finite elements model that reflect the reality 1

[SofieneFEA]

Hi every body,

I’m a new in finite elements analysis method and I have some questions:

- What are the best procedures that must be used to obtain a finite elements model that reflect the reality?
- How we can trust to finite elements softwares?
- Does it exist standards and specifications to guide engineers to do the best mesh that can insure the more accurate results?

 

[rb1957]

1) be very. very careful.
2) gold i trust, but FEMs ... don't turn your back on them or they'll bite you. we don't "trust" the FEA software, we do a bunch of test cases to build confidence the it is lying (by commission or ommission).
3) MSC has published some good QA standards that will detect real trouble with your models (sorry, i don't have a link), but it's up to you to understand the reality of your problem and to decide whether your FEA is following this.

 

[GBor]

Good response, rb!

SofieneFEA, you may want to look at NAFEMS (

http://www.nafems.org).

They sell standards and benchmarks and fund a number of studies of different software packages.

Some of the information that you want to read about is mesh sensitivity analyses (seeing if a refinement in your mesh produces a significant change to your results), but above all, experience will help a great deal. Compare some simple results to textbook questions and hand calculations. Practice using different material models, element types, and boundary conditions.

Get inside the "black box" of your software

Understand how your software is calculating various aspects of your model and learn its limitations...they all have them. Most of them handle linear, static problems just fine and if that is all you are doing, your job is simple. More complicated analyses, make sure you understand the physics behind the analysis screen.

Garland

Garland E. Borowski, PE
Borowski Engineering & Analytical Services, Inc.
Lower Alabama SolidWorks Users Group

 

[sdm919]

You need a good understanding of structural behavior to build a good model. If you don’t know the differences between trusses, beams, plates, shells, etc, how can you choose the right elements for your structure? I guess you get this from college classes and years of experience.

You also need to know something about the theory of finite elements. If you want an overview without too much math, consider the following books:

Finite Element Modeling for Stress Analysis by Cook
Building Better Products with Finite Element Analysis by Adams & Askenazi

To gain confidence in the software you’re using, run simple test cases and try to reproduce expected results. Most software comes with example problems, or better yet, devise simplified cases relevant to your problem. If you’re dealing with plates, get a book on plate theory, make a simple model of a square plate with uniform pressure, and see if you get the textbook solution. If not, don’t give up until you find out why. You'll learn a lot from this process.

Also, use the simplest possible model that adequately represents the behavior of the actual structure. It’s easy to make an overly complicated model, but it will be more difficult to de-bug. Try to get in the ball park with a simpler model, and add complexity only if necessary. And you can’t put too much faith in the answers if you don’t know the input data very accurately. You might have reliable material properties, but how well do you really know the loads? Seemingly small details can have big effects.

Besides NAFEMS, here are a few web sites that might be useful. The last one contains course notes for a college FEA class.

http://femci.gsfc.nasa.gov/index.html

http://homepage.usask.ca/~ijm451/finite/fe_resources/fe_resources.html

http://www.colorado.edu/engineering/CAS/courses.d/IFEM.d/Home.html

 

[craigmj]

1. FEA will always, without question give you the exact answer to the question you ask it. It is the question that needs to be understood the most. If you dont understand that then you are out of luck no matter what element type or material model you choose. In other words you have to know what "reality" is before you can know how well your model describes it. Keep in mind that FEA is just a tool, and the use of this tool is as much an art as it is science...maybe more.

2. You can only trust your FEA model as much as you trust your assumptions. Experience is the only way to trust your assumptions.

3. FEA is not paint by numbers, and its downright dangerous to look at it that way. There are many different phenomenon to simulate and they all can require that you change the rules as you go. That said, there are general guidlines to keep in mind as outlined by countless textbooks and users manuals.

The bible of general FEA (numerically speaking): Finite Element Procedures, Prentice Hall, 1996; Klaus-Jürgen Bathe.

CJ

 

[GregLocock]

- What are the best procedures that must be used to obtain a finite elements model that reflect the reality?
- How we can trust to finite elements softwares?
- Does it exist standards and specifications to guide engineers to do the best mesh that can insure the more accurate results?

Last bit first. I don't know the exact proportions here, but my guess is that most FE analysis' fail due to poor understanding of the boundary conditions, loads, and mechanisms, rather than meshing errors as such. I'll include abuse of rigid elements as a boundary condition/mechanism problem.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[jagad5]

No one mentioned material variability in comparing analysis to reality. The Young's modulus can vary with +/- 3-5%. The ultimate and yield strength can exhibit standard deviations of 5-10%, which means that strong samples could be almost twice the strength of weak ones. Crack growth rates vary for titanium vary by a factor of 10 from specimens cut from the same forging.

This scatter is especially important if your test is a fatigue test. Even in the simple case of repeatedly applying one stress, the range between the 1st percentile and the 99th percentile part can be several orders of magnitude. As a specific example, the range in expected fatigue lives for C355-T6 cast aluminum is a factor of 1000. That is, your strongest casting might last 1000x as long as the weakest one.


Doug

 

[GregLocock]

And then the measured loads for an identical event can vary by 20%, if the event is (for example) driving over a specified stretch of rough road at a constant speed.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.