Eccentric Loads on precast overhead panel?

[UpsideDown]

As shown below, Do I need to consider the moment induced by the eccentric loads from RC slab and thus check the pull out capacity of cast-in plate for panel-overhead panel? Or can I assume the load will be central loaded by the use of EA plus ferrules? I have reviewed comps by others but haven't seen consideration being given to this.

 

[r13]

You didn't show the load, or any meaningful information. What is holding the slab, are we looking into a wall separate by space?

 

[Tomfh]

I’d include any torsional loads. If anything’s going to break it’s those brackets ripping out, and so you want to make sure you’ve captured the lateral loads on them.

 

[UpsideDown]

Sorry my bad, was meant to ask a general question. The details are as below:
1: panel: 200mm thick, 6m wide and 1.5m high
2. loads: 30kN/m dead load (panel sw excluded) and 6kN/m live load (from RC slab). RC slab has internal supports that were not shown.

 

[UpsideDown]

Is there any way to get rid of the torsional loads? I dont think the cast-in plate have much higher than 10kN capacity, which means it will fail when torsion from eccentric loads together with wind loads are considered. What do you normally detail the RC slab to overhead panels?

 

[r13]

Depends on how the slab was designed, you may require, or not require to consider torsion caused by the cantilever slab. Is the slab existing, or a new construction? How thick is the slab, and do you know its design load? Also, what is a "ferrule", a nut?

 

[Tomfh]

Upside down,

I don’t know how you can get rid of the torsion load entirely. The slab itself may be putting negative moment into the connection too. So even if it’s rigid connection the slab itself may be wanting to rotate it off.

I hate these details myself, it just seems wrong to hang lumps of concrete in the air with little brackets, but you end up with them here and there.

 

[civeng80]

Without knowing the width of the panel, what about stiffening up the slab so that it spans the panel, then no torsion resulting in no pull out of the attachments.

 

[r13]

I think "bending" is a better word for this situation. The precast panel has 4 plates, either surface mount or embed in the walls, that would help to hold the panel stable. Attention should be direct to the connection of panel-slab interface.

 

[UpsideDown]

Just to clarify that the slab is not cantilever. Sorry about confusion caused by the attached pic. Like I said it has internal supports. So the slab is 200mm thick, 6m single span. It's residential so lets take 2kPa as live loads. Yes "ferrule" is like a nut in panel and we have bolts connecting it with EA.
The torsion I am considering here is actually not by the bending moment from the slab but the eccentrical loads (reaction from slab) applied on the panel face instead of the central line of the panel. So the design line load on the panel is around 30kN/m, then we have 30*0.1=3kNm/m torsion along the panel. I am wondering how the cast-in plate (shown in purple in elevation view as above) can take this torsion. If not, how do you normally design this?

 

[UpsideDown]

Yeah. I dislike these connection as well but I cannot think of alternative way to do it. I have seen plans from other firms doing it this way without giving consideration for the torsion from their comps. I would expect the cast-in plates (shown in purple) for overhead panel to panel should form a T and C couple forces to resist the torsion with top one in tension and the bottom one in compression and I don't think it will past. (cast-in plates are 200x200x12) with 12 thick capping plate welded to both of them.

 

[Tomfh]

Yes it’s how we do it, and I agree with the basic concept of T/C push/pull via the brackets to resolve the torsion.


You appear to be considering the torsion from the slab edge landing eccentrically on the panel. In my opinion the “eccentricity” is liable to be much greater, because the ferrule connection will be quite rigid, and thus you have a moment connection. This may push your point of zero moment further into the slab. I.e. the slab will cantilever out whether you like it or not.

 

[UpsideDown]

Okay, so how the the cast-in plate for the panel-panel (the purple one in elevation view not the EA brackets) can hold such huge pull/push forcers especially when negative moment from slab is considered? I checked the out-of-plane capacity of the connection and it should be around 10kN only. What cast-in plate for panel-panel connection do you use to make it work?

 

[civeng80]

How wide is the opening?

 

[r13]

I think you maybe getting better perspective by image that, without the precast panel how the slab will perform. Will it falling down? If it will, you get big trouble. If not, then what is holding it up? Then you can fill in the blank. The 4 plates play very important role. They support the panel weight, and resist the pull from the slab.

 

[r13]

If you can provide sketches of plan at the installation, and section cut, that indicate material and thickness of the wall, thickness of the precast panel, and how the slab is connected with the wall, if non-concrete wall, maybe some good ideas will be came about.

 

[UpsideDown]

Please find below the plan view. (the red lines denote 200mm panels)

The 200mm slab is designed to span along Y direction (which is 6m plus 1m cantilever in the middle). The overhead panel I am talking about is at the front over the 6m opening.

 

[civeng80]

There is not much support bottom side, have you checked that the small panels can take the load?

 

[UpsideDown]

Are you talking about the compression capacity? If so it does work. I can also increase the internal panel with to make it a column. I am also thinking about making the two overhead panels as a continuous one and use dowels bars connecting to the internal panel below. My main concern here is the out-of-plane capacity for the panel to panel cast-in plate connection though.

 

[Tomfh]

Upsidedown,

I don’t recall doing a hung panel with a slab framing in like that.

What’s the moment in the slab assuming a rigid connection to the panel? What sort of loads going into the clips?

Generally I add more clips when I’m not happy with the loads going into them.