Split coincident nodes 1

[smithson1984]

Hello Everyone,

I have an issue in a static structural analysis. I am designing a structure which is to be manufactured from many similar plates closely fitted together. I have drawn this within the CAD package that I use with the edges of each plate being coincident with the next one. Now, when I perform a Static Structural analysis Mechanic thinks that the plate bodies share nodes and therefore perfectly transfer the stresses etc. between one another. However this is not the case. I desire to have each bodies edge coincident in space but with separate nodes associated with it so I can then add a frictional contact between them. Does anyone know if there is a node split tool within either Design Modeler or Mechanical, or if not, if there is another way of doing this?

Any help would be greatly appreciated!

Many thanks,

Ian

 

[flash3780]

Ansys Workbench (Mechanical) will join separate meshes with coincident surfaces using contact elements (automagically -- which I find to be a pretty terrible feature). I'd recommend joining meshes where it makes sense to do so (in DesignModeler: Right click on parts to share a continuous mesh >> Select "Form New Part"), and changing your contact behavior where you'd like to simulate contact such that it doesn't bond the surfaces together.

In your case, it sounds like you've got bonded contact where you intend to have standard contact. You should be able to change your contact pair to behave as you intend.

If I'm not understanding your problem correctly, please post a picture.

 

[smithson1984]

Hi Flash,

Thanks very much for the reply.

It seems like you have a decent grasp of what I am attempting to do but I have attached an image nonetheless. This shows the deformation results I am getting. The model has 6 steel 'stools' upon which are mounted two panels which are separate bodies but have coincident edges in the middle of the image. The model also involves a reinforcing core material. There are only currently two contact conditions:

1) face/face between the panels and the stools
2) face/face between the core material inside the panel and the panel's outer shell

So there is no 'official' contact between the two panels themselves however they seamlessly transfer the loads as can be seen by the results. I wish to model it as frictional contact but Mechanic will not allow me to select the relevant faces to do so. I am assuming this is because they share nodes and therefore no 'slip' is allowed between them.

What do you think?

Many thanks once again,

Ian

 

[smithson1984]

PS,

The loads are hydrostatic pressure loads. The panels are at different depths and therefore I would expect different deformation from the two and hence 'slip' between them in the middle.

 

[smithson1984]

Also, this is an image with one panel supressed. The green face is the one which I wish to make the frictional contact with, however I can't find the way to do it!

 

[flash3780]

It's not clear to me why you're unable to put a contact pair between those two surfaces. Can you clarify what is is that you're modeling? Just curious, but it looks like there's some symmetry in this problem that you could take advantage of.

If the panels are separate "parts" in Workbench, you should be able to insert a contact pair between them. Or, in this case, you may need to change the contact pair from bonded to standard contact. Can you tell us what's listed under the "Contact" pulldown?

 

[eltonrcarvalho]

Flash good morning,

Enjoying your first explanation, I believe that i`m having a similar problem, as shown in follow picture. I have a solid pin, what is loaded in the tips with 3000N, and, in de middle, the pin is fixed (fixed support). The problem is: the solution don`t converge. There`s some tipe of singularity, mas i can`t find it. Do you know what occur? The problem happen in the boundary conditions of the fixed region.

Thanks!