Round HSS Splice

[jreit]

All,

I've been trying to find a procedure to design a splice for a round HSS member. It is being used as a beam rather than a column and the majority of the load is flexural. Most of the design procedures I see being used are based on the assumption that the majority of the load is axial. Does anyone have an idea on where to look? Thank you in advance.

 

[CANPRO]

welded splice or bolted splice? are there limits on how far connection material can protrude from the member?

For welded I would use a round plate slightly larger in diameter than the HSS and use fillet welds - i.e. 7" round plate for a 6" round HSS, fillet weld all-around each HSS.
For bolted, I'd use a bolted end plates, maybe just (4) bolts in a square pattern if the loads permit.

Axial or bending I think the procedure is more or less the same - you're going to end up with compression/tensile forces in the wall of the HSS acting on the plates...now you're looking at weld strength per unit length, bolt forces based on state of the HSS nearest the bolt, plate bending that is tributary to the bolt, etc... What I'm saying is you could find your max bending stress and apply it to the entire cross-section and then use your axial splice procedure - I don't think this would be overly conservative, especially if your beam can bend in multiple directions.

 

[r13]

Check the code for splice requirement for flexural members. I recall it was either required to maintain the original sectional strength of the member, or 150% more. Not sure the present requirement.

 

[HTURKAK]

In past we have used bolted CHS flange-plate connection. However you may consider butt splice with internal HSS backing so V type groove weld could be used.

 

[JAE]

For splices of steel members in buildings (in the US), you only need to develop the strength required by the loads, or, in the case of large member splicing to small member, you need to design to the nominal strength of the smaller member (Section J6 of the AISC Spec).

For a pipe-to-pipe splice, typically we simply provide a bevel on one side and use a complete penetration single bevel weld all around. This requires a back-up band inside the pipe for the weld.

If the pipe-to-pipe sizes are the same, and the load is small, a partial penetration weld could work (and avoid the back up bar) but I've always shied away from that as someone in the future might re-use the pipe for another purpose and not realize they don't have full 100% pipe capacity at that splice joint.