Good afternoon,

I’m currently building a model on the 3DEXPERIENCE Cloud, in which fiber‑reinforced tubes are inserted into mating sleeves and bonded. I have a question about mesh setup. Right now, I’m comparing two different approaches for meshing the tubes.

Variant 1:

In the first approach, I mesh the entire tube with a surface‐quad mesh and assign the appropriate composite‐shell section to it.

Variant 2:

In the second approach, I split the tube mesh into three separate meshes. The blue and yellow meshes represent the bonding surfaces. In order to connect all three meshes, I use the “Capture” feature on the middle mesh and select the Condensation mode for the shared nodes. According to the User Assistance, this method should share all degrees of freedom between the meshes at the mesh border. I chose this approach so that I can locally refine the mesh in the adhesive zones. For the sake of comparison, I used the same global element size in all meshes.

Simulation setup:

To compare the two variants, I built a simplified simulation in which I bonded one end of the tube into a sleeve using the standard “Bond” option in the Structural Model Creation App. The connection was defined between the sleeve’s mating surface and the corresponding element set on the tube. A force is applied at the free end of the tube so that I can analyze the response of the bond.

Results:

In the variant with three separate meshes (shown on the right), a distinct jump in stress appears at the interface between the two meshes. The continuous mesh (shown on the left) does not exhibit such a pronounced jump, although in both cases there is a stress peak just downstream of the bond, where the continuous‐mesh variant shows a smoother stress gradient.

Question:

My question is whether node‑condensation does not fully couple all degrees of freedom between the meshes, leading to a non‑uniform transmission of forces and displacements, and thus causes the stress discontinuity at the mesh interface. Alternatively, what options do I have to locally refine the mesh in the adhesive zone while keeping the tube as a single continuous mesh? I have not yet found a way to do this. Edge Distribution didn’t work for me, and I may not be aware of all the available functions.