Menu:

FEA Analysis

Computer-based analysis is a natural partner for solid modeling CAD software. Steve Corboy can provide structural analysis (FEA) and motion analysis for parts and assemblies.

Steve Corboy uses ANSYS software for FEA analysis. Problems can include nonlinear materials, nonlinear geometry, and complex contact. Analysis types can include static analysis, linear and nonlinear buckling, mode shape determination, harmonic response analysis, transient response analysis, rotodynamic analysis and more. It's a very powerful tool.

FEA Philosophy

FEA is a very useful but dangerous tool. It's ridiculously easy to generate convincing-looking pretty pictures showing completely incorrect stresses and deflections of a structure. There's all sorts of common mistakes that can lead to incorrect results, including:
- Incorrect FEA model types - inputting of 3D models from CAD leads to a natural temptation to create models from solid "brick" elements directly from the CAD model, but this is not efficient or accurate for thin-wall structures.
- Incorrect choice of element type and mesh density.
- Applied constraints and loads that have no connection with reality.
- Overly optimistic failure modes, often related to compression loading - members that are assumed to fail either at Euler buckling stresses (easy to analyse, but highly unconservative for many structures), or worse still, material yield stress.

FEA should be thought of as just another tool a stress analyst will call upon when necessary. It's not a tool that is guarunteed to turn everyone into a stress analyst as long as they know which button to press. If you are paying anyone to conduct finite element analyses for you, amongst the questions you should ask are:
- What sort of FEA model are you planning to construct (ie beam, shell, solid), and why is it the right choice for this job?
- How will you determine the correct loads and constraints, and how will you apply them in a realistic manner?
- What failure modes will you be considering, how will you determine the allowable stresses for each mode, and what alignment is there to any national or international design standards?
- How will you know the FEA model is telling the truth?
- Is FEA the most appropriate way to analyse the structure?

Feel free to contact Steve Corboy. for more information.

Traditional Stress Analysis

Whatever did we do before FEA came along? Well, we analysed most structures using hand calculations, often with the aid of some very well refined empirical rules governing the stress of complex structures. For instance - it can still be difficult and inefficient to predict accurate buckling loads for some structures (ie thin-wall) using FEA. But there are empirical rules developed over long periods of time based on real-world testing that sometimes allow for more efficient designs than FEA will allow. Aircraft like the SR-71 were designed in remarkably short periods of time long before FEA was available to engineers.

Some of these traditional approaches also remain a powerful means to determine whether your pretty FEA model is actually telling the truth or not. Steve Corboy has a history of this sort of analysis in the aircraft industry.