Additional Longitudinal Steel for Torsion- Precast Beams 1

[Strucyou]

Hi,
I have been thinking about the additional longitudinal reinforcement required for torsion specifically for a precast beam. Precast beams are often simply supported and resist eccentric loads owing to wide use of inverted L shaped or T shapes with unequal loads. The resistance to torsion is provided by steel plates, anchor bars provided in the dry joints to support the beam. ACI recommends that the torsional bars should be developed inside the support by a defined length. How does this apply for a precast beam resting on a dry joint, lets say a corbel?

 

[dik]

I would try to avoid torsion by taking the loads out that are causing it, by some other means. Designing for real torsional loads increases reinforcing significantly and precast members generally have reduced cross-sectons. Just my $.02.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[KootK]

Torsion consideration in these systems is a bit of a sketchy nightmare in my opinion. Consider:

1) You've usually got unbalanced loads in the erection case.

2) You've usually got unbalanced loads in the permanent case.

3) As you say, it's very difficult to develop the mild, longitudinal torsion bars sufficiently over top of the bearing without resorting to end plates welded to the bars etc.

4) In theory you can use the pre-stressing for this purpose but:

a) There is often a fabricator preference to not have prestressing at the top corners of beams.

b) prestressing strand tends to "develop" even more slowly than mild bars. Granted, you've got the hoyer effect etc.

5) As you've surely noticed, trying to carry torsion into precast supports tends to not work particularly we. Moreover, who the heck knows how one is supposed to incorporate torsion in to a corbel design, particularly if there is net uplift.

6) In my experience, precast engineers are doing this for the most part:

a) Shoring thing to eliminate torsion during the temporary erection case.

b) Finding some way to claim that the beam torsion is coupled to adjacent plank flexure. This, essentially, is using the plank to brace the beams against torsion similar to how we use torsional bracing in steel sometimes. I find that this tends to get a bit dubious at times given:

i) the nature of common precast connections between plank and beam.

ii) the lack of parity, at times, between the scale of the beams and the scale of the plank. Can a bunch of loosely connected, 8" plank really restrain torsion in a 40"W X 48" prestressed beam? Me not so sure.

 

[Strucyou]

Thank you for insights. Bracing action of planks is an important point but needs further research to establish exactly how much it may contribute in the compatibility with the beam under torsion. I mean it will be affected by the span lengths, presence of cast in situ structural topping or as you rightly pointed out the differences in cross sections. Cold joints on the other hand facilitates the development of torsional perimeter reinforcement but even that doesn't come without limitations in precast. The erection of precast beams under torsion connected with cold joints can be challenging at locations where two or more precast beams intersect due to the clashes in protruding steel. Mechanical couplers maybe a way to avoid that but it extensively increases the work on site. Due to this reason, I would still be interested in how to account for the coupling action of planks to capture the bracing action accurately.

 

[KootK]

Bracing action of planks is an important point but needs further research to establish exactly how much it may contribute in the compatibility with the beam under torsion. I mean it will be affected by the span lengths, presence of cast in situ structural topping or as you rightly pointed out the differences in cross sections.

I agree but, in North America at least, the precast industry has basically moved forward with accepting the solution even without the testing.