Design chords to moment?

[Gile_]

Hi

I have this roof truss that contains steel T sections as its top and bottom chord.
The axial force is very close to their capacity already. If I do a combined N&M check the chord will fail due to the low bending moment capacity of T section.

Now my question is do we need to consider bending moment in the chord? If it only fails at the connection are we allowed to allow it to fail since it will just generate a hinge at the point, which is what a typical analytical model of a truss is despite the fact the chord is a continuous member in reality?

 

[GeorgeTheCivilEngineer]

Is this an assessment of an existing structure? If so, are there any signs of distress in the existing structure?

Now my question is do we need to consider bending moment in the chord?
In my opinion, yes.

But the following would be relevant to how important I feel the failure is:
- accuracy of the rest of the model - loading/load location/restraint conditions/joint arrangement
- connection detail - gusset plates etc.


If it only fails at the connection are we allowed to allow it to fail
In my opinion no, because in a real structure the lateral instability of a hinge would not be acceptable. But failure in the order of 105% could be OK. In the order of 150% would not be.

 

[DaveAtkins]

Normally, I would consider the combined axial force and moment in designing a new truss or analyzing an existing truss.

However, if you can prove to yourself that the truss works with all pinned connections, I am OK with neglecting the bending moment. I believe trusses were designed this way in the past, even with continuous chords.

DaveAtkins

 

[structSU10]

One thing to consider is the effect of your web to chord connections. I will always leave at least 10% sometimes a little more to accommodate member strength reductions due to bolt holes. If the system is fully welded, it may not be needed.

 

[Gile_]

Thank you everyone.
It is an extension to an existing structure.
I don’t fully understand this. My understanding is if the chord is not continuous but pinned at each connection then the truss is still stable if the axial forces are within limit. What is the difference between a hinge at the connection and an actual pin connection?

 

[Once20036]

Historically, SJI advised that you could ignore bending moment in the top chord created by a distributed load (roof deck), as long as the panel points were spaced at 2' or less.
For anything more than 2', the moment needed to be considered.

This was recently re-written such that bending moment always needed to be considered in top chord design.

 

[Gile_]

Thanks. May I ask where I can find this? Which book?

 

[Once20036]

I don't want to retype the language, but take a look at the bottom of page 35, here:

https://steeljoist.org/wp-content/uploads/2021/05/StandardSpecs_K_LH_DLH-Series_0820.pdf

This language previous specifically allowed flexure to be ignored based on panel point spacing.

 

[Gile_]

Thank you. I will have a look and maybe come back with questions

 

[wrantler]

Don't forget to consider the compressive resistance of the sheathing in the top chord.
And also the tension in the gypsum wall board for the tension chord.
edit: oops its steel.

 

[DaveAtkins]

My hunch is it would be easier to make it work if continuity is considered, and pins are not assumed at the panel points.

DaveAtkins