If you don't like the Strut-and-tie model 2

[Robbiee]

If I don’t have to use the Strut-and-tie model for analyzing the corbel in the attached sketch, can I design the corbel by:

1- Calculate bending moment about the C.L. of line 2 as V.eh+H.ev
2- Calculate shear at line 1,
3- provide main tension steel to resist the bending moment,
4- provide ties so that the shear friction capacity at line 1 is grater than the shear force,
5- shear friction capacity is only provided by the ties because shear plane is parallel to force or, do we have to account for the main tension steel also to resist shear?

 

[kslee1000]

Isn't the old simple mechanical method prescribed by ACI under CH.11 still valid? I guess it should.

 

[Robbiee]

kslee100,
Thanks. I am familiar with ACI. Here in Canada, the CSA A23.3 uses the strut and tie model in analyzing corbels.

 

[csd72]

It will behave as a strut and tie regardless. It is not that hard, I recommend you get used to it.

 

[Robbiee]

Sorry,
In my previous post I meant to say " I am NOT familiar with ACI". Thanks

 

[csd72]

Looks like you have to do strut and tie then.

 

[Robbiee]

Csd72:
When I am doing a strut and tie model for problems solved in reference books including our Concrete Design Handbook, my models are not the same as those in the books. Even using TEDDS, TEDDS' model is different from the books, the forces are sometimes 25% different. That is why I am looking for a (closed-form solution)that not two engineers will disagree on.

 

[kslee1000]

Ailmar:

In general, your approach is correct. If you can borrow a copy of ACI 318-xx (two digits indicating year it is in effect), "Building Code requirement for Structural Concrete and Commentary". Pay attention to Ch.11.9, "Special Provisions for Brackets and Corbels". From your standing, you will learn it in no time.

 

[kslee1000]

Note, the critical failure plane is line 2, for both tension and shear friction.

 

[hokie66]

I think I am right in saying that corbels in current practice are almost always designed by a strut and tie (or truss analogy) method, so I think you should take the time to learn it. Corbels are not flexural members.

Corbels have been the source of a lot of failures, and are one element where conservatism should be the norm. For the reinforcement, think anchorage, anchorage, anchorage.

In your sketch, the load is getting close to the edge. Corbels are usually loaded at about the centre. If the top corner cracks, there is not much to prevent catastrophic collapse.

 

[kslee1000]

Ailmar:

Hokie's right.

Canada has long before the US in adopting the strut-tie method in the concrete code (in the early 90s?). And it is there to stay.

I have limited exposure to this method, however, the concept is sound (through mechanical means rather purely relies on experimental studies), and I knew manys claim the freedom in constructing the truss model, and the simplicity of use after you have mastered the constraints.

Go for it, because you couldn't avoid.

 

[jrw501]

There are examples here:

http://cee.uiuc.edu/kuchma/strut_and_tie/STM/examples/corbel/corbel.htm

if you don't know how to get started.

 

[WpgKarl]

See the new CPCI Design Handbook - there is a design example for corbels using the Cdn code, both with the S+T model and using the empirical methods (used by PCI). The empirical method is quicker and easier to use and it steers you towards a safe design, whereas if you don't know what you are doing using S+T or design software, you could potentially come up with an unsafe design.

http://www.cpci.ca/?sc=publications&pn=fourth_edition

 

[hokie66]

There was a recent (several years ago) bridge failure in eastern Canada caused, at least in part, by poor design of the corbel supports. Worthwhile reading. It was discussed at some length on this site when the report came out. Suggest you search for it when time permits.