Lateral Restraint to Steel Beams

[Cheers]

Can anyone offer any advice or give references to the degree of lateral restraint afforded to steel beams with open grid flooring fixed to top flange. My initial thoughts are that fixings to the top flange would be capable of providing the lateral restraint, But would connections between the flooring panels be sufficient to develop the required shear for diaphragm action ?

 

[phuduhudu]

The codes usually specify the force required to be resisted by lateral restraints. If it is your top flange in compression then it can be restrained by flooring fixed to the top. You just need to check those shear forces through the panel connections but I would have thought it should be possible. BS5950 specifies the lateral restraint force as 2.5% of the factored force in the compression flange. This can be divided over the length of the flange in compression. Carl Bauer

http://www.bauerconsultbotswana.com

 

[ishvaaag]

Yes, if the floor has some concrete and is properly held to vertical weight or forces, checking the ability to pass the forces. Another question recently asked the same of sheetmetal only bracing, on which the

Guide to Stability Design Criteria for Metal Structures
5th ed
Galambos

has an empirical formulation but needs the effective shear rigidity of the sheet contraption as data, something I have not found anywhere (maybe my PTI manual has something on it whilst dealing with sheet metal roofs).

Of course I wouldn't rely in such bracing for a megastructure member.

In any case is very likely that most well designed even sheet metal would be enough to brace compressive sides.

Respect bracing bottom flanges of continuous beams in compression you brace them directly. This notwithstanding I have one excel readable sheet that says if it can be continuously braced by the sheer web (sorry, no trusses) in the assumption of the slab above being practically rigid to rotation displacemet (which will be for many partial encasements or reasonably thick slabs atop).

 

[sbp]

what if i have a beam consisting of two channels spaced apart about 300 mm with intermittant plates welded at top and bottom? The beam is laterally supported at the ends. Now what shoud be the effective length of compression flange for the calculation of allowable flextural stress?

I think as long as the plate connecting compression flange can take 2.5 % of force in comp flange, the effective length can be taken as c/c distance between the plates and individual beam can be designed. Is this sufficient or the length between the end supports shall be taken for the "combined" section for finding the permissible flexural stress?

thanx.

 

[ishvaaag]

In my personal case and since we have a tolerant code for design of columns built of 2 Cs with battens, I would use the whole member with N=0 if for flexure (I have a mathcad sheet at least for this, maybe 2). Hence I would only need to acknowledge external bracing, akin to that of ends.

Your view I see at least critcable in some aspects. 1) Not stating a maximum distance between battens, you can put 2 or 60 battens and obviously to the component chords it won't be the same; the specification of maximum separations for columns ensure the individual chords won't buckle ever before the built-up member. Then one member braced to another of equal stiffenss (and once joined so will behave) is within the reach of parallel sidewise buckling, obviously for a Lb bigger than the distance between battens. The main conceptual failure here is that the bracing forces even if weak need be effectively passed fo fixed points. Anyway relative bracing (which yours is) is used in the cross braced sections in bridges made of parallel stringers. Just 2 chords or stringers can make the likelihood of the initial imperfections be in the same direction and so I don't recommend it.

Furthermore the required stiffness in any bracing is that at the brace point.

In any case, the built-up member can be checked as a beam with the continuous bracing as per Galambos, but again the effective shear modulus of the continuous restraint is needed. Discrete bracing would have clear distance to use and I would make it to coincide with a batten.

 

[ishvaaag]

In my personal case and since we have a tolerant code for design of columns built of 2 Cs with battens, I would use the whole member with N=0 if for flexure (I have a mathcad sheet at least for this, maybe 2). Hence I would only need to acknowledge external bracing, akin to that of ends.

Your view I see at least criticable in some aspects. Not stating a maximum distance between battens, you can put 2 or 60 battens and obviously to the component chords it won't be the same; the specification of maximum separations for columns ensure the individual chords won't buckle ever before the built-up member. Then one member braced to another of equal stiffness (and once joined so will behave) is within the reach of parallel sidewise buckling, obviously for a Lb bigger than the distance between battens. The main conceptual failure here is that the bracing forces even if weak need be effectively passed fo fixed points. Anyway relative bracing (which yours is) is used in the cross braced sections in bridges made of parallel stringers. Just 2 chords or stringers can make the likelihood of the initial imperfections be in the same direction and so I don't recommend it.

Furthermore the required stiffness in any bracing is that at the brace point.

In any case, the built-up member can be checked as a beam with the continuous bracing as per Galambos, but again the effective shear modulus of the continuous restraint is needed. Discrete bracing would have clear distance to use and I would make it to coincide with a batten.