Effective Length of cross braces

[BAGW]

Hi All,

I am designing cross bracing consisting of stocky tubes say 10x10. They have been analyzed as compression and tension braces. When they are subjected to lateral loads, one brace will be in tension and the other in compression. In such scenario, it can be assumed that the tension member braces the compression member in and out-of plane direction. That is for the compression member half the unbraced length can be used to calculate the compression capacity.

Lets consider a scenario where both braces are in compression (one brace experiencing high compressive force and the other one a smaller force), say under gravity loads. Both the braces have similar length and similar cross sectional properties. In this case, I was in the impression that the braces need to be designed based on full length of the member. But, I was told that, even when both the members are in compression, it can be assumed that they brace each other in in-plane and out-of plane directions. This is because, when the brace with higher compressive load starts to buckle, the one with the smaller force goes into tension. Is it correct to assume the half the unbarced length for the design in such circumstances?

 

[sekwahrovert]

If both braces are in compression, I would not consider them able to brace each other.

I don't know what you're designing, but If it's anything like the towers that we design where I work, it's hard to imagine a controlling load combination in which both braces would be in compression.

 

[BAGW]

[I don't know what you're designing, but If it's anything like the towers that we design where I work, it's hard to imagine a controlling load combination in which both braces would be in compression. ]

I completely agree with you. I case of tower design, under all loading conditions, whether gravity combinations or lateral combinations, the braces will be in compression and tension. I am talking about horizontal lacing in industrial buildings comprising of trusses.

 

[JAE]

Check out this thread (and the referenced paper)
thread507-396523

Also these:
thread726-120298

thread507-258454



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[CARunderscore]

Are the members physically connected where they cross? I would not even begin to think about considering the members as bracing each other if they aren't... they'd still be completely free to buckled in 3 different directions. And even if they are connected, I still don't like the idea.

Could you please elaborate on the situation a bit more? What is the actual physical length of the tubes? What are the members are you lacing together? It sounds like you're talking about horizontal trusses with cross-braced panels, but I'm not entirely sure. Given the size of the members you're worried about effective length for, I suspect you might be talking about bracing for roof trusses or something. A sketch would be nice.

Really, it kind of sounds like you're giving a garbled description of the tension-only member assumption for cross-bracing. Just to make sure, you aren't either:

1. misunderstanding the concept of tension-only cross-bracing or
2. getting an explanation of why cross-bracing is often slender for compression from another engineer who might be misunderstanding it,

...are you? You say you're assuming they can take compression, but the wording of your last sentence... I don't know.

 

[CARunderscore]

Check out this thread (and the referenced paper)

Ah, so allowance can be made if they're connected. I definitely like the idea of using the ~0.7 values rather than 0.5.

 

[BAGW]

Could you please elaborate on the situation a bit more? What is the actual physical length of the tubes? What are the members are you lacing together? It sounds like you're talking about horizontal trusses with cross-braced panels, but I'm not entirely sure. Given the size of the members you're worried about effective length for, I suspect you might be talking about bracing for roof trusses or something. A sketch would be nice.

Yes, members are physically connected at the midpoint. They are connected using the knife plate detail.

The length of the tubes is around 30’. The members that are laced together are HSS10x10x8 tubes.

Attached is the sketch. Correct, I am talking about bracing at the roof (diaphragm). These are not tension only bracing. We have developed a 3D model for the analysis. We see compression forces in both the braces under gravity loading scenario in the top chord level. I don’t know how true this behavior is. In these cases, we have assumed that the unbraced length of the members is half the full length for the design.

 

[ckrkarim]

Be it vertical or horizontal braces, I normally don't consider that they carry any vertical loads. For cross braces like yours, the best bet would be to consider effective length factor 1.0, however, I have been through many situations where I considered effective length factor 1.0, 0.7 and 0.5. Its all about engineering judgment and I guess you can choose an effective length factor for cross braces somewhere between 0.5 to 1.0 and most of the time I choose 0.7.

 

[Lomarandil]

For cross bracing members of that length, judgement is definitely required -- or use of the AISC paper mentioned. The 0.7 factor I mentioned for lacing members may still be appropriate, but I'd imagine the stiffness of your tension brace is significantly less over that length compared to a 3' long lacing bar.