connection of heavy steel beams to existing concrete columns 1

[Eng.Abdulla95]

Hi

I have a 10 m span steel beam which I need to connect to existing 1.2m diameter concrete column, the reactions at the ends come up to be about 500 kN or so.

I need a reference to design the connection. also, what are your recommendation for such scenarios.

Regards.

 

[jayrod12]

We've done curved plates with anchors in these scenarios before. Keep in mind that even though the anchors look spaced niceley on the plate, inside of the column, they get pretty close together depending on embedment.

I didn't personally design them, so I can't tell you exactly what the design process was like, but the final details looked something like the attached

 

[KootK]

1) I thought that I had my own version of jayrod's detail but, upon further inspection, it turns out that I chickened out on it and used the detailing below. I doubt that my solution would work for you as there would usually be aesthetic implications and my load was nowhere as large as yours.

2) Despite what follows, I think that jayrod's detail ends up being the logical conclusion for something like this.

3) When I was considering something like jayrod's detail, one of my concerns was messing with the existing column verts. My plan was to use horizontally slotted holes and to field weld the shear tab such that one could possibly slide the curved plate around the column as required to miss the column verts with the bolts. The slotted holes don't help the bending situation that I'll describe below though as the slots are anathema to that.

4) I doubt that you'll find a detailed design guide for this. When I went through it, I agonized over it terribly and, that, at a fraction of your loads. In addition to damaging the column bars, I also considered:

a) if pounding too many anchors into the column core in close proximity would just turn it to sawdust.

b) how one goes about considering group anchoring effects with the radial setup.

c) if one starts to utilize the bolt columns that wrap around the sides of the column a bit, you get a weird situation where, if you FBD the curved plate, you've really got a fair bit of eccentricity and moment acting on it. Then you have to ask yourself what that means for the plate design and what it means for the bolt design. I'd think that you'd set up a mechanism where the bottom of the curved plate is pushing against the column in bearing and the top is pulling away from the column via horizontal shear in the bolts that are kind of on the side of the column. With four columns of bolts, per jayrod's detail, it might make sense to deal with this effect entirely with the middle two columns of fasteners and then tell the story that the outer columns are mostly resisting just vertical shear.

 

[Agent666]

My one comment on either of those details is that 500kN capacity is a fair old load to be transferring between concrete and steel. As kootk noted with multiple rows and columns things can get rather tricky to define the exact load paths, so making it as simple as possible with direct load paths in this respect gives me far more warm fuzzies than coming up with some elaborate scheme that needs a whole lot of complicated analysis to justify. You'll probably be governed by concrete breakout so this could equals a bucket load of bolts spread out over a large area to engage sufficient concrete.

Do you have a foundation on the level below that you can rest a detail like kootk's baseplate on so you don't need to transfer the load into the column via the bolting, bolting only provided for stability.

Do you know what reinforcement you are dealing with in the column, can you get a logical array of bolts between the reinforcement without risking damage to column reinforcement.

At 1.2m diameter the front face would be relatively flat, enough that I think if you can make a connection possibly work with a flat end plate with grout?

How are any other steel beams (if any) on the job connected to the concrete columns, this might be a good starting point to see what is possible if you can replicate these original details.

Another option might be to create a concrete corbel/column addition to sit the beam on, this could even be a 300mm x 300mm column element with bars drilled into the circular column over the full height of the storey. This avoids all the potential coordination between holes in steelwork and column reinforcement. Get circular column scanned then simply drill in sufficient bars anchored into the new column addition and the circular column core. Once you are safely past the verticals and spiral you can drill all the way to the other side of the column if you like. Then assemble a cage and pour. Might depend if anyone involved can deal with the loss of space from the additional concrete though.

 

[KootK]

I can't resist adding that I hate circular columns. As someone wise once told me here on this very forum: circular concrete columns are for architects. Not only is a square column be easier to build in the first place, it's a lot easier to attach crap to in the future.

 

[Eng.Abdulla95]

Thanks for the replies and solutions. I also find the following publication from sci
,

https://www.steelconstruction.info/File:SCI_P102.pdf..

@jayrod12 solution seem more applicable to me. Spacing the bolts and increase the number seems the right decision to reduce shear and pull out effects.

@KootK you are right for large loads we are more comfortable seating the beam on the column with some bearing destance. But this is an existing column. Anyway 1.2m dia column should provide me with enough space to seperate my bolts.


Thanks for the thoughts.

 

[KootK]

But this is an existing column.

You don't actually think that I've failed to appreciate that it's an existing column do you?

Spacing the bolts and increase the number seems the right decision to reduce shear and pull out effects.

It sounds as though you may be missing the rather critical point that there may be a lot more going on here than just bolt shear/pryout.

How many bolts do you think you're likely to have here? 1.2m is a big column and, if you can get the job done over a fairly flat-ish column surface, that would seem a very promising solution.

 

[Eng.Abdulla95]

@KootK, I faild to read your design, sorry my bad.

yes I think I have 100 mm cloath as part of interior clothings so I think I can implement your details on my design.

Thanks alot

 

[KootK]

yes I think I have 100 mm cloath as part of interior clothings so I think I can implement your details on my design.

That would be wonderful if that's the case. I'd figured that you'd be limited by bearing stress at the bottom of the side column thing with my approach. I'm guessing that you need something to the tune of 50 in^2 bearing to make a go of this. That could be hard to come buy if you've only got 4" depth to work with. That's why my initial assesment was that jayrod's details would likely be the way to go in the final analysis. It just presents some complexity that may need to be addressed. I think that you're first stop is to come up with a quantity and arrangement of bolts assuming that they resist just simple, vertical shear. With that in hand, we can start talking meaningfully about the more nuanced behaviors.