Corbel or Cantilevered beam 1

[slickdeals]

Folks,
This may seem to be a stupid question, but can anyone offer an explanation:

I have a stub column which is to be designed to carry a horizontal reaction. It has no axial load on it, meaning it is a cantilevered beam. The stub is 3'-0" high and 4'-0" wide and 2'6" deep.

My options are to design it as a cantilever beam and provide shear reinforcing parallel to the direction of force or to design it as a corbel and provide closed hoops perpendicular to the direction of shear. It may be cast at a different time, with rebar already doweled in during the first pour.

How do I decide which method to design it by?

 

[swivel63]

cantilever beam. how big is the load?

 

[youngstructural]

This is similar to the difference between a wall design or a column design... It's mostly engineering judgement, but should be based on what the behaviour should be like.

Given your shape described, this is certainly not shaped like a beam, it is a corbel.

I would use strut and tie.

Cheers,

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

 

[slickdeals]

shear Vu = 480 kips, ultimate.
Moment arm of 18"

 

[slickdeals]

/ /|
/ / |
/____/ / <--- direction of force
| | /
| |/
-------------

 

[swivel63]

ok. that's a big load.

strut and tie.

 

[kslee1000]

Don't know which code you are using. For ACI, you will have to meet certain geometric limits to call it a corbel. For your case, just design it as a cantilever beam.

 

[slickdeals]

I have a very small shear span-to-depth ratio. Of the order of 0.67 < 1.0.

 

[swivel63]

the shear stress you'd get by designing it as a cantilever beam would be monstrous, right? something on the order of like 500 psi or something like that?

 

[kslee1000]

Think this is an odd case of deep beam. You will require facial reinforcing in both directions to prevent crack from happening.

 

[slickdeals]

If designed as a cantilever beam, i would need 8 legs of #4 bars (Av = 1.6 in2) spaced at 6" o.c. and a flexural reinforcement of 7 in2. fc' = 4000 psi.

 

[kslee1000]

If feel necessary, you can bundle two legs to allow more room for concrete to pass through, and to match the tension bars.

 

[rapt]

It is a deep beam and should be designed by strut tie theory and also checked by flexural theory just like any other deep beam.

 

[kslee1000]

The beam can be classified as a wide beam, for which the nature of the load, uniform/concentrate, has significant influence on how the beam is going to behave. For concentrate load, the stress tends to spread across the width, with concentration under the load. For such case, strut-tie method seems superior without resorting to FEM analysis (for non-uniform stress distribution). For loaded uniformly along the width, in my opinon, either method (beam/struct-tie) could be used with confidence. Reinforcing details hold the key for success.

 

[rapt]

kslee,

There is a big difference between the results from beam theory and strut/tie for what is defined as a Deep Beam. As this fits the dimensions of a Deep beam according to most codes, I do not think that you should be making the recommendation that you have.

 

[kslee1000]

One has to judge the predominate behavior in design method selection. Both mathods satisfies the code, wasn't that current code pushing for strut-tie method?

 

[youngstructural]

Strut and tie always applies, regular flexural design ONLY applies if you have span/depth > 10 (St. Venant)... Otherwise your D region effects outweigh your B region effects and your flexural solution will be erroneous.

Once again: Strut and tie. I would love for anyone to give me a good reason to approach this differently. Always happy to learn, but pretty sure I'm right on this one.

Regards,

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

 

[kslee1000]

YS hit the main nerve.

Prior to strut-tie method, the ACI left the deep beam design largely in the hands of the engineers, except some provisions for shears. Lack of specific method, for bending, the use of conventional flexural method was widely spread, including me, though we knew well the beam theory really does not apply. Ever since the advance in computer software, I used to double check the design against computer model, to predict, or to match the stresses distribution generated.

I learned strut-tie method by self-learning while working on a Canadian project. I have never mastered it, but I appreciate its clear concept, and allowrance for creative thinking.

With one leg in each camp, I wouldn't reject either one in dealing with deep beams. But at times, one could be clearly superior than the other. The choice, again, purely depends on the anticipate member behaviors, and loading conditions. This is my ultimate opinion on this matter.