Eccentric truss connection

[volunteera]

I am analysing an open top offshore container with Staad pro. It is a 20ft container made out of HSS truss walls with eccentric KT-connections. As one can notice from below pictures, diagonal web members of side and end walls (which are released from translating moment in all directions) are eccentrically attached to top and bottom chord joints. To illustrate this I modelled each of the eccentric distance into a separate beam member. This allows to transfer axial load from diagonal webs into chords correctly which adds value into moment diagram of the chords.

Another option would be attaching the webs directly to the joints and using offset tool however then it ignores additonal moment value originated by diagonal web axial force in offset distance from the joint.

My questions are:
What is your preference in modelling of such KT type eccentric connections?
When analysing such structure do you release vertical web from moment translation as well?

 

[MotorCity]

For a traditional truss analysis where all web members share a common work point, and connections are made with gusset plates, I pin the ends of all vertical and diagonal web members. However, for your situation above, I would bet that the web member connections to the truss chord will act more like a moment connection than a pinned connection due to the robustness of the connection and the chord bending due to the offset work points.

 

[volunteera]

Motorcity,

That's why for conservativity I pinned the diagonal webs. This will pass all moment on vertical web to stabalize. I think this is the worst case scenario.

Thanks for commenting on.

 

[MotorCity]

I don't think that pinned ends is necessarily conservative. If you model them as fixed, a moment will be transferred to the wall of the HSS. This could lead to localized failure modes such as punching, yielding, crippling,etc. As I mentioned, you may have modeled it as pinned, but my gut tells me it will act more as a fixed connection. Because the workpoints do not coincide, you are definitely imposing a moment on the chords