Torsion in concrete beam 2

[JStructsteel]

Back to my torsion issue from before, would torsion be calculated at the edge of support, or distance "d" from face of support? Im ACI-318-14. I am using PCI as a reference, and ACI 318-05 allowed it at "d" distance. I cannot find such reference in the ACI 14.

Anyone remember if its been removed?

Thanks
Jim

 

[le99]

I think these are the provisions you are looking for (no change from earlier editions):

9.4.4.2 For beams built integrally with supports, Tu at the support shall be permitted to be calculated at the face of support.

9.4.4.3 Sections between the face of support and a critical section located d from the face of support for nonprestressed beams or h/2 from the face of support for prestressed beams shall be permitted to be designed for Tu at that critical section unless a concentrated torsional moment occurs within this distance. In that case, the critical section shall be taken at the face of the support.

 

[hokie66]

I have always considered the torsion at the face of the support. Not sure of the intent of 9.4.4.3, as torsion is resisted by the support, and not at a distance out.

 

[le99]

@hokie66,

I think the argument is the same as the critical location for shear. This provision has been in ACI since 19xx(/).

 

[JStructsteel]

Thanks, I have done it at the face, seems more conservative. Not a huge d distance anyway.

 

[le99]

I concur. The provisions say you shall compute Tu at the face, then determined whether it can be ignored or reduced. If required, the segment between the face and d can be designed for the Tu at d, instead of the full value at the face, unless there is a concentrated torsion present.

 

[hokie66]

If you think about the difference between diagonal tension which is typically referred to as beam shear, as opposed to torsional shear, it makes sense to take torsion at the face of support, and beam shear where a stirrup will cross a crack. But then, the beam shear and the torsional shear stresses are additive, so where torsion is a factor, face of support it is, at least in my book.

 

[Lomarandil]

Sure, at the face is easy and conservative, but I understand the logic of the provision. A spiral torsion crack extends into the member similarly to a diagonal shear crack, and would be intercepted by a stirrup the same way. (Particularly for beams exceeding the threshold torsion and explicitly designed as such).

----
just call me Lo.

 

[KootK]

1) I agree that one can indeed do the "d from support" thing if desired, based on the same logic that we use for shear.

2) Again based on the same logic that one uses for shear, one needs to be cognizant of how the load is introduced to the beam. For some arrangements, "d from support" is appropriate; for others it is not.

3) In this particular instance, matters are complicated by the fact that only a portion of the beam will be involved in transmitting torsion to the support. This means that:

a) We've got some interesting decisions to make about how much of the cross section we want to claim is participating the torsion resistance and;

b) If the support connection works for torsion, odds are that the beam proper will be just fine too.

4) Because of the nature of the component here, this is something that really cannot be allowed to develop torsion cracks under service loads.