Concrete Edge Breakout - Combination / Interaction on adjacent edges

[NRam]

Hi all,

Struggling conceptually with understanding this issue. So I have a few questions.
See attached document and images for the background.

1. Where does the V_parallel = 2V_perpendicular provision from ACI 318-19 Section 17.7.2.1c come from?

Even in the commentary, the code merely states
"The maximum shear that can be applied parallel to the edge, V||... is twice the maximum shear that can be applied perpendicular to the edge."
This seems somewhat arbitrarily chosen. What's the mechanism here??
I've seen some posts discussing this that merely state that it ought be checked. But does anyone know the rationale behind it?​

2. What would you consider a safe interaction equation or methodology?

a. Would you consider them completely independently? This approach seems unreasonable to me. The stresses would combine in someway.
b. Would you consider a linear interaction? Meaning Vd/Vc + Vd/Vc > 1

This to me seems more reasonably, and what I would design for, but could result in over conservatism.
This would require each embed above to be below 50% capacity.​

How would you tackle this issue, and what load would you design the anchors for?

In advance, thank you!

 

[Eng16080]

I just spent way too long thinking about this with the hopes of arriving at some clever solution, and I've come up with little. This is a really good question, and ultimately, I'm not sure the best approach. I agree with you that there would be some interaction between the forces along the two sides, and I wouldn't want to ignore that.

My approach would likely be a conservative one, like you mention in 2b. I would probably design each of the two connectors independently for the total combined load, and if that leads to a reasonable solution, I would leave it at that. I imagine this would be quite conservative, although I can't say for sure, considering the complex interaction of the stresses.

Some other brainstorming (not necessarily endorsing these as good ideas though):
[ol 1]
[li]With the two forces being perpendicular, perhaps this is similar to a combined tension and shear interaction, as per Section 17.6, and perhaps using that interaction equation makes sense somehow.[/li]
[li]You could extend the connectors of the two sides either up or down, alternatingly, to minimize the "interaction" overlap area.[/li]
[li]Maybe joining the two connectors to instead create a single "L" type bracket would minimize the risk of the concrete failure at the common edge.[/li]
[li]Encasing the column with a steel collar wrapping all 4 sides would presumably prevent this failure mode although I'll admit this is perhaps crazy.[/li]
[li] This is also crazy, but imagine cutting the square column in half into two equal, separate triangles with each connector going to it's own triangle. Now, if each triangle piece can reliably resist the connection forces, I would think the full square with two joined triangles would be at least as strong. (Admittedly, analyzing a triangular piece of concrete is perhaps an even greater problem than the one presented here).[/li]
[/ol]

One other thing: Regardless of the solution you come to, I think you'll want to stagger the connectors to avoid interference or near interference at the shared corner.