Breakout of I-Beam Cast in Concrete 1

[EngDM]

Has anyone here ever analyzed a steel I-Beam cast in concrete for breakout? Some ideas I have are to add nelson studs to the web of the beam, or add weldable rebar to the sides/bottom, but I have never actually seen any calcs or procedures for this. I've got some pretty high tensile loads to resolve. Just wondering if there are some "typical" details or simple details that have been developped that I could look into.

Top of concrete is top of top flange, equipment is then welded to the top flange.

 

[KootK]

Does this need to be a moment connection?

I feel as though the "typical" would be to weld the beam to a cast in embed plate rather than embedding the beam. What prevents this?

 

[KootK]

The closest analog that I can think of for embedment would be steel coupling beams from the world of shear walls: Steel Coupling Beams

Top of beam at top of wall just makes a difficult thing more difficult I'm afraid.

 

[AskTooMuch]

We often cast in steel beams or columns in concrete as fireproof. I have never seen anyone consider any breakout analysis, everyone just use standard concrete fireproof detail.

 

[EngDM]

Does this need to be a moment connection?

I feel as though the "typical" would be to weld the beam to a cast in embed plate rather than embedding the beam. What prevents this?

The beams are specified by the equipment manufacturer, it is how they connect their equipment to the slab on grade because the loads transfered into the beams are quite large. It's my responsibility to design the slab however, and as such the breakout/connection to concrete.

 

[KootK]

Ok, so the beams are actually posts/columns?

How thick of a slab on grade are we dealing with?

Biaxial shear and bending?

The SOG is cast around the posts?

Is there a local SOG thickening or anything like that?

 

[KootK]

Or is this mostly an axial load thing?

 

[EngDM]

Ok, so the beams are actually posts/columns?

How thick of a slab on grade are we dealing with?

Biaxial shear and bending?

The SOG is cast around the posts?

Is there a local SOG thickening or anything like that?

The top of beam runs parallel to top of slab, almost like a rail would. Their equipment connects to my beam with a moment load, and that moment load needs to transfer into the slab below. The moment is an overturning moment, which would presumably rotate about the bottom flange toe.

 

[KootK]

So like this?

Are you allowed to have stiffeners etc in the beam?

 

[EngDM]

Yes like that. I assume I could do stiffeners, but I believe that would be up to the supplier to specify. I'm wondering if nelson studs off the bottom instead of the sides are just better for this.

 

[KootK]

How thick is the slab?

How deep is the beam?

I don't like stiffeners on the bottom flange. Too much plate flexibility between the load origin and where it would be resisted.

Reversible load?

 

[EngDM]

How thick is the slab?

How deep is the beam?

I don't like stiffeners on the bottom flange. Too much plate flexibility between the load origin and where it would be resisted.

Reversible load?

W12 beam, slab thickness TBD, can be whatever is required. Load is not reversible for this one.

 

[KootK]

Something like this would be my Cadillac solution if stiffeners were viable.

Will the load be applied at discrete points at some spacing? Or more or less continuous?

 

[EngDM]

I believe it is more or less continuous, at least that is how the manufacturer presents it and notes as such. Are the stiffeners explicetly required, or if the top flange was stiff enough to not cantilever up would this be okay? I'm assuming that the flange rotating clockwise pushing into the concrete does not require the stiffeners, and that the stiffeners are only required to hold the flange down.

I do like the idea of releasing the tension how you have shown. Where does your compression plane wind up? Bottom flange of beam?

 

[KootK]

When I take away the stiffeners, I tend to feel as though doing nothing at all might be almost as good as anything else that you might do. And this kind of suggests the breakout mechanism that you mentioned .

 

[KootK]

Are the stiffeners explicetly required, or if the top flange was stiff enough to not cantilever up would this be okay?

Enough stiffness in the beam plate elements obviates the need for stiffeners. That said, I'm not sure that much stiffness would be practical. This ties into my question about reversibility. And, to a lesser extent, I would have the same concern if the load were significantly varying even if it did not reverse. I'd not be too excited about the top flange developing a gap with the concrete over time, as I would anticipate.

I'm assuming that the flange rotating clockwise pushing into the concrete does not require the stiffeners, and that the stiffeners are only required to hold the flange down.

In my conception of it, the primary purpose of the stiffeners is to make the whole beam act as a unit torsionally, without significant cross sectional distortion. This allows one to spread out the tension and compression portions of the resisting mechanisms plausibly, without having to sweat a bunch of weird prying mechanisms that no one can evaluate accurately.

Where does your compression plane wind up? Bottom flange of beam?

In the slab. No matter how you handle this, the beam torque becomes slab flexure somehow.

 

[KootK]

How concerned are you about maintaining SOG reinforcing continuity across this thing? I would not relish the idea of the beam becoming a big contraction joint.

How long will one of these beams be?

 

[KootK]

I feel as though it would have been something like this on a set of hand drawn, "pre-anchorage" drawings from the 70's. And for modest loads, I'm sure that it would be fine. Our ancestors were rather clever in spite of, or perhaps because of, the unavailability of accessible FEM.

The connection story told below barely even uses the beam below the top flange for torque resistance.

Which takes us back to the statement below. Basically, you're delivering the load to an embed plate that is the top flange.

I feel as though the "typical" would be to weld the beam to a cast in embed plate rather than embedding the beam.

 

[Boof19]

KootK, I gotta say man I just love your contributions to this site. I hope you never stop

 

[human909]

I agree with the above. Once aspect of consideration. How often are your ultimate design loads reached? The description of the equipment sounds like a rotating mass undergoing an upset and sudden stop condition.

If this is the case then this is an extreme low likelihood scenario and service loads might be a fraction of the ultimate load. So some of Kootk's detailed and thorough concerns might be overkill.