What is the the ACTUAL method of analyzing trusses?

[LockeBT]

In almost every single structural analysis books I've read, truss analysis is performed under the conditions of "ideal trusses":

1) Distributed forces along the member are neglected and forces are summed up and applied as point loads at the nodes/connections.
2) Members are only designed for axial loads (tension or compression).
3) Nodes/connections are considered to be pinned.

I'm ok with the 3rd condition since it's more conservative than assuming that there are some fixity at the nodes/connections. No worries there.

However, let's focus on the 1st and 2nd conditions. It's a bit counter-intuitive for me given how that's not a conservative approach. A member subjected to combined flexural and axial loading (mainly compression) has reduced capacity because it has to satisfy the unity equation. Even if the member is fully braced along its compression edge I still have such reduced capacity that I will need to upsize said member compared to the it being designed for axial load alone.

I have always analyzed the truss as a whole (using the ideal truss method) to figure out the axial forces in each member. Then go back and apply distributed loadings such as wind, dead and live loads on said members (especially the ones in compression) and check it for flexure + compression. It just makes more sense to me. And more often than not I end up using a larger member.

Does anyone else do this?

 

[jayrod12]

Thats how I do it for the loaded chords. However I don't too much worry about bending due to self weight in the other members. Meaning, I generally check the top chord for combined axial and bending (the odd bottom chord as well if having significant loading hung from it).

 

[hokie66]

I agree with both of you. Sometimes, a truss is loaded at the joints. Then, IMHO only axial load need be considered. Self weight of the members is inconsequential for local bending.

 

[LockeBT]

hokie66,

Thank you for the confirmation. Asides from large steel trusses used for bridges, what other applications have you seen trusses loaded at the joints? For me personally it's quite rare. I'm done mostly residential and commercial work. The rare occasions of me doing industrial work I still find trusses not loaded ideally.

jayrod12,

Thank you for the input. That confirms it for me. I feel better about my engineering judgement now lol.

 

[Celt83]

Be careful “ I'm ok with the 3rd condition since it's more conservative than assuming that there are some fixity at the nodes/connections. No worries there.” this is not always true. Moment reversal due to fixity could result in lateral-torsional buckling limit states with much longer unbraced lengths than the pinned joint case.

With most of today’s software it is fairly trivial to consider both the member self weights and between joint loading.

My Personal Open Source Structural Applications:

https://github.com/buddyd16/Structural-Engineering

Open Source Structural GitHub Group:

https://github.com/open-struct-engineer

 

[LockeBT]

I have just been trained on this new software for transmission tower design called PLS-TOWER. It's a very niche program for structural engineering in the utility sector. Anyhow the program does not consider between-joint loading (mostly from wind load) for the members. It follows strictly the ideal-truss analysis.

That made me a bit uncomfortable which stemmed my further research into this. Just based on my observations:

1) Tower members are mostly angles which inherently do not have good flexural capacity.
2) Tower members at the base see huge axial load and have longer unbraced lengths. And even if they are braced, they're not braced in the direction of the principal axis (the weakest axis), rather they are braced in the other 2 stronger axes which don't govern the design anyway. Any introduction to flexural forces on top of the axial force will easily fail the member in unity. The program ignores that.


Oh boy that's something I did not consider. Which now makes the "ideal truss" approach even worse. It's too...ideal lol.

 

[Celt83]

For a transmission towers depending on location ice loading along members could be fairly significant.

My Personal Open Source Structural Applications:

https://github.com/buddyd16/Structural-Engineering

Open Source Structural GitHub Group:

https://github.com/open-struct-engineer

 

[phamENG]

Locke - I think you have the right idea in general. The ideal truss analysis is only one piece of the puzzle. As long as you don't have big gusset plates with lots of bolts for your web members, you're unlikely to have enough fixity to really worry about it. Celt is right, of course, that you need to at least have it in the back of your mind to recognize when it does matter.

For trusses loaded only at panel points/nodes: lots of large and widely spaced roof truss and purlin applications (think churches and other wide open spaces), open web steel joists systems with truss girders (the truss girder will be laid out so the individual joists land on panel points of the truss girder), most wood truss girders, etc.