Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

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

Stent Meshing using Tet Elements

suz_02

Student
Jan 22, 2025
12
1738687738960.png

For the stent image provided (mind the low resolution image), are there any suggestions of how to mesh this for explicit analysis. I had an issue where due to the large number and small size of mesh elements, the stable time increment was very small (x10^-11) which meant that the simulation would take a long time to run. I am trying to increase the stable time increment so that my simulation will run much faster.
1738687880950.png
I have been trying to use the virtual topology to merge all of the faces together to create a single face and then playing around with partitions to try and make the mesh. However I have ran into issues where the stable time increment cannot be calculated due to mesh issues. There have also been found to be sharp angles in the geometry which can alter the mesh, I have used the ignore topology for those that were found.
 
Replies continue below

Recommended for you

Can't you simplify this geometry a bit, reduce it to midlines and mesh with beam elements ? It will be hard to get a tet mesh of a reasonable density for explicit analysis here.
 
Would this mean that I would have to somehow convert this geometry into its midline’s, and then is beam elements an option you choose when you sketch the part? I have not performed meshing with beam elements in a while.

Could you please give me some more information or resources on how I can do this?
 
I don't think you can do it automatically in this case. You would have to manually draw a 3D sketch with lines representing this geometry (likely in some CAD software, not in Abaqus/CAE).
 
Algorithms to compute centerlines using CG (computational geometry) libraries are available in various packages (CGAL, for sure). However, if you are not a power coder, I wouldn't recommend going down that pathway. A simpler route might be to look at 3DSlicer; it has a VMTK plugin for computing centerlines. You could give that a try. Don't know how well it is going to work but it won't take more than an hour or so to give it a shot.

Now, the underlying geometry is almost (but not quite) axially symmetric so you could carve out one or more longitudinal repeating series of struts, point and click a whole bunch of times to create centerlines manually for that one or more series of struts, and then finally copy/mirror those in the circumferential direction. It will be painstaking work that does not advance your skill-set but it will get the job done.

What is the ultimate intent of this work? My guess is you'd like to deploy this stent in an anatomy and compare the results with post-intervention CT scan. First thing to think about is - how will you do the comparison? Depending on the ultimate goal and resources available, you may not want to do it manually or just by visual inspection alone. Also, if the comparison is simply at one time-point (say, diastole), then I am not sure what is novel about the work - maybe it is, maybe not - I do not know. But, if you automate the comparison over multiple time-points, you may have something novel.
 

Part and Inventory Search

Sponsor