I am trying to simulate a compression test which is displacement controlled. The project type is static non linear. Here I use a custom material i.e. AA6060 (aluminium alloy).
But while running this it shows the simulation failed.
Can anybody help me to find out the problem?
Sandwich structure- uniaxial compression | SimScale Workbench
I have a few suggestions.
#1 For me it seems like the core structure seems pretty coars for such a simulation, I would suggest increasing the number of elements on this structure.
For this you can use a regional refinemet or a local element size
#2 As seen in the result control the core structure is penetrating the top layer. Therefore a physical contact needs to be defined between the bottom face of the top layer and the wally of the core pyramids.
It’s a bit abstract, but this crash test attenuator case is simal to your setup.
please have a look at this case in order to see how to set up the physical contacts.
Thanks for your reply.
Actually, for meshing I mainly focus on the faces of interest to save my core hours. The top and bottom face plate have little importance for the analysis and I want no data from these two regions.
To avoid rigid body motion I set remote displacement as BC, made the z axis unconstrained and the upper plate as undeformable. Isn’t that appropriate in this case?
The top faces of the pyramid and the upper plate are already in bonded contact if I again add physical contact that detects conflicting contact issues. In this case what is the appropriate contact, please let me know.
Indeed, I am writing in details because I have only 4 simulations left for my current plan that’s why I need to be sure before a final approach.
A physical contact is defined by replacing the master-slave assignment to avoid conflict.
Mesh refinement is also done.
Sandwich structure- uniaxial compression | SimScale Workbench
Would have a look to this now to check if the set up is alright to start a new run?
Should I delete the remote displacement ?
Hello smohaimenul_is,
I was able to create a working case.
What I did is I chose the top surface as the master and the side surfaces of the pyramids as slave faces.
I kept the top faces of the pyramids as bonded contact.
I also refined the surfaces of the pyramid to improve the contact calculation.
This is the result I got.
And here is the mesh.
In regards to the number of simulation runs, if you are a student and this is a project for your university, please have a look at the academic plan here:
Hi again,
I was able to run the case but I could not hide the top plate to see my deformed structure. Is there any way to do that?
Sandwich structure- uniaxial compression | SimScale Workbench
Yes, by clicking on the eye icon next to the part name on the right-hand side panel, you can hide parts. For example, fig 43/44 in this tutorial.
Cheers
Thanks for your reply.
In fact, in my geometry, there are 10 different bodies but after simulating, in the solution fields there I can see only 8 bodies. That’s why I could not hide the top plate. Is there any way out?
It seems to work as expected to me - we have 10 volumes visible in the post-processor and I can hide the top volume as well.
Oh, now I got it! After hiding the ‘STEPS’ , all 10 bodies appear.
Thanks for assisting me.
