Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

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

Meshing a solid plate with irregular surface due to corrosion pits

Status
Not open for further replies.

alima2065

Civil/Environmental
Nov 22, 2011
31
Hi everybody,
I am trying to mesh a rectangular plate with hexahedral elements (not tetrahedral) in Abaqus. But one side of the plate is covered by elliptical pits as shown in the attached figure (pits are random in place and geometry). As a result, the Abaqus mesher can not do it automatically But I need it without using manual partitioning. I appreciate any suggestion or any software that can cope with such a problem.
Capture_hz9mb5.png
 
Replies continue below

Recommended for you

Partitioning will be necessary in this case but maybe you could automate it with Python scripting (not an easy task though). Or you can try preparing the geometry in some advanced mesher that allows easier automation, such as Hypermesh or Ansa
 
Thanks for your reply. Actually, the problem is that when the number of pits increases (assume 100 pits for example that can overlap) the partitioning also does not work, and meshing reaches an error even when I tried manually. Do you think Hypermesh or Ansa can do meshing in a completely automatic procedure? Do they have an API? because I need to mesh it in an iterative manner.
 
These programs are both very good meshers and allow for some automation but I’m not sure to what extent. However, you may have to use tetrahedrons instead. They are not that bad when used properly and sometimes there’s no other choice.
 
Yes, I see. Because I want to model the ductile fracture of the plate (large plastic deformation) it is important to use fine mesh in pits. Therefore num of the elements is high and using reduced integration is a great help. on the other hand, the first-order tetrahedrons are so stiff and the results for one single pit showed inaccuracy and so I need to use second-order tetrahedrons that it means more and more computational effort.
 
Indeed, with many holes such tet mesh will require quite large computational resources but with proper choice of element size and local refinement around the holes you may reduce this requirement significantly. If you have access to HPC, that’s the way to go here.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor