Reinforced Concrete/FE Design Strips 1

[abusementpark]

A lot of the modern FE element software packages that design reinforced concrete floor systems rely on the "strip method". Here the user or program selects "design strips" which have a series of individual "design sections" over which the program integrates the stresses determined from the FE analysis to generate design forces for each section. I've used this first hand with both ADAPT and RAM Concept software packages.

One thing you quickly realize is that the results of your design is heavily predicated on how you select your design strips. Obviously, your design will become less stringent as you pick wider design strips. However, at what point are your strips too wide that you are missing critical behaviors in the slab? It sure seems like a lot is left to judgement. One thought is to try and pick design strips similar to what you would get using ACI's code methods. But that doesn't always apply in unique situations.

Has anyone comes any technical papers or articles on this topic? I'd like to know how some of you guys have been approaching it.

 

[rowingengineer]

The distance between zero shear is normally used at the reference point for edges of strips. The UK concrete society publication TR43 discusses this in detail for a PT slab. I would then divide this strip up as per the code of reference into middle and column strips.

If you are using FAE software I would suggest reading a few of the past discussions about the modelling effects.

ANY FOOL CAN DESIGN A STRUCTURE. IT TAKES AN ENGINEER TO DESIGN A CONNECTION.”

 

[rapt]

The strip widths nominated in codes (ACI, AS, BS) and all logical. There is no reason to look for any other. For complicated floor arrangements, it is still possible to apply those limits if you understand how slabs work.

Making a column strip too wide (or as ACI/PTI suggest for PT and using the full panel width) leads to grossly underestimated levels of stress and deflections.

You cannot check crack control on concrete unless you use the actual tension at the cross-section you are interested in. Averaging this stress over 6-8m thus reducing the peak enormously with the average results in a gross-underestimate of the real tension stress causing the cracking. Same for the effects of cracking on deflection!

 

[handex]

Abusementpark,
We have access to a different concrete fem package but typically don't use it for design. Generally we cover the slab in 1m to 3m wide strips and use the integrated forces from the output into a spreadsheet.

As a point of interest, what is the general workflow for the programs like ram concept meant to be? Do you guys just import a stripped down cad model, assign geometry and loads, input design strips and parameters and press run and design? What modifications are usually made from the output to when you chuck it on the drafties desk? If it's being used for complicated floors how do you check the results? By the way I'm not trying to have a go at the program etc I'm just generally interested.

 

[handex]

just bumping this thread for responses

 

[abusementpark]

As a point of interest, what is the general workflow for the programs like ram concept meant to be? Do you guys just import a stripped down cad model, assign geometry and loads, input design strips and parameters and press run and design?

That's essentially it. The other important piece is setting all the rebar information and parameters.

What modifications are usually made from the output to when you chuck it on the drafties desk?

It really depends on a case-by-case basis whether or not you think it is worth it to "penny-pinch". We usually like to chose a good base line for the program to check that will be typical rebar mat in a slab (i.e #5 at 12" o.c. typical) and go in and add a couple bars in the higher moment regions as required. The program generally adds in rebar only in dsicrete locations and along the length of the design strip that it is needed. The output is never "drafter-ready".

If it's being used for complicated floors how do you check the results?

It depends how much you want to check. You can always study the stress contours and reactions to see if they agree with your intuition. There are simple ways to check your input as well. Obviously, it is sometimes impossible to verify the precise load distribution of a complicated floor structure. Sometimes a big challenge is finding the proper way to different support conditions.