I am trying to simulate two flanges connected by bolts. The flanges are a the tip of a square profile and they’re teared apart by a bending force.
I can’t find a way to get a result. Do you have any tip? The errors span from Jacobian matrix errors ,when I try to simulate the contact between head bolts and plate with “Tie contact”, to non convergence when I simulate the contact with friction. Do you have any tips to try to get out of my condition?
Please see the attached model “ ANALISI_GIUNTO.pmx”
Negative jacobian with tie constraints means that you should disable Adjust for them. Convergence is much more tricky and can have many causes (although it usually originates from contact). To find them, it’s best to simplify the model (reduce its size e.g. to segment with one bolt and gradually disable some advanced features such as the 3 nonlinearities).
At the first glance, it seems that you could utilize symmetry in one plane. Also, you should improve the mesh - refine it for shells and use hexa or at least second-order tetra elements for the bolts. You will find many examples of similar models on the forum if you search for solid bolts. Such as this thread: COMPRESSION ONLY for simulating bolts and nut faces Or this one: Error during analysis of baseplate fixing
Displacement-based preloading is much better for convergence.
In the second step, where you apply the actual loading, you should freeze the preload (Displacement → Fixed).
Chek distance between hole center and plate edge. Seems to be too small (EN1993-1-8)-. For the stresses involved, 5 kN and about 1 kNm on the joint, the number of bolts seems to be exaggerated, as well as the thickness of the plates (15 mm) .
@Matej While testing this, I’ve noticed a small issue - deactivating steps with preload doesn’t deactivate the pretension section keyword (preloads within these steps have to be deactivated too):
I am a Mechanical Engineer from Brazil and I am using a simple bolt-and-support assembly to learn how to properly set up contacts in PrePoMax. I’ve been trying to get this to converge for 3 days without success.
As a new user, I cannot attach my .pmx file yet, but I am providing screenshots of my setup and the convergence monitor.
Model Setup:
Assembly: A simple support plate with bolts.
Contacts: Surface-to-surface with friction (Hard contact).
Solver: CalculiX.
Issue: The simulation fails in the first increments with “No Convergence” or “Zero Pivot” errors. The residual forces in the monitor stay very high and never drop.
What I have tried so far:
Swapping Master and Slave surfaces.
Enabling “Adjust: Yes” in the Contact Pair.
Reducing the initial time increment to 0.001.
Refining the mesh around the holes.
Could someone please provide a step-by-step guide on how to correctly set up contacts for this type of assembly? I want to make sure I am following the best practices from the start.
Please let me know if you need more details. Once I am allowed to attach files, I will upload the .pmx. Thank you in advance!
You should be able to attach files now. Please share the .pmx file. If it’s too large to fit directly on the forum, close the results before saving the file. If it’s still too large, you can use some hosting website.
