Beam shear allowance

[Chyclets]

When you design beams.. do you make the Vc larger than the service load to make sure the beam won't form any diagonal shear crack in spite of putting adequate stirrups? Note that the contribution to Vc comes from the dowel action, aggregate interlock and contributions from the uncracked concrete. From the Vc formula = 2 * sqrt (Fc) b d, let's say your fc is 4000 psi, b is 12", d is 19", then 2 *sqrt(4000)*12*19 = 28840 lbs. Do you make sure the service load of your beam won't exceed 28840 lbs so Fc won't be exceeded.. or do you not mind about it and just see to it the stirrup contribution from A(fy)d/s plus the FC exceeds the Vu? My concern is that whenever diagonal crack forms after exceeding Fc, the crack may not be predictable so is it better to make Fc (or beam) large enough to avoid any initial cracking from forming at all at service load? Or do you trust the stirrups to handle it (to prevent cracks from getting wider much like the longitudinal bars holding the tension crack?) What is the usual that you do?

 

[Lion06]

I think that is not practical in most cases. I've had to design many beams that go right up to 8*sqrt f'c, and 10*sqrt f'c for deep beams. It's not ideal, but sometimes it's what is required architecturally.

 

[JAE]

In all my past years of concrete beam designs I've never looked at service loads vs. phiVc. Never had a problem with cracks.

 

[Chyclets]

JAE, you never had problems with cracks because your actual load doesn't exceed the shear resistance contributed by concrete or Fcr? Or are you saying there are cracks but you never had problems with it? What are the appearances of diagonal cracks. Can this be seen with the naked eye outside the beam?

 

[hokie66]

Chyclets,
I believe you are overthinking this. Code provisions for beam shear are just conservative rules which have been found over time to be satisfactory, and these provisions should not be taken as an attempt to mimic actual structural behaviour. The rules just lead to a relatively simple design approach which agrees reasonably with testing which has been done.

 

[Chyclets]

What I was asking was, if your load exceeds 2 * sqrt (Fc) b d, the shear resistance taken by the concrete is gone and there is diagonal crack which is to be resisted by the stirrups. Since the total resistance is supposed to be Vcf (concrete) + Vs( Stirrup), a diagonal crack can make Vcf less, is it common for diagonal crack to form just like tension crack in moments that is resisted by the longitudinal bar?

 

[hokie66]

As I said, you are overthinking. Just because the load exceeds that arbitrary limit, that doesn't mean the concrete resistance to shear is gone. Your best advice is to just follow the rules, and don't worry about the actual mechanism of failure...unless you are doing research. To quote one respected text about the current method, "This approach has no rational justification, but on the other hand it leads to a relatively simple design procedure and provides a reasonable correlation with available test data..."

 

[bookowski]

Chyclets - As others have said it is not typical to do what you are suggesting. If you have a special case of an exposed beam or some other reason that you are particularly concerned about cracks it might be worth looking into, but for other cases you are going to end up with very large beams. Since Vu would usually be around 1.45 x Vservice you would end up with very little shear reinforcement and very large concrete area.

See attached photo of service cracks in a large beam. After a rain they really stand out - nice diagonals at the end and vertical at midspan. This might have been a case where someone should have thought more about service cracks.

 

[hokie66]

Even on that very long cantilevered beam which bookowski showed, the cracks do not go through the compression zone. That is why Vc is not the whole story. The concrete in compression is still carrying a lot of shear.

 

[Archie264]

>>>To quote one respected text about the current method, "This approach has no rational justification, but on the other hand it leads to a relatively simple design procedure and provides a reasonable correlation with available test data..."<<<

The above quote reminds me of a saying, "If it's stupid, but it works, it's not stupid."

I wish that philosophy was adopted more often, truth be told.

 

[a2mfk]

hokie66- I was wondering if I could email you directly about something not relevant to this forum, but I know you would have good information about? Not sure about the policy on that or how we exchange email.

 

[hokie66]

a2mfk,
I would be happy to help if I can, but the site doesn't like us to post our email addresses directly. I'll look into it, but one way might be if you can contact someone else who knows my contact details. Ron knows, so maybe you can contact him? Tell him you have my permission.

Go the Seminoles! Beat War Eagle!