Simple Shear Tab Connections to Column Webs

[laurelerin22]

Hello,

I'm currently designing some simple shear tab connections into column webs, which I believe are pretty common. We are using shear tabs that extend beyond the flanges of the column, so that large beams can attach without copes or erection issues. We are providing welded stiffeners at the top and bottom of the tabs within the column flanges. These shear tabs have a single row of bolts (see picture attached for clarification).

Do we need to account for bolt-group eccentricity on the single row of bolts, if the distance from the face of the column flanges/stiffeners to the bolt row is only 3"? It seems that standard practice is to NOT account for an eccentricity, unless the bolt row is more than 3-1/2" from the stiffeners, but I've found some conflicting information about this scenario and the AISC manual is not clear. RAMConnection does not account for any bolt group eccentricity when we model this type of connection in the program with the bolt row located 3" from the stiffeners.

Hopefully this is a pretty basic question for any of you connection designers out there. Thanks for your help!

 

[laurelerin22]

I need to make a correction... RAMConnection DOES account for the eccentricity, but you have to input an actual load on the connection, as opposed to just looking at the connection capacity, for it to consider the eccentricity. I'm still unclear on where the eccentricity would be taken to (i.e. just to the face of the column flanges/stiffeners, to the c.g. of the weld group for the shear tab, or all the way back to the column web).

 

[JoshPlumSE]

What eccentricity do you use? I think that is a darned good question. My belief is that it would be the FULL eccentricity from CG of bolt group to the column web. AISC's design examples doesn't directly address, so there isn't a definitive reference I can point towards.

However, you might look at Tamboli's Handbook of Structural Steel Connection Design and Details. I'm looking at the first edition figure 2.49 and it implies the full distance. Though that is for a skewed connection. Then it references Kloiber and Thornton's: "Design approaches to to shear connections for skewed Members in Steel Structures" published by ASCE in the 1997 Proceedings of Structures Congress

 

[jdgengineer]

Maybe I'm misunderstanding the question, but I think that this is covered in the AISC Manual. They go through the design checks on page 10-102 (black book or red book)

If the distance from the bolt centerline to the connection to the column is 3.5" and the number of bolts is equal to or less than 9 the AISC says you can ignore the eccentricity. For other cases they say to size the bolt group accordingly to resolve the eccentricity. Alternatively, they say that the eccentricity can be taken out by other means. They reference research from Sherman and Ghorbanpoor.

 

[laurelerin22]

The AISC manual is not clear about the difference that stiffeners make in locating the eccentricity. The debate is whether the "connection to the column" (eccentricity) has to be taken all the way back to the column web or just to the face of the column flanges/stiffeners. The research by Sherman and Ghorbanpoor argues a scenario where you can take the eccentricity to the c.g. of the weld group for the shear tab with stiffeners. It makes a huge difference in the number of bolts you need to reach a certain connection capacity.

I've seen connection designs from steel detailers that obviously don't account for an eccentricity (assuming that the stiffeners designate the "connection to column" and therefore allow the shear tab and bolts to be designed as "conventional"), and I've had discussions with other engineers who argue that the stiffeners don't matter and an eccentricity has to be considered all the way back to the column web (and thereby forcing an "extended" shear tab design with an eccentric bolt group).

 

[wannabeSE]

I thought stiffeners were added to provide torsional stability for the extended plate (whether it needs it or not). There is an MP4 presentation on this from last years Steel Conference on AISCs web site. Scroll down to 27a or 27b at

http://www.aisc.org/content.aspx?id=32756

 

[structSU10]

I would think it depends on how you want to design the column - if you want no moment in the column design the bolts to take the full eccentricty from the web, if you can live with a moment in the column, reduce the eccentricity to the bolts, and apply the moment from the remaining eccentricity to the column. The eccentricity has to go somewhere, and should be accounted for somewhere in the assembly.

 

[laurelerin22]

@wannabeSE - Yes, typically the stiffners are provided for torsional stability of the for the extended plate. However, I've read that you can also size them to take the moment from the bolt group back to the column web.

@structSU10 - After doing some additional research, this is the conclusion I've reached. This question was asked in the March 2012 issue of Modern Steel Construction, and the answer was the same as yours (see attached). If you account for the additional moment in the design of your columns and design the stiffeners for the moment forces, then, yes, you can reduce the eccentricity on the bolt group.