I am trying to model contact between a wire inside a tube-like structure (See picture). The wire is fixed to the tube at its end, the other end is free (the one that extends outside the tube). When a pulling force is applied, the wire causes the tube to bend.
However, I have a problem with the behaviour of the wire. It does not have flexibility. It behaves like a rigid beam and simply goes through the tube. It should be restricted to stay inside the tube somehow. But I dont know how to. Thus,
1. The wire is not flexible (I have used cbeam elements and once I used shell elements, none worked)
2. The tube is modelled with shell elements.
3. There is contact between the wire and the tube.
4. The wire should stay within the tube, instead of going through it.
I tried surface to surface contact, but it has not worked. How do you solve such a problem?
Contact NO PENETRATION only runs when you have surface-to-surface contact, then unless you mesh the wire with 3-D solid element the contact definition is not possible. But for the picture, it seams that dimensions of both bodies wire & tube are totally different, that meshing the wire with solid elements is not practical.
This is a complex problem if you want to study the contact between tube & wire, a linear static solution is useless because apart of contact you have other nonlinearities like geometric nonlinearity (ie, large displacement effect).
Here I simply suggest to try to capture the "effect" of the wire to stiffen the tube when bending loads exist, forgetting the contact effect: simply imprint the wire in the tube surface, mesh the tube with shell elements, and the wire with CROD 1-D element, use the same element size (ie, merge nodes) between wire & tube, and run a geometric nonlinear static analysis. To stabilize the model you can use symmetry in the longitudinal plane and study 1/2 model,OK?. Additionaly of tensile load in the CROD element (the wire) you can consider gravity
Good luck & best regards,
Thank you for the reply. I will try your suggestion, but I think ROD elements will not work in this case because they do not have rotational DOFs, thus I will use beam elements.
It depends, please note in my comment I suggest to merge nodes, then both Shell CQUAD4 & CBEAM are compatible elements, then both elements share rotational DOF. If you use CBEAM elements + merge nodes, then both elements are welded. Instead, using CROD elements only tension & compression stiffness is shared. Let´s suppose the wire only works in tension, then using CROD is better approach.
Use both CBEAM & CROD and compare results, is important to run nonlinear geometric analysis to extract the membrane stillness of the tube meshed with SHELL elements, OK?.