Different type of column to beam connection

[hitch22]

Hello,


I am trying to connect a vertical member (double angles are connected to a gusset plate which is connected to the steel plate I am trying to design) to an existing concrete beam. What makes this connection different to me is that no bolts are to go into the concrete. To achieve this I am trying to "hug" or "wrap" around the concrete beam using rods (see attached diagram).


Right now I am trying to estimate the plate thickness. The issue is that I am not sure how to model this situation. As can be seen in diagram *, I think the bottom plate can be treated as a beam with two pins. Lateral movement would be prevented with the use of angles which I did not include in the sketch.


The top plate could be modeled as a cantilver beam (diagram **). The cantilever would start where the gusset plate meets the steel plate. It's easy to see that the gusset plate is thin and so a cantilever might not be a valid assumption (rotation is possible at this location?).


Does anyone have any ideas as to how to model this?



Thank you.

 

[paddingtongreen]

If you don't need to strongly pretension the bolts, you don't need the grout on the top of the beam. If possible, I would use back to back channels running parallel to the beam with single cross channels at the bolts, bolts hung from the shear centers.

Between the bottom plate and the bottom of the beam, I would put two strips, parallel to the beam and near its edges. This would minimize the moment in the plate. Unless the load is very high, I wouldn't bother with the grout.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[hokie66]

The top plate is a simple beam with a concentrated load at the middle. M = Pl/4. Use same plate at bottom, although loading condition is not as severe.

 

[hitch22]

I've returned to this problem and I am concerned about cracking the concrete beam at the corners of the cross section. While I can find the reactions as concentrated loads ( half of P) how would one compare this against the maximum stress which concrete can resist? Since I don't know what area of the steel plate will be touching the corners I am not sure how to view this load as stress.

 

[JAE]

The bottom plate is the difficult design issue. As hokie66 indicated, the top plate is a simple span with concentrated load in the center.

The bottom plate will not have a uniform load delivered to it by the concrete beam. There will be a concentration of distributed load on the edges of the concrete beam, diminishing to zero at some point as the bottom beam rotates at its "supports" (the hanger rods).

The exact distribution of the stress across the bottom of the beam is tough to analyze. You are correct to worry about high bearing load at the concrete beam corners. It does depend a lot on the stiffness of the bottom beam relative to the stiffness of the grout.

One thought would be to add a partial width bearing plate with an elastomeric pad. The bearing plate might only cover the center 80% of the concrete beam width with the elastomeric pad between the partial bearing plate and the concrete. This would direct more load towards the center of the concrete beam and away from its two corners - thus avoiding corner cracking.

I hope this description is understandable. If not I can post a sketch.

 

[hokie66]

We don't know the forces involved, but I agree that concentration at the edges of the beam could be a concern. If so, do as JAE suggested and use a centralized bearing plate.

 

[hitch22]

Hokie,

The force from the vertical member is about 100 kN which is quite large I would say.

Jae,

So you're saying to insert a smaller steel plate between the concrete beam and a larger bearing plate? This is to steer some load away from the corners of the concrete beam. I'm sorry, but I am not picturing what holds this smaller steel plate to the concrete beam.



Do you think adding stiffereners (running transverse to the concrete beam)could possible keep the bearing plate from cracking the corners.

Oh thank you very much for your thoughts.

 

[koohi]

IF you use a enough stiff system for bottom plate the distribution of stress in bottom surface of concrete will be uniform then the corner crack of concrete column dont be happen.

 

[JAE]

hitch22, yes - the smaller steel plate could be welded to the bottom "beam" to secure in place. By shortening this plate smaller than the width of the concrete beam, the load is directed away from the corner. The smaller steel plate would be "held" to the concrete beam just like your original bottom beam would be "held" to the concrete beam - by the pressure/stress on each other.

koohi point is partially correct - by increasing stiffness to some level, you minimize the build-up of stress near the corners. But the problem may be that you can't provide or afford that large of a bottom beam to achieve this. And even so, there will always be some build-up of stress on the corners no matter how stiff. Even if uniformly distributed, I'd still want to keep stress away from the concrete corners.

 

[BAretired]

I agree with JAE. The area of the plate has to be sized to satisfy bearing requirements against the underside of the concrete. The height of the extra plate must be sufficient to prevent the "beam" plate from contacting the concrete.

BA