Bracing Design

[dccd]

I read it somewhere that the X bracing is only designed to take tension not compression why ? I knew that when the steel is subjected to compression, it tends to buckle, so the steel will be generally weaker when it's subjected to compression especially when the span is long. But X bracing is only designed to take not designed to take compression ??

 

[rather_be_riding]

There's no requirement that you have to design cross bracing to only take tension; however, due to the space requirements and the fact that each brace will handle movement in one direction in tension, it's often most convenient to do so. Quite small rods and plates can handle the bracing loads in tension. Once you decide to try to add compressive capacity to these members, you usually end up with a significantly larger member to avoid buckling. As such, it's simply more efficient in many cases to design tension only bracing members.

For example, it might not be uncommon to see a cross brace made of 20 mm rod (or M20 threaded rod even) and the length of each bracing member might be 6 m or even greater. The compression capacity of a 20 mm rod at such a length is essentially nothing at all and yet it still boasts a reasonable tensile capacity. Two opposed (X) braces will form a satisfactory system where a single member would need to be quite a bit larger to offer any compression capability.

 

[jimstructures]

dccd,

Properly designed X bracing can handle either tension only bracing (think PEMB buildings, either rod or cable) or tension & compression bracing ( think heavier Crane bracing or heavier mainframe building / seismic bracing), either angle, I section or tube (rectangular or pipe section).

Jim

See many AISC examples.

 

[OldDawgNewTricks]

Tension-only bracing is economical for the reasons mentioned. But it is not allowed for high seismic design in tall/heavy/important buildings where better performance is required.

 

[SteynvW]

Just remember as well that if the bracing does not extend from the ground (anchorage point) al the way to the top with one diagonal, other members have to be able
to handle compression. For example, with roof bracing the purlins have to be able to take the compression or another member added for these forces.

 

[ldeem]

Main point is to be consistent with your analysis method and design method. So if you design X-brace tension only then make sure our analysis model is also set up for tension only.

There are various schools of thought on compression buckling lengths for X-bracing. In my work (heavy industrial / mining) I assume full length of the brace as buckling length out of plane and 1/2 the length for buckling in plane - assuming braces are connected in the middle.

Lastly I have been moving toward single leg bracing using pipes to take both tension and compression. I have found many places where bracing has been replaced due to corrosion or damage but the intermediate connectors are never installed.

 

[human909]

Lastly I have been moving toward single leg bracing using pipes to take both tension and compression. I have found many places where bracing has been replaced due to corrosion or damage but the intermediate connectors are never installed.

I've been doing the same for similar sort of construction. Though more ease of modelling, ease of locating bracing around equipment and consistency.

While fabrication costs of a HSS member is significantly more this is partially or mostly offset by using half as many bracing members this cuts down on installation costs. In general if I try to use cross bracing I end up having to use HSS in several areas any way. I find using HSS just makes the whole design and installation process easier for all including the engineers, steel detailers and installers.

Tensioned roof rod bracing still makes sense for portal frame constructions. But I don't design too many of those.