my email so that we can discuss further things in private without spamming around here. The STL requirement is not a problem for me. From your website I see that I could use STEP or IGES which is pretty much all I need. Cheers, Arno