Combining Solid Section with Gmsh Shell

[nrosenblum]

Dear Forum,

I am trying to create a very primitive aircraft wing structure. I have a frame consisting of ribs and spars on the inside and then a skin on the outside.

So I created this frame:

It is covered by this skin:

I used the skin to create the frame in order to create the contours in the frame. So the frame combined with the skin looks like this:

I want to used 3D solid tetrahedral elements for the frame and then use Gmsh quad shell elements for the skin. I have offset the thickness of the shell elements by 1/2 of the thickness of the shell elements. I then created a TIE constraint between the frame and the skin. I have tried selecting parts, surfaces, etc. and I get this extremely peculiar behavior that makes absolutely no sense to me as shown below:

When I run a simulation of ONLY the frame, this is what I get (upward wing tip load and fixed at base of canitlever):

When I run a simulation of ONLY the skin, this is what I get (upward wing tip load and fixed at base of canitlever):

Any ideas about what I might be doing wrong? Can I combine 3D solid and Gmsh shell elements together?

TIA,
–Neal

[FEAnalyst]

You can use solid and shell elements in the same model if they are properly connected (usually with tie constraints). Check the mesh with expanded shells (for that, you can use the “Check Model” option instead of submitting the analysis normally). But even more importantly, instead of running a static analysis, use a frequency step and check the deformed mode shapes. They may indicate issues with connections.

The meshes are rather coarse, maybe too coarse for tie constraint to connect the nodes.