Multi-Planar RHS Connections

[MegaStructures]

Hello,

I am looking for a little guidance on how to analyze multi-planar connections for tubular connections (Rec. HSS). I understand that there is a little more that goes into weld design for a tube-tube connection than is given in Blodgett's elastic weld procedures, due to the flexibility of the tube face and there are also some additional failure modes in these connections, mostly again issues with local strength in the tube face.

AISC Ch. K gives some guidance on how to account for the uneven load distribution in a weld for in-plane HSS-HSS connections, but does not give any guidance on multi-planar connections. CIDECT, really is the committee responsible for most of the guidance in Ch. K and does offer some very limited guidance in their DG3, which basically amounts to "apply a reduction factor of 0.9 to a uni-planar connection for multi-planar joints". My structure is a space frame and the joints don't closely resemble a uni-planar HSS connection, thus I don't see that any of the force distribution modifiers apply to my joints. I am tempted to use Blodgett's weld design method with (conservative) 2D weld geometry and use this 0.9 reduction factor, without considering force redistribution.

Anyone have any experience designing these type of joints that might be able to offer a little guidance? I do work in an industry that requires load testing of the structure after construction that will be completed at a load greater than the load that will ever be on the structure in service that will help to verify results and I plan to do UT of select joints after test to prove soundness of welds.

“Any idiot can build a bridge that stands, but it takes an engineer to build a bridge that barely stands.”

 

[DrZoidberWoop]

I think your reasoning for utilizing the reduction factor is sound and conforms to the latest traditional design methods from AISC-360-2016 and CIDECT DG3-2009.

First thing I found was a concise rundown over at Steel Tube Institute:

https://steeltubeinstitute.org/resources/multi-planar-welded-connections/

Side note: your connection looks a little similar to the example used on pg 81 of the 1996 version of CIDECT-DG3.

 

[MegaStructures]

your connection looks a little similar to the example used on pg 81 of the 1996 version of CIDECT-DG3

Somewhat similar for sure. Looks like that one is two uni-planar trusses, mine has tubes in a lot of crazy angles.



“Any idiot can build a bridge that stands, but it takes an engineer to build a bridge that barely stands.”

 

[r13]

I tend to simplify the matters - a 3D joint in equilibrium with planar reactions on all faces of the tubular support. The weld stress maybe additive if the edge weld overlaps.

 

[DrZoidberWoop]

I've done a few heavily skewed HSS space trusses like that, but I can't say I've been in your exact position. In my geographic region, EOR's often delegate connection design to the fabricator, so I've been able to get the exact loads and either 1) talk them into using nodes or other load collection elements or 2) design their ridiculous skewed HSS with heavily customized weld calcs.

It's a pain when a rectangular HSS frames into the corner of another rectangular HSS, and it's even worse when they're relatively thin-walled. Not to mention the assumptions involved for distributing axial loads into each weld.

 

[HTURKAK]

I have experience with CHS elements for space truss similar structures. IMO, the connection shown at the picture you have posted too complicated for welding and inspection .
My suggestion will be , simplify the connection with using intersecting vertical plates and with the fillet welds sized for member loads and with allowances for uneven load distribution.

The snap is from the book Handbook of Structural Steel Connection Design and Details by Akbar R. Tamboli

and I will suggest you to look the book HOLLOW SECTIONS IN STRUCTURAL APPLICATIONS by J. Wardenier, J.A. Packer....