16 Jan 4 common mistakes in FEM and CFD analysis
In recent years, simulations techniques have gained more and more interests because of their capability of providing engineering information about products efficiency, resistance, reliability in a cheap and quick way. They allow every engineer or designer to simulate a virtual test lab without requirements of prototyping, but already in a CAD design phase. This great advantage comes with a price: are the results of our simulations comparable with real tests or they are just like a beautiful video game that produces great images and videos?
In this post we are investigating 5 common mistakes that one should avoid when defining this type of simulation. Let’s begin!
This is probably the most difficult task to achieve, despite its “simple” nature. CFD and FEM software have many “switches” which can be turned on and off by any user. Each of these add a physic behavior to the simulation. In CFD simulation, for instance, can turn on/off turbulence, combustion, rotation, heat exchange, radiation, phase change, chemical reaction, gravity, gas compressibility and so on and so for. What someone usually experience is that in any simulation only a limited number of these switches have an influence, whilst all the others have minor and negligible effects, and for these can be turned off without losing results significance. Actually: avoiding useless models allow us to reproduce more accurate results and use a lower amount of time!
Obviously the user experience is important in this phase, but at the same time it is a matter of common sense. If the simulation is about an impeller, we probably don’t need heat exchange or gravity, whilst rotation and turbulence play major roles. In case of force convection gravity plays a minor role and can be neglected. Inside a turbomachinery it is impossible to forget gas compressibility effect.
Mesh is a geometrical infrastructure any CFD/FEM software uses to make the calculations required. Even though most of the times it is seen as a waste of time, meshing is the fundamental part of any analysis. Verification of mesh quality after its creation is a fundamental step in order to obtain good results in a latter phase. What are the hints here? Quite simple: try imagine the solution you are looking for and make your mesh more refined in those areas where you expect high gradients (i.e. temperature variations, stress variation, …) ! What if you can’t imagine the solution? Well, make a first attempt mesh and check the results you are getting after a CFD/FEM run. Then make a second mesh using these results as an experience.
In deep: check out this post.
Steady simulations are the most common in CFD and even if the simulation is steady, CFD solution is an iterative process where, iteration after iteration, the residual (evaluation of the error) becomes smaller. Usually the whole process takes some hundreds iterations at least (thousands for big geometries). You can’t expect good results if your simulation runs for a lower amount of iterations. A practical hint? Check your residuals are below 10^-4 before stopping the simulation!
In deep: this blog post talks about residuals.
Most of you might ignore it, but in FEM simulation you have singularities. These are points where stress concentrates due to the geometry we are simulating (i.e. in sharp corner). A simple linear elastic material model return an infinite level of stress in these points, which is obviously an nonphysical behavior. In real materials, in fact, we have local plastic deformation that limits the highest value of stress reachable. For this reason the user running the simulation has the duty of detecting these points and see what he can do to simulate the correctly (modify the local geometry or use a more advanced type of material model).
CONSELF wants to make state of the art, cutting edge technology, available to every professional in the globe.
CFD and FEM simulation software is a very powerful tool, with its adoption optimization and innovation can be achieved in every field. To make this instrument accessible to everyone means lowering costs, but mainly to develop an infrastructure that favours a super-easy adoption by market new entrants.
CONSELF is highly committed and will pursue this goal working side by side with professional and industries to define the best strategies and solutions.
Ruggero Poletto – email@example.com