Shear capacity of cast in plate with rebar

[Cos0Sin90]

Hi,

I have a steel member connecting to a RC column. I avoid using anchors or shear studs due to edge distance issue. Instead, I have this cast-in plate with rebars that weld to it.

I am wondering if anyone can guide me on how to check the shear capacity in this case. I have gone through some PCI examples, but it seems like they don't really talk about this scenario. Most of the examples are related to shear studs.

 

[Cos0Sin90]

Do I need to consider edge distance in this case?

Or do I rely on shear friction so that the rebar will be just in tension? Then does the interface between the steel plate and concrete provide the adequate friction as concrete to concrete?

 

[cliff234]

The best guidance I can give you is to point you to ACI 318-19, Chapter 17. The minimal edge distance in your sketch is a problem regardless of whether you are using welded rebar or headed studs. It looks like the bars closest to the slab edge will be virtually useless. You might want to verify that the welded rebar in your detail does not hit the reinforcing steel in the concrete. It looks like there will be interference.

 

[dik]

Can you use a single rebar, or headed stud that is larger diameter? and more edge distance? Can the rebar be horizontal and slope down, going back into the concrete mass?

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[dik]

I've designed numerous connections like this. The load goes into the concrete as tension in the attached rebar, if it is horizontal... The concrete and column reinforcing have to accommodate the shear. Can you make the bearing plate longer and move the anchors away from the face of the column?

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[Cos0Sin90]

Thanks everyone.

@dik
The shear comes from combined shear and torsion. See below sketch in plan view. If I move the rebar away the torsion increases as well. (I have like 150kN shear at steel beam end). Any suggestions?
If the interface is roughened, won't the rebar in tension based on shear friction theory so edge distance is not an issue?

 

[dik]

Can you make the plate longer to provide added cover for shear connectors and just design as a normal shear connection. With added clearance, I'd be using headed studs. Torsion, or flexure in two axis?

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[Cos0Sin90]

Thanks dik.

The column is 350mm deep only. If I extend the plate and have two columns of shear studs with 100mm spacing, the edge distance is only 125mm still, which doesn't give me much capacity.
I am thinking either:
1. provide hairpin so concrete breakout is not a problem.
2. have a L plate with one column of shear stud on this face (so that edge distance increases) and another column of rebars on the other face to take the tension.

Anything I miss or should be careful about?


Back to the very first question, if I weld rebars here, why can't I use shear friction so that rebars take tension only? I am comfortable with dowel bars to take the shear even if they are close to the edge when the interface is concrete-concrete. So does that mean if the interface is steel plate - concrete the shear friction theory does not apply? Is it because the interface is too smooth?

 

[Gus14]

Do you have moment reaction? That will make things even worse. I tried to design these type of connections before, but then it got too complicated and I chickened out.

 

[Cos0Sin90]

@Gus14, moment is relative small due to small offset so I don't worry too much about that.

 

[dik]

350 is lots... could use a 300x300 plate and stay a couple of inches from the edge with the anchors... could use 4 headed studs in lieu of rebar.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[canwesteng]

There isn't really a difference between concrete-concrete and steel-concrete, other than lower coefficient of friction. The problem is what happens to the shear once it gets into the concrete? It's going to need to be transferred to ties.

 

[Cos0Sin90]

@dik with 125mm and edge distance it still doesn’t give me too much. I just put more rows to reduce the shear.

@canwesteng if it is based on shear friction theory, the shear will become tension in the rebars, and since they are rebars not studs it will rely on bond stress, just like starter bars?

 

[BAretired]

The shear comes from combined shear and torsion. See below sketch in plan view. If I move the rebar away the torsion increases as well. (I have like 150kN shear at steel beam end). Any suggestions?
If the interface is roughened, won't the rebar in tension based on shear friction theory so edge distance is not an issue?

Torsion, what torsion? I don't see torsion. Shear in what? The beam? Show the forces with arrows.
If tension parallel to beam axis, the yellow bars shown below provide a good anchor.
If compression parallel to beam axis, you probably don't need the yellow bars, but they wouldn't hurt.

I would not rely on shear friction so close to the free edge.

Why not cope the beam and weld it directly to the embed plate? Omit the plate or bar extending from the beam.

 

[Cos0Sin90]

@BAretired. The torsion comes from the shear face from steel beam times the distance between the shear center of the anchor group and where steel beam is connecting to the MS plate.

The sketch is on plan view

 

[bones206]

I typically use a product ESR report to determine capacity for these types of deformed bar anchors. For example:

https://icc-es.org/wp-content/uploads/report-directory/ESR-2907.pdf

 

[BAretired]

@Cos0Sin90,

Let's go back to your first sketch. There is torsion in the column, but that has nothing to do with the connection under discussion. No matter how the connection is designed, the torsion in the column is unchanged. I am suggesting a couple of anchor bars as shown in yellow in the sketch below to prevent anchors near the free edge from popping out.

 

[Cos0Sin90]

@BAretired, I don’t see how torsion does not affect the connection. The horizontal shear in my first sketch comes from the torsion, otherwise you will just get vertical shear.

I do agree with what you suggested. I believe having hairpin around the stud will also do the work.

But what I don’t really understand is, if it is a concrete to concrete connection we would simply have dowel bars without worrying about the edge distance based on shear friction theory. What makes steel plate to concrete different?

 

[BridgeSmith]

As far as I know, you can't apply shear friction to a steel plate on concrete, because the steel plate is flat - there's no 'bumps' on the plate to provide the interlock of the rough surfaces that makes it work. Shear friction works by having 2 interlocking surfaces that have to move apart in order to slide, which is what causes the tension in the rebar.

The resistance to shear at each bar in the connection, as you had it shown originally, will be limited by either the shear capacity of the rebar, or the concrete breakout, whichever is smaller.

With the yellow bars in the configuration BAretired showed those would develop the tension strength of the bar, in the direction parallel to the bar, assuming adequate capacity of the weld.

Not sure what kind of capacity those yellow bars would have for shear in the direction perpendicular to the bar, but if the plate's embedded so there's solid concrete up against the edge, there would be substantial capacity from the plate bearing on the concrete along the edge.

 

[phamENG]

I'm confused about the force, too. When you say torsion, do you mean beam end moment, or do you mean torsion in the beam about the beam's longitudinal axis?

It would be best if you could provide a 3d/isometric of the connection with all of the associated forces acting on it.