Hey
We’re working on a project for one of our customers and could use some advice on resolving a specific challenge. Due to our NDA, I can only share the components in question rather than the full assembly.
We are conducting a modal analysis on a complex machine and have determined that the analysis should also include preloading. The setup involves four rods, fixed at both ends, which are intended to stiffen the assembly and improve the structural response.
Approach Taken
To apply loading, I opted to replace the 3D rod elements with connector elements. Specifically, I used a beam section where both the material and rod radius could be assigned. When running the modal analysis without preloading, the results were nearly identical to the 3D rod version.
However, issues arose when I introduced a static step and attempted to apply loading to the connector, as it turned out this was not possible.
I suspected this was due to all DOFs being constrained in the beam element. To address this, I tried using a translator connector with the x-direction set free. I re-ran the analysis without preloading, but this time, the results were significantly different—which was unexpected. In this case the material can't be defined so I added elasticity to the connector element which I believe is the stiffness of the rod since the unit is kg_s2. Therefore, I calculated the axial stiffness. Yet, as I said the result is different.
Key Questions
What causes the discrepancy between beam and translator elements?
I can’t share images, but I observed that the 4 rods rotate around the middle axis through the X-support, which might be an important clue in evaluating why the beam element works but the translator element does not.
What workaround would allow me to apply preloading to a beam element?
Given the current limitations, is there a way to introduce preloading while keeping the beam representation?
Any insights or suggestions would be greatly appreciated.
