Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

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

Meshing of a complex geometry tissue in Abaqus

Status
Not open for further replies.

sepantafa

Bioengineer
Aug 29, 2012
10
Hi everyone,

I am trying to do a finite element analysis of the Temporomandibular joint disc (TMJ disc) in Abaqus. I already got the IGES model of disc from Hypermesh but the geometry needs so much partitioning to become suitable for at least Tetrahedral mesh. Would you please tell me if the partitioning is the only way and if it is the what would be the best way to do it?

I think there should be easier method to do it since I have sen much more complex geometries (like brain tissue) which has been meshed by Abaqus. I would appreciate if anyone can help me.
 
Replies continue below

Recommended for you

Do yourself a favor, don't bother making these geometries using MRI/CT scans; just make realistic but simple geometries using SolidWorks. Use boundary or loft feature to make a 3D model and export it as an .SAT or as a .STEP file which can be easily imported into ABAQUS and meshed in seconds.

 
Is there no way to decimate the model?

TTFN
faq731-376
7ofakss
 
Another option is to smoothen the geometry as best as you can. Geomagics is pretty good at it and MIMICS and Simpleware have functions to accomplish that. Also, freely available Slicer3D may also have these options. Finally, another freely available and user friendly software called IAFeMesh may get you a hex mesh for this geometry.

 
Thanks for your quick responses. Actually the real geometry of this tissue is of high importance due to its thickness variation along the anteroposterior and mediolateral directions. On the other hand, According to literature,recently, they have been using MRI/CT even more than before. I am also planning to consider collagen fibers in the analysis which makes it totally important to have the most realistic geometry.

I am just wondering if this is the normal way of meshing such a complex geometry and if the answer is yes, then how they do meshing on much much more complex geometries in Abaqus? Don't you think it's better to do the meshing in Hypermesh (which I think is a better software for meshing although I am not sure if it also needs partitioning or not) and then import the model including mesh to Abaqus as a .inp file?

Btw, @IRstuff: I am not sure what you mean by decimating it, but i want to run an analysis on the whole model.

I appreciate your helps.
 
@IceBreakerSours: you're definitely right, Abaqus is not a good solid modeling software. I used to work with Solidworks and it's quiet more user friendly in terms of making solid models. At the moment, the University doesn't support Solidwork, otherwise I would import 2D image stacks of tissue into it and make a 3D reconstruction which could easily be meshed.

Btw regarding the meshing software you mentioned above, do you mean I can do the meshing there and then import it to Abaqus? Is the meshing procedure different from Abaqus (doesn't need partitioning)? because I know on of my friend used Hypermesh to mesh the model with Tetrahedral element and he said the software did it automatically although the mesh quality might be not that good. Then he imported the meshed model into abaqus and did the rest.

@rstupplebeen: I've tried working with Virtual topology which makes it smoother as you said but I was wondering if the partitioning is the only method to do mesh such a geometry.

Tnx
 
Yes, almost every program I mentioned previously provides an output format that ABAQUS can import. I am not sure about Slicer3D.

Also, try to zoom in into the geometry and check for some small features like edges. You must ignore those features using virtual topology. Finally, quadratic elements can conform to curved geometries but before you use quadratic elements, make sure the geometry is free of any issues. Note that quadratic elements are not recommended for contact (unless they are modified).

 
And I have another question regarding making a fiber-reinforced model. the collagen fiber can be modeled as fibers inside the tissue. Basically, there are two main methods to model fiber reinforced materials , one is to use spring btw the nodes which makes the fiber directions limited to node distances and the other one is to use continuum model and define two separate material for the fibers and the medium surrounding the fibers including the interaction btw them.

Since I am new to abaqus, I don't know from where to start. Do I have to simply make holes inside my models and pass tube shape fibers through them and then define a material for them? in which step should I use the spring between the nodes? What is the SPRINGA element? I found all these keywords while i was searching but I still haven't managed to implement any technique yet. I would appreciate if anyone can give me a clear framework, and ideally, an step by step procedure although i know it's different case to case and you might need to know more details.
 
Use the Holzapfel material model. Up to 3 fiber directions (anisotropy, to be accurate) can be incorporated. In the Holzapfel model, fibers don't take load under compression (ground substance does). See the Abaqus documentation for details.

 
Use Hypermesh if you have it. You can check and correct mesh quaility and even downsample if you need to.

We use Hypermesh or Amira and dont really touch Abaqus CAE until the last step of setting the model up. Simpleware is also really good especially for setting up contact surfaces.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor