The Unwritten Code of Practice in FEA Analysis

Well before the days of finite element analysis (FEA), the days when the cowman roamed the range with his herds. Life was hard, cowmen didn’t take a bath, they just dusted off. They were bound by unwritten rules and codes of conduct based around loyalty, respect for others and fair play.

​“A man's got to have a code, a creed to live by, no matter his job”.
John Wayne

​Finite Element Analysis and all types stress analysis practice also have a code and creed to live by. There may be the temptation to present FEA results in a way that is misleading if it means getting the project signed off quicker. Some get away with this kind of behaviour for a while but all eventually get found out. If the design fails the design code requirements, it could be because there is too much conservatism in the assumptions but removing conservatism could mean more advanced analysis and more cost. It’s often, quicker and cheaper to change the design rather than run advanced analysis. Honesty and transparency are some of the most important attributes of the FE Analyst.

​“Do not go where the path may lead, go instead where there is no path and leave a trail.”
Ralph Waldo Emerson

​Just because your company has been doing things a certain way for a long time, this doesn’t mean it’s right or there’s not a better way. Finite Element Analysis can help you to push the boundaries, challenge the way it’s always been done and find ways to improve your product. Use FEA to explore new ideas, alternative designs and new ​materials.

​“Never approach a bull from the front, a horse from the rear, or a fool from any direction”.

​Write an analysis plan and make sure you will get the results you want before you start building FE Models. What are the project goals? When you get the stress results, what are you going to do with them? What type of stress is appropriate for your material? There is a temptation to build FE Models that are too large and complicated. Try keeping it simple and go for linear static analysis or even a hand calculation if you can. The approach you take right from the beginning will determine the success of the analysis.

​“Don't interfere with something that ain't botherin' you none”.
Roy Bean

​Be focused and tailor your analysis to answer specific questions. If your steel component only has a few hundred load cycles with a yield strength of 250 MPa and your FEA is telling you that the stress is 65 MPa, there is no point in chasing down mesh convergence to within 1% of convergence, 5% convergence is fine. Don’t spend time building a beautiful mesh on areas of the model that are only really used for their mass contribution. You can probably guess where the high stress is so use your engineering judgement for a more targeted approach.

“If you find yourself in a hole, stop digging”.
Will Rogers

​We’ve all been there, the ingenious plan to prove how good you are by building a FEM with multiple contacts, non-linear materials and all the bells and whistles has backfired. The model is so close to convergence that you can’t turn back now because there is too much at stake. As each day passes your project manager is starting to ask questions but your model is still not working. Knowing when to bail and try something different comes from experience.

​ Finally, and possibly the most useful quote;

​“Never squat with your spurs on”.
Will Rogers

Figure 1 - A Bending Moment

So long.