Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Finite Element vs. Strip Design

[Subject Prev][Subject Next][Thread Prev][Thread Next]
My comments on finite element vs.strip design for post-tensioning design were not based on a statistically accurate survey but I did talk to a fair number of individuals, including engineers at different levels. I want to write an article on code changes, and I needed to know what to assume people are using for design.

I don't have any particular bias;  I have both types of program.   I like doing finite element analysis because the contour plots are lots of fun,  but for most designs it takes me a lot more time to do a finite element model than to do a strip design.  I can do a standard 60-20-60 one-way slab and beam parking garage design in about a day with a strip design program.

As a general note,  finite element analysis is very precise;  concrete construction is not.   I know this for a fact because I used to work for a concrete contractor and while I was supposed to be a project manager, if we had a Saturday pour and too many people were still in jail from Friday night bar fights, I ended up shoveling concrete with everyone else.  As a result, I saw things I have for some reason never seen while working as an engineer.

Concrete is also a non-homogenous, inelastic material that is difficult to model precisely in a software program without making a lot of assumptions and  simplifications, many of which are "transparent" to the software user.  The precision of finite element analysis does not tranfer to the design of concrete structures.

In most design firms,  the newer graduates are most adept at finite element analysis, and thus they become the office gurus.   Most of them have very little design experience, some of them have never been on a construction site, and few of them have read Chapter 18 of ACI 318.  Because they have no way to judge whether a result is reasonable, the finite element analysis program lets them make alot more mistakes a lot faster.

On a related issue, the 2005 version of ACI 318 will take the limit on tensile stresses for two-way post-tensioned slabs back down to 6 sqrt(f'c).  The increase to 7.5 sqrt(f'c) in the 2002 was an oversight; an unfortunate manifestation of the fact that post-tensioning is really not well addressed in 318.

Gail Kelley