L shaped rebar welded to steel column 1

[LQQD]

Hi everyone!

I am reviewing a retrofit project where I found a connection detail I have never seen before, I have serious doubts that this can work.

There is concrete overhang beam which is partially interrupted by a circular HSS column, the beam top bars, which are working at full capacity, are 5#8 bars. Only two bars are continuous over the column going through the sides of the column, the remaining 3 bars are welded to the steel column.
This 3 bars are welded by their 90° leg (flave-bevel goove weld) perpendicular to the direction of the bar/load.

My thoughts are that this load will produce a high stress concentration at the top of the weld due to excentricity, from the vertical distance between the center of the bar and weld.
Some conifnement may be present by the concrete mass restraining the bar from bending, but seems to risky to ask 100% of ultimate capacity to that bar/connection, it looks to me like the weld will tear apart like a zipper.

(I think I have seen similar things when a U bar is welded upside down to an embedded base plate, usually with lower diameter bars, I never liked this solution and can't find any referene now)

Does anyone used or seen such a detail before? do you have a reference for this either allowing or banning this kind of weld?

Find drawing attached.

Thanks!

 

[KootK]

You're instincts are spot on, that is a horrible detail and those welds may well just unzip as you have intimated. You know, fracture mechanics??

 

[Tomfh]

Yes they'll peel off the post.

 

[271828]

Agreed with the above.

The HSS might bend locally also.

 

[JohnRwals]

I think rebar welding could fail due to zipper effect, also.
But, isn't it dependent on how much bending moment is?
May 2 rebars are enough.
Or, development length is enough.

 

[LQQD]

Thanks averyone, if someone has different opinion it's welcome.

@JohnRwals, all bars are almost at their full tension unfortunately, 2 bars are not enough.

I will let you know if the designer has a different opinion, a justified one, when he replies my report, I don't see this happening anyway.

Thanks!

 

[EZBuilding]

I would also expect that if you need 5 #8's worth of negative moment capacity to the right of the column, you'll need just about the same to left of it...

 

[KootK]

Much depends on the beam proportions but I suspect that you'd be much better off bundling the disrupted bars on the side or placing them in two layers. You'll sacrifice a little flexural depth in exchange for a far less questionable condition.

 

[JLNJ]

Drill some holes through the pipe and run the bars clean through. Add a couple of extra holes for grouting and Bob's your uncle (once you figure out the shear transfer).