Analysis of steel truss under Uniform Load 2

[flamby]

I am analysing a truss under UDL. Fairly simple truss but one peer said since the loads are not acting at the pinned joints, bending moment will be induced in the members and these bending stresses should add up to the axial stresses.

One way to get out of this dilemma is I design the whole thing as plane frame since the joint are actually going to be welded. But I want some more thoughts on this from you guys.

Thanks.

 

[Ron]

Flame...not all situations fit the "mold" of typical analysis. You can do a couple of things to ease your mind:

1. Resolve the UDL as point loads applied at the panel points of the truss. Analyze as a truss.
2. Analyze entire truss as connected beams (inducing bending from the UDL...plane frame analogy).

Compare your results.

In reality, the true response of the system will likely be somewhere between these two approaches.

We usually model trusses as having only an axial load capability in the members. This is generally erroneous, particularly with welded trusses, as there is a bending component induced due to the fixity of the joints. The true picture is further complicated by the fact that you must consider, once deflection starts, that you now have moving reference points thus affecting rotation about those points (the truss connections move globally, while the stresses are reacting locally!)

 

[dik]

If the truss under UDL is what we consider as a joist, then the Canadian code does not require consideration of moment for the top chord if the panel points are less than 2'.

For smaller 'trusses' the redundencies are so great that the secondary effects from bending become less significant. Without quantifying it, the more redundencies you put into a system, generally, the stronger it gets. Although the web members have small moments at each end, the effective length of them is reduced, increasing the compressive resistance. If you're looking at a large truss, then I would consider the moments.

Catch the following parts of S16.1 (Canadian Code):

Open-web steel joists shall be designed for loads acting in the plane of the joist applied to the compression chord, which is assumed to be prevented from lateral buckling by the deck. For the purpose of determining axial forces in all members, the loads may be replaced by statically equivalent loads applied at the panel points.

The compression chord shall be continuous and may be designed for axial compressive force alone when the panel length does not exceed 610 mm, when concentrated loads are not applied between the panel points, and when not subject to eccentricities in excess of those permitted under Clause 16.6.11.4. When the panel length exceeds 610 mm, the compression chord shall be designed as a continuous member subject to combined axial and bending forces.

The code also requires provision of a point load for floor and roof joists.

 

[flamby]

Thankyou Ron and dik for your helpful comments. Since the panel I am doing are 1600mm, I have to consider moments as I do not dream of any country-code allowing me not to consider it.

I was wondering if there was a software which can analyse a truss and yet consider bending on the top/compression chord. I feel much work lies ahead for me as I have several trusses of different spans and loadings.

I tried the method suggested by Ron and my first assessment is that the plame frame method is very conservative to the point of being expensive. I have to examine it in much detail though..

Thanks again..