Hello,
i am new to PrePoMax but looked at various youtube tutorials regarding the topic bolt connection. I tried to setup my own litte model step-by-step like in the tutorials but it always says “no convergence”.
It’s basically two plates connected via screw. On the top of the bigger plate there is a fixed BC. The plates a connection together with a hard surface contact. The screw is splitted in to two halfs with boundary layer like in the tutorials.Am i missing something? (unfortunately i cant attach the .pmx file because it says a new user cant attach a file)
- Is the other plate unconstrained ? You may need to add some BCs to it. Using symmetry would also help remove some RBMs.
- Did you try with tied contact or tie constraints between the bolt head/nut and the plates ?
- Try reducing the contact stiffness. The default is usually too large.
- Disable Nlgeom if you have it enabled.
- Try with displacement controlled preload.
- If there are two steps, fix preload in the second one.
You should be able to do it now.
-
Is the other plate unconstrained ? You may need to add some BCs to it. Using symmetry would also help remove some RBMs.
→ my “real” usecase model is more complicated and cant be split with symmetry thats why i am not use it in this simple test. Yes the plate is unconstrained because on step2 (not yet in the model) i want to apply some forces not BCs. Add a additional BC on the unconstrained plate seems not realstic to me but i think i understand why it is necessary on step1. Just by setting the lower face of the plate to DOF12 = 0 nothing has changed (bolt direction is dof3).
-
Did you try with tied contact or tie constraints between the bolt head/nut and the plates ?
→ tie constraints
-
Try reducing the contact stiffness. The default is usually too large.
→ so i have to change from hard contact to Linear with reduced K?
-
Disable Nlgeom if you have it enabled.
→ Nlgeom is “false”
-
Try with displacement controlled preload.
→ did not solve the problem
-
If there are two steps, fix preload in the second one.
→ right now only one step is created
Infortunately I still cant upload a .pmx file because its to big (20mb)
Alternatively, you could also add soft springs as flexible supports to eliminate RBMS.
And it didn’t work either ?
Yes, you can determine it using the approach shown here: Snap-fit contact snagging problem - #17 by JuanP74 - Analysis issues - CalculiX (official versions are on www.calculix.de, the official GitHub repository is at https://github.com/Dhondtguido/CalculiX).
You could use some hosting service. I will send you instructions via pm.
oh my fault the results where still in the .pmx file. Now it should work.
Beispiele_Verschraubung.pmx (1.1 MB)
It seems that you’ve missed one step in this procedure:
You should create a compound out of the two bolt halves to make sure their meshes are connected.
You will see them separating if you run a frequency analysis:
Try with this fixed version:
Beispiele_Verschraubung mod.pmx (1.2 MB)
1 Like
wow you are right i messed up at the frist step, the geometry… Thanks a lot for your help und your tutorials.
