Steel connection for beam supporting girder

[KingKongdoor]

I have a situation where a cantilevering beam is supporting a perimeter beam/girder. See attachment.
Does having the perimeter beam be continuous pass the supporting beam with a splice a couple feet away from the supporting beam put eccentricity into my double angle connection? I think not but im second guessing myself and can't find a design example with this situation.

 

[cliff234]

Yes. In my opinion this connection should be designed to resist the eccentricity. This will not unduly penalize the connection. Most of the time when a beam frames to a supporting girder you want the connection to be flexible enough to permit end rotation so that the connection will act like a pin and allow the end of the simply supported member to rotate. In your connection you don't want that cantilevered member (the beam passing by the supporting beam) to rotate - so you want to account for the eccentricity in the connection design. This can usually be accommodated by using double angles that are thicker than what are usually used for simple shear connections, and by specifying that standard holes be used (versus short-slotted holes). Short-slotted holes are necessary to accommodate tolerances, camber, etc. in simple shear connections, but in this connection using standard holes should not be a problem. Tolerances can be taken up elsewhere.

 

[KootK]

I vote no for practical design purposes. The supporting cantilever beam tip will of course be twisted to match the slope of the perimeter girder at the joint. And that will induce some torsion in the double angle connection. That said, the amount of twist should be small and torsional flexibility of the supporting cantilever beam should be high.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[JoshPlumSE]

Yes and no. I wouldn't call it an eccentricity to the connection. To me this is a deformation compatibility issue. The spliced beam will have some rotation at the point where it is supported. The amount of rotation could be approximated based on the flexural rigidity of the back span beam.

This deformation will cause one of two things to happen.

1) The connection will be flexible enough (short slotted holes) to allow the rotation without imparting any force into the supporting member.
2) The connection will be rigid enough that it will impart some torsion into the supporting beam... until the two rotations are equal.

That being said, I beams are very flexible in torsion. So, I have a hard time believing that the torsion that develops will be significant. To the point where the torsional force induce in the beam is practically equal to zero and can be ignored for design.

 

[KingKongdoor]

The top flange of the supporting cantilevered beam will be welded to a non-composite concrete/metal deck system. I don't think the cantilevered beam would be allowed to twist.

 

[KootK]

It'll twist, and warp, unless the bottom flange is similarly restrained. But yeah, if the beam will truly be restrained from twisting, then the connection should probably consider the forces that would induce that twist.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.