torsion at ends of one way slab

[struct_eeyore]

Hello all,

I've got a condition where a fairly long - 15' span - one way slab that supported on each side by continuous concrete beams, with which it is mono, and which are supported on columns ever 12-15' - detail attached. I've tried to detail the connection to achieve fixity at slab ends, but am wondering if simply due to the span and deflection of the slab itself, the perimeter beams would induce torsion into the support columns. Anyone have any insight before I spend my weekend digging through ACI to no avail?

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1708653043/tips/PLAN_SECTIONS_MOD2-Model_jqhbb2.pdf[/url]

 

[Tomfh]

Yes in reality there will be frame action going into the columns, some of which will go via torsion through the beams. This is one of those "compatibility torsion" situations.

 

[ChiEngr]

You wouldn't induce torsion into the columns, correct? The beams would be subject to torsion and would consequently induce bending moment into the columns. I think you can check the beams to ensure that the threshold torsion is not exceeded, which would result in you not having to design the beams for any torsion. However, I would check the columns for any induced moments as described above.

 

[BAretired]

A one way cast-in-place slab spanning 15'-0" should have a thickness of L/20 if simply supported, L/24 with one end continuous and L/28 with both ends continuous, where L is the clear span. This 6" thick slab is restrained at each end, but not fully fixed, so excessive deflection may be a problem.

I've tried to detail the connection to achieve fixity at slab ends, but am wondering if simply due to the span and deflection of the slab itself, the perimeter beams would induce torsion into the support columns.

There will not be full fixity at the slab ends, but the slab will induce torsion into the beams, which in turn will induce bending into the columns. Where else is the fixity you strived to achieve going to go?

Perhaps you should consider either a thicker slab or a two way slab, particularly with the 1.5" cover that you have specified top and bottom. Alternatively, you could use concrete joists or perhaps precast hollow core slabs.

 

[struct_eeyore]

BA,

The slab is doubly reinforced w/ #5@6 T&B - I'm getting a long term deflection of ~ L/264. (I did increase the depth by another 1/2" to clear 240). Unfortunately, the architectural requirements are such that I have to do CIP, and w/o any intermittent joists.

You don't think that confining the slab bar hooks within the beam would create sufficient fixity to extend the 'pin' location to over column centerline?

 

[BAretired]

You don't think that confining the slab bar hooks within the beam would create sufficient fixity to extend the 'pin' location to over column centerline?

I have no idea what you mean by this. I have not suggested that the slab bar hooks within the beam would create insufficient fixity, but even if we assume the bars generate full fixity, that moment must be transferred to the columns in bending. If the beams are pin connected to the columns, the slab is a simple span with a positive moment of wl²/8 because the beams will simply rotate to accommodate the end slope of the slab.

We seem to be having a serious communication problem.

 

[struct_eeyore]

BA,

Let's see if I can wrangle the language here a bit better -

By developing full fixity across the beam/to slab connection, I can effectively treat the edge beam as part of my slab - i.e. I don't calculate torsion that would otherwise be generated at the beam to slab connection if we assumed it was pinned (~4" from the centerline of the beam for an 8" W beam). Instead the only torsion is the compatibility torsion that results from the slab trying to deflect, and bow the column along with it.

The column, i probably forgot to mention, are concrete, so they will act fixed to the slab edge beam.

 

[BAretired]

Okay, that is pretty much as I expected. You wondered in your original post "but am wondering if simply due to the span and deflection of the slab itself, the perimeter beams would induce torsion into the support columns".

We now agree that the perimeter beams induce bending in the columns, not torsion.