Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Applying displacement BCs to nodes

Status
Not open for further replies.

barrind

Mechanical
May 24, 2006
31
0
0
NL
Not sure if I am going to be able to explain what I want here, but here goes.

I have a model, part of which I need to remesh to get a much finer mesh. Try as I might I cannot get the seeding right for the mesh to propogate properly and give me a nice coarse mesh far away from this region using bricks. I always end up with either far far too many elements and/or badly distorted elemnents as the mesher tries to get round changes in geometry. I have been advised it is best to chop the model up and rejoin using ties, such that each part can have its own mesh. This sounds ok but can it be taken one stage further? Tell me if this is possible:

I put partitions in as if I were going to split the model into several parts, but don't split it and leave it as a full model. Then I apply a decent mesh, enough to get a pretty good result for overall displacements of the various parts of the model. I therefore have an odb with nodal displacements for the full model. I then go back and do the geometry cuts to remove the unwanted parts of the model and give the new part a finer mesh. Is it then possible to apply nodal displacement boundary conditions to the parts of the model which would have been ties in the other method? I realise there will be more nodes on these faces with a finer mesh so I am assuming some interpolation based on x,y,z coords will need to be done.

I bet this is as clear as mud.
 
Replies continue below

Recommended for you

I believe if you apply boundary condition on a edge or face. Then the boundary condition will be applied on the nodes within the edge or face automatically regardless of how fine the mesh is.

Tie surface give noise to the results I believe. So tie surface close to the area of interested need should be avoided. (Tie used to be quite bad but I think ABAQUS have introduced the surface to surface tie which supposed to have less noise).

However, I still prefer to avoid using tie unless I have to use it, but that just my personal taste.
 
Depending on how much time you have on your hands, careful partitioning and selection of the meshing methods will let you achieve the mesh you want on your global model.

Submodelling will likely save you considerable time and efort.
 
Status
Not open for further replies.
Back
Top