Recommended reading to become a better analysist? 6

[PerKr]

Apparently, with the last re-organization of our R&D department I was considered to be sort of our "expert" on FE analysis and mechanical strength calculations.
Yes, given the limited education I have regarding FEA, this is worrying.

The software we use for FE is Catia V5 and we currently have the GPS(2) and GAS(2) products (possibly the EST(2) as well) though it's not entirely unlikely that we could get access to further products (like FMD, FMS and GDY) as time goes by.

I am familiar with basic (mechanical) FE Analysis but want to learn more, to become a better analysist, and since a proper in-depth education is out of the question, self-studies is the only way.

So, two questions then:

1) What litterature would you recommend I have a look at?

2) Do you know of any litterature dealing specifically with the FE capabilities of Catia V5?

 

[GBor]

Check

http://www.nafems.com.

They have some free resources and some for purchase. I would suspect that they have something specifically dealing with CATIA. Then check the CATIA site for sample problems and tutorials. Learn each type of problem that you package can solve by performing their simplified models...then build up slowly. Then find some textbooks with examples worked out by hand. Get to where you can match the results within a few percent, and I would say you would be on your way to a successful FEA career.

By the way, the English word is 'Analyst'...you should probably start your career by learning to spell it correctly

 

[PerKr]

By the way, the English word is 'Analyst'...you should probably start your career by learning to spell it correctly

Haha, thanks. I thought "analysist" looked and sounded a bit odd after I had posted, just couldn't get my brain to connect Happens to me a lot it seems

 

[GBor]

If you are in a country where English isn't the first language, perhaps you can start a trend! "Analysist" - one who conducts "analysises". I kinda' like it!

 

[rstupplebeen]

Full course notes fome U of Colorado.

http://www.colorado.edu/engineering/CAS/courses.d/

Rob Stupplebeen

 

[SAITAETGrad]

90%+ of the materials out there are not going to make you a good analyst. While everyone needs a strong background in the nuts and bolts of the method, you are an analyst and not a programmer so the materials you study should be towards that end. For the analyst there is little out there but a few jems exist.

MacDonald, B. "Practical Stress Analysis with Finite Elements". This is a great primer on the subject. You should start simple and this book takes a basic approach to explaining many common principles analysts MUST know such as boundary conditions.

Cook, R. "finite element modeling for stress analysis". Another great book for the analyst, it will give you a much better understanding of the correct use for elements. What I like about this book is that it gives you enough theory to understand the elemnents and then explains the sources of error and common pitfalls. This is probably the best book out there on the subject specifically for the analyst.

Morris, A. "A Practical Guide to Reliable Finite Element Modeling". This book is about model validation, a key subject not often covered. Model validation is proving your model is representative of reality. Model validation should not be confused with model verification which is accounting for the errors introduced by mesh limitations (quad angles, etc. etc.).

AGARD LS 147. The lectures from Ian Taig in this series are invaluable. You should be able to find this document on the internet with ease.

There are other docuements such as "Rules for Modeling" a collection of notes that was circulating on this website a while back that was credited to Ian Taig and Boeing, Bombardier, and Airbus references that you may be able to find if you are resourceful.

 

[MechEng2005]

The course I took for BSME used "Fundamentals of Finite Element Analysis" by Hutton (From McGraw). It was good for theory and explaining what the "black box" is doing. You better remember linear algebra though. The problems in it are pretty simple, so they can be checked by hand. It's not so good for a practical reference.

That's the only "training" I have in FE, so I can't say how it compares to other books. My recommendation would be to check out local university bookstores, and maybe you can find a decent book used for cheap.

I would recommend getting both some references on the overall theory/application, and some program specific references. Each program works just a little different that to be the "expert" it seems your company wants you to be, you will want to know specifics of the program.

-- MechEng2005

 

[GregLocock]

First thing I'd do is go to a test lab and watch and help test parts - statics, fatigue and modal analysis.

The detailed mathematical gobbledygook is all fine and wonderful, but most FEA analysese don't fail to predict the real problem for mathematical reasons, they fail because the FEAer didn't understand the physical situation.



Cheers

Greg Locock

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

 

[jhardy1]

"Building Better products with Finite Element Analysis" by Adams & Askenazi - highly recommended, if you can get hold of a copy. It was published in 1999, and I think it has been out of print for some time:

http://www.amazon.co.uk/Building-Better-Products-Element-Analysis/dp/156690160X

It is a rare book in that it is actually about how to USE finite element analysis as a practical design tool, rather than concentrating on the theory of finite element analysis. (You may need to find another reference to brush up on the theory.) The examples are based on Pro/Mechanica, but the philosophy and methodology will work with the finite element software of your choice.

Hope this helps!

 

[Twoballcane]

I agree with Greg. I do believe that to be a good analyst you have to be good in hand calcs and testing. You have to be able to correlate to your FEA. FEA is GIGO, if you don't have anything else to back up your FEA study, I would consider it useless.

It would be a good idea to start out with some hand calcs to figure out what is going on physically. For static/structureal, I would calc out all of the bending and shear stress and eventually get to the Von Mises stresses. This always helps me truly understand all of the variables. Once done, you should have a good idea what the FEA should be. It would be even better to get some test data to compare with hand calcas and FEA data. In this way you have more confidence with your FEA to further engineer the product.


Tobalcane
"If you avoid failure, you also avoid success."

 

[Twoballcane]

Oh sorry forgot to add, I do a lot of heat transfer and vibration/shock stuff and my two go to books are Cooling Techniques for Electronic Equipment and Vibration Analysis for Electronic Equipment both by Dave Steinberg. These two books go more into practical than theory.



Tobalcane
"If you avoid failure, you also avoid success."