The ability to run simulations for projects allows designers and engineers to verify assumptions, validate approaches, and identify the most cost-effective – and optimized - solutions before finalizing a design and sending it to production.
Not to mention, simulation helps companies meet cost targets, stay on budget, and meet tight deadlines.
With the aid of well-designed simulations, manufacturers can expect to produce fewer prototypes, create more innovative design solutions, shorten the physical testing phase, reduce time-to-market, and encounter less rework.
Essential Elements of an Effective Simulation
To get the most out of a design simulation, follow these seven best practices to make sure you’re well positioned to render the most accurate and relevant results:
1. Assemble a Well-Trained Team
Accurate simulation starts with a well-trained team of designers and engineers who understand how to best leverage available data and technologies to optimize results.
To ensure that everyone has the right skillset, make training available to the right people in the company.
Establish management support by communicating the value of simulation, so not only will leadership make simulation a priority but also allocate the resources required for the proper education of your team.
2. Start With a Simple Model
Identify the key assumptions and start with a simple model to test those factors.
You can also start with a pilot project that has a smaller scope and shorter timeline.
You can always add complexity later on to bring the model closer to a real-life scenario after you have identified the key components necessary to help you achieve the desired outcomes.
3. Conduct Simple Analysis To Prove the Model
After you have set up a model, conduct simple analyses to validate assumptions and verify your design approach.
Analyses should be done early and often. The most successful companies have design engineers who use simulation during the concept/design phase to explore alternatives and hone in on the most effective solutions.
4. Perform Mesh Convergence Studies
Finite Element Analysis involves building a simplified model that represents a real world object. These models are constructed of building blocks known as elements. The system of elements is known as a mesh. In the past, and to some extent today, these meshes were manually designed by an experienced analyst. More commonly today, analysis software includes a meshing engine that generates a mesh automatically, typically from a 3D CAD model.
Without any user input at all, these engines can generate a mesh that will run successfully and provide results. The results generated using this “default” mesh should be considered a first step, and the results preliminary. These preliminary results will likely change as the mesh is refined. Refining a mesh usually means making the elements of the mesh smaller, either globally or in key areas. Think of it as adding more pixels to an image, the picture gets clearer.
This refinement comes at a cost. A finer mesh takes longer to run. This is why the “default” mesh is a good starting point. It’s a rough estimate, but can run fast and give you immediate feedback about your model. Not every model or design candidate is worth the extra time and money that it takes to run the finer meshes. As the design and your model evolves, more time should be invested into a better mesh.
You’ll know when to stop refining, because the results of the analysis stop changing dramatically when the mesh is refined. The results of each mesh refinement “converge” toward a final answer. A model where the results are no longer influenced by the mesh is said to be mesh independent. The process of seeking this point is known as a mesh convergence study, and is a fundamental part of any simulation analysis.
5. Verify Your Data and Results
Simulation results are only as good as the data you enter into a software application.
During the early stages of a project it’s rare to know everything about a real world scenario, and, therefore, some inputs may have to be assumed. If the limits of the assumptions are understood then the simulation performed will still have tremendous value.
To ensure you’re getting the most accurate results, start with high-quality data. Double check that you have all the units set up correctly. For instance, are the units used in the imported CAD data matching the units you’re using in the simulation?
Then you need to make sure you’re displaying the desired result type.
If you’re also running a physical test, you can correlate the results to see if they’re in the ballpark. If not, you may need to look at the assumptions and adjust.
Note: This is the most important part. Any other information you have should be put to use to confirm you’re “in the ballpark” such as physical test results, calculations, similar designs, common sense, and experience.
Be suspicious of your own results.
6. Peer Reviews
Having a second pair of eyes to review the simulation model can help avoid careless mistakes. Develop a checklist and workflow to help reviewers make sure they have covered all the essential elements.
Including peer reviews as a standard component can help improve the accuracy of the simulation. After all, every mistake found and fixed is one that won’t go into the final design, enter production, and cause a flaw in the final product.
In addition, managers need to foster a positive peer review culture in which finding mistakes is considered a means of learning, growing, and communicating.
7. Outsource the Initial Baseline Simulation
There are many moving parts when it comes to setting up an accurate simulation model. In addition, the ability to identify and use historical product knowledge can enhance the quality of your designs.
You can engage a simulation engineering service to set up an initial baseline simulation, which can improve the accuracy of all subsequent simulations. This allows you to eliminate expensive prototypes, avoid field failure, and reduce excess materials usage.
Learn more about Sim Specialists, our simulation engineering service that offers a complete solution for all your simulation needs.