Hi everyone, I have a question that has been bothering me for a month.
In Abaqus Explicit, I used shell (S4R) to simulate a compression experiment. I chose 'general contact' with 'all* with self' for the contact settings. However, there is a distortion as shown in the picture starting at the step 0.
I have tried many solutions but none have worked, such as controlling hourglassing and changing mass scaling. It seems that the problem lies in the contact setting of 'all* with self', because if I manually add the surface pairs that will come into contact during compression, such distortions do not occur. But adding contact pairs one by one really takes time. How could I set up the contact while keeping All* with self?
I really appreciate if you have time to look at my issue!
If it is at step 0, the explicit solver hasn't been used yet. It can be related to your graphical user interface, your boundary conditions or an error in your mesh.
S4R elements are 2D with reduced integration to save calculation time. I'd check the dat-file that's sent to your solver for the coordinates of your nodes on such elements to find if it's mesh related of BC related. In that case get into more details.
If I look at the mesh it seems to S4R layers are close (connected?) to eachother. Is that your intension? Check the diference with the areas where it is going according to plan.
Martijn Jannink Hi Martijn, thanks for your reply! These layers(surfaces) are connected to each other. I think the problem lies in the "ALL* with self" algorithm. It detects some surfaces are already contact each other at step 0, so it creates some displacement on the surface to make the distortion. But I do not know what settings I should change...
Is indeed likely, but I don't know where to find the right UI buttons either.
You can leave the elements out of surface selection and see if the error changes.
If the other surfaces went according to plan, check difference in meshsize and node distances. I don't know which contact algorithm you are using, but some require master and slave sides where element sizes (master vs slave) are relevant.
- Badges
- Science topic
- Similar topics
- Psychology