Hello,
I’m trying to make a telescopic support. I built a very rough model for practice. I can’t get the clearances around the inner profile right. I think the clearance is causing the contact alignment issue. To enforce kinematics, I added a light spring. How can I initially connect the inner body to the outer soft spring?
Can anyone suggest a better solution? Could I replace the contact with a soft material with a v=0 value? I’m currently analyzing the entire model. I know it will be a bit easier with 1/2.
I take every response seriously and would be grateful for any ideas.
The plastic material is causing the problem.
160.00 0.00
170.00 0.00
180.00 0.01
190.00 0.01
200.00 0.02
210.00 0.05
215.00 0.20
0.00 0.20
I can’t seem to refine the mesh at the expected contact point.
Is there another way to do this?
Telescopic beam (x).pmx (4.9 MB)
How about making a tighter gap and maybe using Adjust in contact ?
Do you mean that you want to remove it during the analysis ? This should be doable with some keyword edits. Or are you just looking for a way to define the connection ?
Theoretically yes - sometimes contact is replaced by element layers with specific linear elasticity (e.g. compression-only, proper E and v). But adjusted contact with supporting soft springs should be better.
Yes, symmetry would help here, also by reducing the number of needed BCs.
The last point is wrong, try extrapolating it further instead. Solvers don’t like entering plastic plateau.
Extruded mesh can be refined only on the source surface, then it will propagate (extrude) the refinement along the whole profile. You could split the parts into more subvolumes (also smaller ones around the contact region) and use the Transfinite algorithm.
It would be better to use C3D8R (more of them through the thickness) instead of C3D8 elements.
Or maybe you could try with shells.
1 Like
Thank you for the quick advice 
Sure, later I will share refs to some research papers that could be helpful. Their authors often choose shell elements. They also include so-called “sliding pads” that connect both profiles:
1 Like
I know it can be reduced to such a model and it will be safe. But with the test results, it can’t be made dependent on displacement. The force is unknown. The article is too general.
I’m trying to force contact for a structure with play. The weak spring is there to make it easier to engage the contact at the beginning of the iteration.
Here are the papers I’ve found:
- “Finite Element Analysis and Stress Testing for Telescopic Boom of Telehandler” H. Yan et al.
- “Buckling Analysis of Telescopic Boom: Theoretical and Numerical Verification of Sliding Pads” A. Pavlovic et al.
- “Static and Dynamic Analysis of Telescopic Boom of Self-propelled Tunnelling Machine” D. Derlukiewicz et al.
- “A one-dimensional dynamic model of thin-walled telescopic boom considering complicated cross-section deformation behaviors” L. Zhang et al.
- “Stress analysis in contact zone between the segments of telescopic booms of hydraulic truck cranes” M. Savković et al.
- “Structural Optimization of the Telescopic Boom of a Certain Type of Truck-Mounted Crane” Z.W. Chen et al.
Most of them discuss full models (both segments). Sliding pads could still be included there to provide means for load transfer.
1 Like
Unlike Abaqus, CalculiX doesn’t have any automatic/contact stabilization features for such cases. Theoretically, using implicit dynamics with high damping could work, but from my experience, this procedure struggles a lot to converge. So springs might be the best way. Or even better dashpots (since they work only when necessary), but those are not supported in PrePoMax and require keyword edits.
