Hello,
I am trying to create a tie constraint between a beam (I-profile) and an end plate. The input setup seems reasonable to me. I searched for contact pairs and applied a TIE constraint between them. However, when I review the results (despite the analysis failing, it still displays results), the deformed shape shows the beam undergoing rigid body motion, as if it has flown away.
I’m wondering what I might be doing wrong in defining or applying the tie constraint.
Keep in mind that master surface should be the one with a coarser mesh. It’s good to mesh in such a way that the nodes are close to each other but you can also adjust the tolerance parameter. Then run a frequency analysis to see if the connection is working properly.
Thanks @FEAnalyst , I appreciate your help. I ran a frequency analysis, and the end plate and beam appear to be tied correctly. The error I’m encountering seems to be related to the spring elements I’m using to simulate the bolted connection between the end plate and the column flange. Specifically, I created a reference point at each bolt hole (on both the plate and the column flange), applied rigid constraints between each reference point and its corresponding hole edges, and then used CalculiX keywords to connect the reference points with springs as follows:
*ELEMENT, TYPE=SPRINGA, ELSET=SPRING
10001, 26153, 26177
10002, 26155, 26179
10003, 26157, 26181
10004, 26159, 26183
10005, 26161, 26185
10006, 26163, 26187
10007, 26165, 26189
10008, 26167, 26191
10009, 26169, 26193
10010, 26171, 26195
10011, 26173, 26197
10012, 26175, 26199
*SPRING, ELSET=SPRING
850000.
I am starting to think the issue is due to these spring elements only limiting translation in Y direction (axis of the element), any idea how to make the spring (bolt) limit translation also in shear directions?
You should add more springs - one for each degree of freedom. So 3 springs per bolt. You can use SPRING2 elements and specify their degrees of freedom:
@FEAnalyst I can’t really imagine that. So should it be 3 springs going between the same 2 nodes at each bolt location one for x, y and z accordingly ? or do I need different nodes ? Is beam elements a better approach than springs ? Thank you!
Same two nodes but separate spring elements with different DOFs. You could try with beams but springs are easier to define.
Check this thread for more details: Why the top plate displacement is zero?
Thanks @FEAnalyst for your prompt reply. I added the 3 spring in X,Y and Z direction but still seem to get the same behaviour in the frequency Analysis while the static step analysis never converges. Do you mind having a look at the file if you have a second. I am new to prepomax and trying to get better.
There’s no DOF parameter for the *SPRING keyword. Use this syntax to define SPRING2 elements:
*SPRING, ELSET=elset_name
dof_1, dof_2
k
Here dof_1 and dof_2 will be the same, e.g. for Z:
*SPRING, ELSET=elset_name
3, 3
10000000
Perfect! The frequency analysis now works perfectly! Really appreciate your help @FEAnalyst . Now I’ll spend some time to figure why the static step is not converging.
