Concrete Anchors - Load Transfer Fundamentals? 1

[psychedomination]

Hi there,

I have been struggling to wrap my head around load transfer, especially when using concrete anchors (both cip and post installed). If someone could provide an explanation to help me better understand, it would be greatly appreciated.

I understand that for normal reinforced concrete structures, if you have a RC column connected to a RC footing, you would fully develop the steel reinforcement from the column into the foundation to make sure that the moment is adequately transferred into the foundation.

However, for anchors, when a moment is present, it is normally converted into a couple, and the anchors are put in either tension or compression. The anchors are checked to see if the steel would fail in tension or shear and if the concrete would fail in break out or pull out. With this said, I wouldn't have thought it necessary to transfer loads from the anchors into the rebar if the normal checks for steel failure and breakout etc are satisfied?

Would the first section (typical lightly loaded strip footing) shown in the attached sketch
suffice if the above checks are satisfied, or would there need to be a transfer of load from the anchors into the reinforcement (more like the second section)?

What would be the failure mechanism for not transferring the load and what would be the reasoning for transfering the anchor load into the reinforcement?

This is probably a silly question but one that I haven't really come to fully understand yet lol, so bear with me.

Any clarity would be appreciated.

 

[Tomfh]

Anchor design doesn’t ordinarily rely on an assumption of lapping to the structures reinforcement.

Obviously it’s better if you do lap with the steel, but it’s not generally designed that way, hence anchor embedments tending to be much less than steel lap lengths.

It’s a weird grey area. You could cast in a piece of straight rebar 250mm, and lap it 200 with the footing foundation, and most engineers will say insufficient lap, not ok.

And yet if you call the piece of steel an “anchor” and glue it in 250mm then now it’s ok.

 

[bugbus]

I believe this happens a lot without us even really thinking twice about it.

The common occurrence I see is where moment connections are formed in concrete with straight starter bars. E.g. a column connected to a pile cap with straight starter bars (i.e. not anchored at the base with a hook). Alternatively, a wall projecting from a footing, again with straight starter bars. This situation usually won't pass the strut-and-tie check, because without a hook, the bars are not anchored at the base of the pile cap and it is hard to imagine a load path from the tensile reinforcement in the column to the tensile reinforcement in the pile cap. Where the embedment length of the bar is equal to the development length, the point at which the bar is fully anchored may well be at the top surface of the pile cap. As far as I can see, there is no strut-and-tie model that can make this work.

Nonetheless, I believe this is still a valid moment connection, and I see this all the time in my world. Particularly where the projecting member is somewhat smaller than the supporting member, this is never really questioned. For an anchor, we are relying on the tensile strength of the concrete to resist pulling out, and to transmit the concentrated anchor force to the rest of the structure. It is the same for a straight piece of reinforcement. There is obviously a potential failure surface that would allow the straight reinforcement to pull out, but the tensile strength of the concrete is more than sufficient to avoid this happening.

I have seen a lot of posts on this forum which state that the tensile strength of concrete should never be relied on. To that I would say that we do this all the time. Whether it is a cast-in anchor, a moment connection without a neat point of anchorage (e.g. ending in a hook) that can satisfy a strut-and-tie model, or even the anchorage or lapping of reinforcement itself. If concrete really did have zero reliable tensile strength, it would simply not be possible to develop or lap reinforcement. Splitting cracks would immediately form and allow the bars to slip without resistance.

 

[Ron]

Rebar failure (slipping) is a completely different mechanism than anchor pullout. For rebar, the concrete is so confined laterally that it doesn't fail in the typical shear-cone, but fails locally along the rebar with multiple mini-shear failures, closely spaced.

For anchors in tension, the concrete is not restrained at the top, thus a shear cone is developed at failure. The area of a shear cone is significantly larger than the shear area around a rebar splice, thus less embedment is necessary.

 

[RPMG]

Ron is correct. It is theoretically possible to have the sectional failure at the top of a pier because the rebar was undeveloped, due to anchor tension.