K-Bracing

[joemarch]

Is it a right assumption to model the k-bracing in a frame as tension only members? I've entered a frame into RISA and specified the k-braces to be tension only members. In the results it shows the members taking compression. Should I model the bracing as compression members?

 

[UcfSE]

Open the RISA help file and search for "tension only sections". That should explain why you are getting some compression in your member even though you specified "tension only". Generally I find RISA's help file to be pretty good.

 

[mrMikee]

In general I would not model a k-brace with tension only members. Under a full load condition you would end up with a tension force in one brace and a zero force in what would be the compression brace. This puts a large bending moment in the cross member. Of course the members can be sized for this if that's what you want.

The last time I looked (and I don't use RISA now) I think RISA calculated compression forces in tension only members until a buckling value was achieved. As UcfSE says check the manual, there is an explanation for this.

Regards,
-Mike

 

[DaveAtkins]

mrMikee is correct--I have never liked this, but RISA lets tension only members take compression until they buckle.

DaveAtkins

 

[UcfSE]

The information in the help file also provides some information on how to adjust how much compression the "tension only" member can tolerate by adjusting the unbraced length. I don't like designing tension-only braces either. You don't get a good feel for what is going on the the other brace while the tension brace is deforming.

 

[14159]

Do not use RISA's tension-only members. I talked this over with them a couple of years ago and it seemed like they didn't even have faith in their implementation. RISA is a great program as long as you're doing basic linear-elastic stuff. Anything advanced or nonlinear is beyond the program's real capabilities.

14159

 

[aggman]

Version 5.5 of RISA has true tension only members, Euler buckling members, and compression only members. I have used both types and have checked them on small 2-D frames. The results it gives are good as long as you model the structure correctly. I have had problems with this in RISA before version 5.0 came out.

 

[dbuzz]

I'd model and design them to take the compressive force.

 

[jec67]

Typically, K-braces are designed as taking both tension and compression forces.

K-braces will also take aprtion of the gravity load from the beam/girder to which they are connected. Consult your steel design manual for additional information on this aspect of the design. I do not have my copy in front of me at this time, but I believe Jack McCormick's steel design book has a treatment on this topic. Perhaps it is Blodgett's weld design book??

 

[str334]

To elaborate on aggman's post, the true tension only member in RISA-3D takes NO compression, and the Euler member will take compression up to buckling, at which point it fails, so which to use is up to you.

The tension only feature works just fine in RISA-3D, and to my knowledge there is no lack of confidence in it on Risa's part.

There is no reason to hesitate in using them, assuming, as aggman pointed out, you model your structure correctly, but of course that's true for all software.