Stiffness Matrix of Frictionless Plate Support: Puzzling Differences

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
4 replies · 2K views
Mohamed_Wael
Messages
19
Reaction score
0
ImageUploadedByPhysics Forums1469943346.394802.jpg


Hi all,
In the attached photo, you can find a plate supported along the edge by "frictionless" support and I am trying to obtain the stiffness matrix using the node at the center. I wonder why the Kxx and Kyy (highlighted) are not equivalent or even near to each other, any suggestions?
 
Physics news on Phys.org
Im not familiar with ANSYS, but are those arrows your BCs (the supports)? It looks like you have "BCs" at the centerpoints between the edge nodes...Is that the case?

Also, are the BCs on the upper right hand side of the upper right half the same as the BCs on the lower left (asking because the arrows are different)?

Also, can you explain what you mean by the stiffness matrix at the node at the center? The Stiffness matrix applies to the entire structure (or a substructure), and doesn't describe a single point.
 
First thanks for your care,
1-actually I wanted to divide the plate in such homogeneous way trying to get equal Kxx and Kyy, but this lead to triangular base and not circular edge, but when u define the boundary conditions you define it for the circular edge geometry, that's why they look like flying supports.
2-I defined a frictionless support allover the edge they should be the same, I don't know why Ansys mark them in this way
3-when I extract the stiffness matrix I should define the DOF by selecting the nodes and the required DOF, that's what I mean
4-I have just tried changing the type of support to fixed support and I got equal Kxx and Kyy! i will attach the image.
thanks again
 

Attachments

  • Capture2.JPG
    Capture2.JPG
    38.9 KB · Views: 644
Good to hear.

The "support type" (constraints or BCs) is very critical in FEA. When I was learning, my constraints were the cause of most of my mistakes. It helps to think of things in terms of nodes instead of physical shapes (that is, as a series of triangles instead of a circle).
 
What kind of elements are you using for this model? Have you considered using elements with mid-side nodes to allow for a curved outer edge?