Clip Angle Flexibilty

[peinma]

I have a lightly loaded secondary beam with welded clips each end (to a beam webs). The clip is welded on 3 sides, so not detailed as simple shear connection. Design for moment of eccentricity of the beam reaction and/or beam moment of fixed beam (wl^2/12)? Checking both but looking for thoughts. Thanks.

 

[BAretired]

If this is already built, you should provide a sketch. Otherwise it would be usual to detail it as a simple shear connection.

Even if the secondary beam is rigidly connected to the supporting beam, and assuming that the supporting beam is a wide flange, it is unlikely that you will be able to transfer significant moment because it has to be carried in torsion.

BA

 

[peinma]

Already built - please see attached. I don't need fixity for the beam design, but want to check the welds on the clip. The 3 sided weld on the beam web does not allow flexibility for a typical simple shear connection, so I need to size the weld for moment. Is the moment the reaction x the eccentricity of the connection (about 6") or the end moment of the beam assuming it is behaving with fixed ends. My thought is there would be some flexibility/yielding in the web of the W12 if the welds on the clip angle are capable of resisting the reaction with the connection eccentricity. Thanks.

 

[TehMightyEngineer]

I see people mess that up all the time.

Maine EIT, Civil/Structural.

 

[KootK]

The 3 sided weld on the beam web does not allow flexibility for a typical simple shear connection, so I need to size the weld for moment.

Not necessarily. While I agree that the weld will inevitably see some moment, I believe that most designers would ignore it in this situation and design would proceed as follows:

1) Assume that the angle to supported beam connection is a moment connection.
2) Assume that the angle to supporting beam connection is a pinned connection transferring only shear (see BA's comment regarding torsional flexibility).

The philosophy is similar to extended single plate shear connections in the AISC manual (Link).

In the Canadian steel manual, our standard clip angle connections come with welds on the angle heel by default. Angle flexibility be damned it seems. It works in practice and is supported by some dated research done at the University of British Columbia.

Theoretical chit chat aside, your connection is fine. No matter how you chose to look at it, you should be good to go for a simple shear reaction delivered by a W8x10.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[BAretired]

If the two sided weld to the W8 is adequate to resist a moment of M = R.d where R is the simple beam reaction and d is the distance from the c.g. of the W12 to the c.g. of the two sided weld group, the connection cannot fail, but it is not a connection to be recommended.

The actual moment occurring at the c.g. of the W12 is indeterminate and depends on the welds of each joist to the beam, the web thickness and torsional stiffness of the W12 as well as the support condition at the opposite end of the W8. It is, in my opinion, a bad detail and one which should be avoided in future work.

BA

 

[paddingtongreen]

The main problem that I see here is that the weld will unzip if any appreciable moment is placed on it because of the stress concentration at the end of the weld. Normal weld would be for the outstanding leg with returns, usually about half an inch or so along the top and bottom. The returns are to have the lift off point for the welding rod away from the highest stress point because it can be a point of weakness.

It raises eyebrows when this unzip happens because the split stays within the weld material and that is stronger than the matching base material. It took me ages to fight off QA/QC on a nuke when a piece of mishandled steel unzipped through the stronger weld material.

Michael.
"Science adjusts its views based on what's observed. Faith is the denial of observation so that belief can be preserved." ~ Tim Minchin