unbraced length 1

[Cahaya]

I design equipment with plate and C channel stiffeners welded to the plate to resist internal pressure loads, ductwork for power plants, to be specific.

If the free flange, away from the plate, is in compression, what is the maximum length for the channel to run without cross bracing if it is designed to a maximum stress of .6Fy in the composite shape? What is the relationship of allowable stress to unbraced length when length is greater than the first case?

 

[fegenbush]

Please see equation F2-5 of AISC 360. This will tell you the maximum unbraced length for your design.

The spec can be downloaded for free at

http://www.aisc.org/2010spec

 

[Cahaya]

Thank you for the response. But I don't believe this case is for a channel welded to a plate. It seems to apply to channels and symmetric shapes bending about their strong axis only. In the case I stated one flange is restrained against translation and to some degree against rotation.

Did I read it incorrectly?

 

[fegenbush]

My error, I had forgotten that you were using a channel. I do not recommend this as an efficient shape for your application. My preference is to use a WT or ST shape.

If you insist on using a channel, chapter F12 is applicable. I will summarize for you: Determine buckling stress by analysis. In the commentary, it is recommended to use Galambos Guide to Stability Design to calculate the buckling stress (6th edition free if you are an AISC member) or to use one of the many choices given in the previous parts of Chapter F.

 

[Cahaya]

That section, too, does not apply to channels on plates. It's for asymetric shapes.

 

[fegenbush]

In what way is the combined section of a channel on a plate symmetric?

If the shape is built up from plate or hot rolled, the stability equations are the same provided that the two members are adequately connected. If they are not adequately connected, they act independently, and a previous section of Chapter F applies.

If there is some way that I am not understanding to connect a shell and a plate, please provide a sketch.

 

[Cahaya]

It's not symmetric but its not free to bend in the weak axis as the plate restrains translation along its plane and provides some restraint against rotation of the bottom flange.

 

[fegenbush]

I think I see your problem. I will restate the question and you can determine if that is what your asking.

"I have a member that is built from a channel and plate (see attached picture). The tension flange is braced by welding to the plate, but the compression flange is free. Since the plate welded to the tension flange produces some resistance to rotation of the cross section, how would I determine the maximum unbraced length of the combined section? AISC only shows the unbraced or fully braced condition and this is one of intermediate fixity."

If you are using the properties of the combined section, you should not consider the plate as stability bracing. It is either part of the section or part of the bracing, but not both. If you are using only the properties of the channel, there may be some resistance to buckling due to the plate, but the conservative approach is to neglect this effect and add bracing as necessary since the bracing is connected to the tension not the compression flange.

 

[Cahaya]

Add bracing as necessary, certainly. The question is how to determine what 'as necessary' means. And of course the plate both stiffens the channel against bending and provides lateral stability. Code suggests 32 thickness of the plate as the element to be added for calculating stiffness about the strong axis. But how to determine what length of the top flange can be sustained and at what stress level. That is the question.

 

[WillisV]

This limit state is referred to as constrained-axis flexural torsional buckling. It is not covered in the AISC Specification. You can find general information on it in Timoshinko and Gere or more recently it is covered in AISC Design Guide 25 and in the 2nd ed. of the AISC Seismic Manual (in the collector beam section).