I wanted to propose the possibility that the software creates coincident nodes between surfaces without having to create a “compound part” with the “split compound mesh” property (see my discussion Bolts shear contact not working correctly).
Currently, the software still performs the analysis, but provides incorrect results.
In my opinion, this may be dangerous, especially considering that the procedure I used (incorrectly) is the classic one used with similar software (such as Ansys, and presumably Abaqus).
It seems that Prepomax already has all the necessary tools, so I hope this implementation isn’t too challenging.
1 Like
This is not possible to achieve without the compound in the background. The problem with compressing this 2 step process into a one step process is that creating a compound is not always successful and can also create additional parts/regions when the geometries overlap.
How would the user prescribe the condition for a coincident mesh?
I don’t understand where the problem was. You forgot to split the mesh?
Hello Matej,
No, I didn’t even create the “compound part”. “Compound part” is something new to me, I applied the contact condition on the surface of the single parts imported directly in Prepomax like I do in other softwares.
In online Prepomax videos where there are contact surfaces, they do not use this procedure that involves using “compound parts” with “split mesh.” The calculation is performed as I did initially (without compound parts).
I’ll be honest, this still puzzling me, but with time I’ll find when is necessary this procedure and when not.
My bad. I didn’t think the procedure could create problems.
I take advantage of the message to compliment you for the software.
You did on your own what EDF couldn’t do with Salome (a very powerful tool, but too complicated and time-consuming to use for everyday use).
Have a good day
Aha, I understand. This process is not necessary. The problem is the solver. CalculX is obviously more sensitive to the shape of the contacting surfaces. Creating a compound part only produces a coincident mesh, making it easier for the contact algorithm to get a smooth contact stress field. You could get the same results using un-coincident mesh but with a much smaller element size.
Thank you. 
I really appreciate it when users find the PrePoMax useful.