As sophisticated modeling technology has become available and widely adopted, the demand for innovative design is growing, leading the industry away from box structures and putting pressure on engineers to carry out more complex designs. Most engineers today face complex design challenges involving transfer girders, hanging columns, cantilevers, and sloped members. Column layouts supporting structural slabs are no longer uniform; increasingly irregular layouts and loading demand more unique and detailed design.
The need for flexibility in design makes the case for finite element analysis (FEA) over the traditional strip method. FEA can help solve challenging structural analysis problems, yet adoption has been slow. Why?
With the rise in more complex, irregular structures, the strip method falls apart. Instead of reverting back to the strip method, engineers should focus on the results of the analysis and look at where forces are actually occurring in the slab and then only placing the reinforcing in the slab where it is required.
Rethinking strip design methods
First, there is nothing wrong with strip design methods. With a regular slab, no significant openings, and regular loading, the traditional approach of strip design could even prove to be slightly more economic than FE design (although slower in design time).
Difficulties arise when layout and loading are not perfect. In this case, the engineer makes judgements to come up with idealizations that will result in a safe design. Inevitably these will err on the conservative side. The challenge becomes keeping them safe without being overly conservative and potentially wasteful. This approach moves analysis and design toward an art rather than a science, leaving room for interpretation and creating an opportunity to get the design seriously wrong.
On the converse, using FEA breaks those constraints. Design is faster and more flexible while simultaneously accommodating irregular slab geometry, significant openings, and irregular loading. Reinforcement can be efficiently placed so that extra bars are only provided where they are required.
With FE, analysis and design become more of a scientific process with greater certainty and fewer errors.
Things to consider
Pragmatic design based on FEA still requires consideration of averaged design moments in localized strips. These can also be considered in the form of rectangular patches with associated reinforcement design. Problems can arise when trying too hard to extend averaging requirements in local strips and patches to exactly match the design strips considered in the traditional idealization.
The traditional strip method results should not be viewed as “correct” or totally accurate. Instead, they are simply acceptable and safe design forces in the regular situations where the method is applicable. This becomes quite apparent when comparing the similarities and differences between traditional approaches used in different countries working to different codes; all codes embrace a traditional strip method but there is quite significant variation in the proportioning of forces to the strips.
In an FE design, it’s important to embrace the concept behind the traditional strip idealizations without getting drawn into the need to replicate the method exactly. You’ll need to accept that averaging over slightly different strip widths is perfectly valid and will be safe, as long as you avoid averaging over greater widths than would have previously been assumed acceptable. So, the “art” that remains is using averaging to a reasonable degree. In addition, you don’t have to average everywhere, just where needed. Basically, keep it simple when possible.
You’ve run FEA, why not use that to place the reinforcing instead of being forced into the overly structured and rigid strip method?
Information provided by Trimble (www.trimble.com).