Inquiry about the theory of Tension-Only Bracing

[MrFurleyEIT]

I've got a question about the theory of bracing for steel structures, primarily vertical bracing but it applies to horizontal bracing also. Bracing primarily provides a load path for the lateral loads to be transferred to the foundation by being able to take tension and compression, right?

It seems that it is popular to design the bracing as tension-only bracing because the member sizes will apparently be smaller and lighter, than if it were to be designed as taking both tension and compression. Is this fundamentally true?

The compression has to go somewhere, though, doesn't it? And if not to the bracing - when the bracing are designed as tension-only bracing - then to the main framing members, such as columns or beams. Is this a case of "robbing Peter to pay Paul"? That is, my bracing sizes may be smaller because they can take only tension, but my other framing members might have to be larger.

Also, are there cases where it is more economical to design bracing as taking both tension and compression?

Hope someone experienced in the design of bracing members (bracing engineers, anyone?) provide some feedback. All are welcome to answer, though. Thanks a bunch.

 

[JAE]

Many times a diagonal brace extends across a longer distance and has no intermediate roof members, deck, or girts to brace it as does a beam or column. Thus, the brace tends to be much more susceptible to compression forces as its unbraced length is large.

Yes, the tension only brace does introduce extra force responses into the column/beam system around the brace.

In low seismic areas, tension only bracing is common. In areas where design and detailing must follow the AISC Seismic criteria, tension only bracing is prohibited.

 

[SHBH]

JAE is correct. Tension only bracings are discouraged in high seismic zones. Reason being;while under compression they tend to buckle and under reverse load effect when they experience tension, sudden snapping at the bracing junction occurs which eventually lead to the failure of joint under fewer load cycles.
But if your concern is to use tension only bracings in normal cased (low seismic zones) then i don;t see any problem, except the fact that they tend to exert more compression on beams and columns at the junction. But in this case try to use more slender members so that during compression they can easily buckle , otherwise in case of stocky member they may tend to resist the compression and in case of high compression permanent yeilding may occur which eventually lead to the failure of these members and subsequently leaving these members unsuitable to take any further tension.

 

[MrFurleyEIT]

Thanks for the quick responses, JAE and SHBH. Now, if I may expand my question a little bit. There is also the case of compression-only bracing, or struts. These are apparently very common in steel structures. Under what conditions do we specify these compression-only bracing, and is it possible to design these same beams for compression and tension, or is redundant? Once again, thank you so much for all your responses and please keep them coming.

 

[MrFurleyEIT]

While I await responses from my previously posed questions, I have got another questin about determining the area required, A for a bracing member. Say, I have a tension-only bracing member with an axial load of T. Allowable stress is 0.6Fy. Now the area required for the bracing member will be:

A = T/(Allowable Stress * 0.85)

What is the 0.85 factor in the denominator of the above equation?

Another question is that since my bracing members are all tension-only members, is it correct to say that the slenderness ratio (KL/r for tension members = 300 max) is not to be considered in tension-only bracing. Thanks.

 

[Lion06]

the kl/r max of 300 is only a suggestion, not a requirement.

 

[hokie66]

I take it your .85 factor is a capacity reduction factor.

Generally, compression braces or struts can take tension as well.

 

[nutte]

StructuralEIT is right that the kL/r max of 300 is only a recommendation. However, it's coming from the industry experts, so I would call it a good one and adhere to it if at all possible.

Where did your 0.85 come from? For ASD, the required area would be T/(0.6Fy). There is no need for an extra 0.85 reduction.

 

[csd72]

Is this equation for angle bracing.

If so, then the 0.85 factor may be to allow for attachment to one leg only and the uneven stress distribution resulting from that.

The Australian codee uses this factor, I cant recall seeing it in US codes though.

csd

 

[Lion06]

csd may be right about the 0.85 factor. This could just be a factor similar to "U" for the effective area of a tension member (page 5-34 in the green book).