Written by Megan Jenkins on May 23, 2019
March 28th, 2019
approx reading time
Only 30 days have passed since I joined the SimScale team. As a CAE sales engineer, my primary focus was to learn how to use the SimScale simulation platform with all of its functionalities. It seemed like a long way to go back then, but now I’m starting to feel like I’m on my way to becoming an expert. So here are some tips on how to learn SimScale in 30 days:
My learning process started by completing the tutorials available on the website. I started with the basics: Stress Analysis of a Connecting Rod. This helped me learn how to set up a simulation on SimScale and familiarize myself with the user interface and capabilities of this 3D simulation software. Following every step, I successfully completed Tutorial 1.
If you are new to the SimScale platform like I was, you can always search through the tutorials available on the SimScale website by going to the Resources tab and clicking on Documentation and then Tutorials. It will bring you here.
After reading the documentation and completing all of the tutorials, it was time to attempt my first simulation.
I started by uploading the CAD geometry: a piston-crank assembly. The file was a .STEP file that contained three solid bodies: the piston, the connecting rod, and the crankshaft.
When I meshed the geometry, things became more complicated. I started with the default mesh settings using the type: fully automatic tetrahedralization, a mesh order of 1, a fineness of 2 – Coarse, and 1 core for processing.
What followed was an error message:
So, I checked the mesh operations event log:
The three possible issues that needed to be considered were:
Since the CAD geometry was uploaded as a .STEP file (the recommended file type), I was able to rule out that issue.
Next, I tried adjusting the meshing parameters to generate a finer mesh using both fully automatic tetrahedralization and parameterized tetrahedralization types. However, each iteration resulted in the same error meshing.
That meant that the problem was with my CAD geometry. It appeared to be clean, but upon closer inspection, there were overlapping surfaces on the pins’ contacts at both ends of the connecting rod.
After removing these extraneous surfaces, the CAD geometry meshed successfully, and I was able to move forward with the simulation.
In my third week, I learned how important the quality of the mesh is for a successful simulation, as it has a great influence on the solver convergence and final simulation results.
A common misconception when generating a mesh is that the finer the mesh, the better the results. While finer meshes can give a higher accuracy, they are also more expensive, requiring increased memory and CPU time. Thus, the best mesh is one that has been constructed with an understanding of the physics of the problem in mind and can accurately capture the physical behavior of the model in a computationally efficient manner.
Continuing with the piston-crank assembly, a coarse mesh and a fine mesh were generated using the platform’s ‘fully automatic tetrahedralization’ mesh operation type.
The coarse mesh is sufficient for the volumes, but it is too coarse near the contact regions. The fine mesh has the necessary detail in these contact regions but it is unnecessarily fine on the volumes. Knowing this, I tried to improve it by making manual refinements.
As a general rule, a coarse mesh can be used in regions of low stress and strain. Near corners, concentrated loads or stresses, areas of contact, and in locations with an abrupt change in material properties, a finer discretization is required to fully capture the physical behavior.
There is a mesh operation type called ‘tetrahedralization with refinements’ that allows the user to add manual refinement zones. Maintaining a coarser mesh on the volumes while refining the mesh size at the contact zones produced the following result:
This refined mesh integrates the best qualities of the previous coarse and fine meshes.
The simulation run was successful and therefore I could produce an animation in the post-processing software called ParaView.
For my latest simulation, I was inspired by one of my favorite hobbies: rock climbing. Carabiners are an essential piece of equipment for the sport and I thought it would be interesting to make a static analysis simulation.
An important topic that I have not discussed yet is material plasticity. For solid mechanics, the default setting for material behavior is ‘linear elastic.’ The linear elastic assumption is valid when there are small deformations (strains). When stresses are high enough to induce yielding in the material, there is a transition from linear elastic behavior to plastic behavior due to non-reversible deformations in the material. The stress-strain plot below shows this principle:
A rock climbing carabiner could be subject to high loads and large deformations. Thus, plastic material behavior should be considered in the simulation.
Stress-strain curves are widely available online and are typically presented in graphical form. For aluminum 6061-T6 (red), the stress-strain curve is as follows:
On the SimScale platform, the values from the material stress-strain curve must be extracted and entered into a spreadsheet (.csv file) with the strain values in column 1 and corresponding stress values in column 2.
Things to keep in mind:
In the simulation designer, you can define materials and assign volumes to them. Many commonly used materials are already predefined and can be imported from the material library (available at the bottom of the Material’s window).
When the material behavior is set to ‘Plastic’, the option to ‘Define the value via a table’ will appear. Under the ‘Source’ parameter, you can upload a spreadsheet and define the desired stress-strain curve for the defined material.
The plastic material behavior has been fully defined and you can continue with the rest of the simulation.
My first 30 days at SimScale have come and gone. It has been a busy time and I feel that I have learned a lot.
I really hope that my tips will help you guys get familiarized with the platform. To make sure I give you a starting point, here’s a short list that was of great help for me:
I hope this article helped has given you a better idea of how to learn SimScale. Good luck with your projects and I’m really excited to see them in the Public Projects Library. As for me, now I feel ready to face more complex simulations!
To discover all the simulation features provided by SimScale, download the document below.
Subscribe to the SimScale Blog
Written by Megan Jenkins on May 21, 2019
This computational wind engineering guide covers types of wind analysis, how to improve your designs, and more wind resources...