Highly loaded beam to girder simple shear connection

[MJC6125]

If I have a wide flange girder beam supporting a large reaction from another wide flange beam (only one side of the girder) and I'm using a simple shear connection, how is the eccentricity of the connection resolved? I understand that the connection itself should be designed for the eccentricity, but I'm wondering where it goes after that. I don't think we intend for that moment to go back into the supported beam. But I also don't think the girder beams are usually checked for torsion or connections are provided in the girder beam to resist the torsion that the connection would cause. Is the resulting moment typically small enough that this can be ignored. If so, what if I have a very large load and large eccentricity?

This question applies to double angle and shear tab connections. Appreciate any input.

Thanks.

 

[KootK]

I don't think we intend for that moment to go back into the supported beam.

I think that we often do for very large loads. For smaller loads, this effect is usually ignored.

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.

 

[BridgeSmith]

The moment on the supported beam due to the eccentricity of the connection is accounted for automatically if the span length used for the beam includes the length of the connection, i.e. the span length is to the web of the supporting beam. That assumes moment is transferred through the connection (it's not a hinge). Otherwise, the eccentric moment becomes a torsional moment on the supporting girder.

 

[WARose]

There is (of course) a bit of eccentricity......but it is typically ignored. A good back check for that is: what is the slope of the loaded beam? The girder will have to rotate a bit to match that.....and ergo you will have a compatibility torque. But usually the rotation is so small that it is no issue.

 

[TehMightyEngineer]

Most structural software can include offsets in beam to girder and beam to column connections to account for connection eccentricity. This will put a small amount of torsion into your girders. Generally it has little affect except for extreme cases. One nice feature of computer analysis is that it becomes relatively trivial to have the computer account for and verify the effects of these sorts of thing in a structural model.

Ian Riley, PE, SE
Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[KootK]

The primary condition that can cause some problems is where the supported beam is close a girder support that provides significant torsional restraint. Then it's a challenge to get the rotation that you need for redistribution without taxing the connections. And, even then, it's nowhere near as big of a deal as it is in concrete.

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.

 

[NS4U]

The torsion you put into the girder is self limiting. In other words the girder cant twist more than the ends of the supported beam will rotate. Because the twist in the girder is limited by the supported beam you can ignore torsion in the girder.

As long as your connection (even if very long) is strong enough to to take the moment that it sees then I don’t think you need to consider torsion in the girder.

 

[dik]

you design the connection to accommodate the shear and moment and consider the moment to be transferred to the beam CL. If the connection cannot rotate, then there will be little torsion transferred to the Girder.

Dik

 

[cliff234]

Unless I am missing something, the answer to your question is in the AISC Manual. For standard beam-to-girder connections (per the tables in the AISC Manual) the torsion is ignored. For standard double angle connections, connection eccentricity is generally ignored when selecting connections per the Manual. This practice has been validated (for standard connections) through both research and through the use of these connections in thousands of buildings.

 

[BridgeSmith]

I think everyone is pretty much saying the same thing.

In a standard connection, there is moment transfer through the joint, which means negligible torsion in the supporting girder (unless it is braced against twisting very close to the connection of the supported beam) and the supported beam span is assumed to go to the web of the supporting girder. The shear and moment on the connection would be calculated the same as for any other point along the beam.

If the connection creates a hinge in the beam, then there is torsion induced in the supporting girder, which can be calculated easily.

The only question would appear to be, which type is the "simple shear connection," in the OP?