Steel beam top flange lateral restraint by trapezoidal sheeting

[mikesg]

Dear colleagues,

I am working on a steel project and am struggling with the recurring question whether (and when) the roof sheeting provides lateral restraint to the top flange of the purlins when calculating for gravity loads assuming the purlins work as single-span pinned beams.
Our code contains a statement that they do without providing a way to actally calculate this based on the properties of the beams and the sheeting. Russinan SNIP for steel structures also states this. Eurocode3 on the other hand provides formulae to calculate the required and provided strength and stiffness (EC1993-3-1, pt 10.1) based on the geometry of the roof, sheeting and beam section. However if I check the sections and distances between purlins that are typically accepted in our practice it turns out that the purlins cannot be considered as fully restrained. Especially if the sections are hot rolled and taller than 140mm (5.5") and the sheeting is more than 40mm(1.6") high. Thin and torsionally weak cold formed sections pass the check very well, but we don't always use such.

Should I turn everything upside down and tell my colleagues, including the supervising engineer that we've been all wrong till now? That the code misleads us? I think no, because the facts prove that buildings designed this way are still functioning and have proven to be able to carry heavy snow close to the design load values. On the other hand, I don't feel it is correct to simply accept that everything is fine, always, for any size of beam with any type and size of sheeting - the correct thing would be to make a calculation to check.

Now the question - what is your practice in this matter, how do you check whether the purlins can be designed as laterally supported for gravity load? Can you provide me with pointers (from your code, or some article, etc.) containing formulae and recommendations?

Thanks for your valuable input!
Mike

 

[JoshPlumSE]

Absolutely tell your colleagues. However, it's not that you've been doing it WRONG all these years. But, that the other code offers some additional guidance. They (and you) can use that additional guidance to justify cases where your engineering judgment is telling you to do something a bit more conservative than your code suggests.

My take on it is more along the lines of:
1. Should those non-complying situations be considered FULLY braced? Probably not.
2. Should those situations be considered totally unbraced? You could. But, that may be overly conservative.
3. Can you use the Eurocode for determining when you should use an unbraced length greater than zero, but less than the full length of the member? Yes, it may not tell you exactly what to do. But, the code isn't a cookie recipe. This is what engineering judgment is all about.

 

[mikesg]

>the code isn't a cookie recipe
Very well said, and fortunately it is not a recipe, it would not be interesting otherwise

> Can you use the Eurocode for determining when you should use an unbraced length greater than zero, but less than the full length of the member?
A good pointer. I will research in this direction and write here the result if I find something worth it.

Thanks!
Mike

 

[chicopee]

For what it's worth, I would not rely on roof sheething for lateral restrain.

 

[JAE]

I know that all pre-manufactured metal buildings certainly depend on their roof panels to brace the roof purlins (usually Z shapes).

However, these buildings are the first to collapse during our mid-western USA severe storms and tornadoes.

 

[DST148]

@mikesg: AISC 360-10, Appendix 6 provides guidelines on strength and stiffness requirements for beam bracing.
We typically use 1 1/2" / 3" deep ribbed metal deck on hot rolled Wide flange/Channel shapes or open web steel joists. We not only count on the deck for providing bracing for the purlins / beams subjected to gravity loading but also as diaphragm for lateral loading. The deck manufacturers provide allowable shear values based on the depth / thickness / span(of the deck) , and the spacing of the fasteners attaching deck to the structural member. A lot depends on the magnitude of the gravity loading and the span of the purlins / beams.

 

[paddingtongreen]

Doesn't the bottom flange control in the wind uplift condition? I used to find it so. The effective length of the top may be reduced by the deck but the bottom is not.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[mikesg]

@JAE: I agree that sheeting can brace Z shapes (as well as other cold formed thin wall sections), and calcs support this. But hot rolled shapes are a problem definitely

@DST148: If I understand correctly, I should use formulae A-6-9 and A-6-10 modified as per point 6.4.2b for continuous bracing and compare the required strength/stiffness to provided ones?

@paddingtongreen: It is close, but does not control in our usual cases. Snow is quite heavy here and projects are in terrains with obstacles.

Many thanks to all of you for your support and for the information provided!
Mike

 

[ishvaaag]

One of the keys of structural survivability is keeping extant the projected shape of the structural items. JAE points to cases where structures where the roof sheets are used to brace use to fail first in severe thunderstorm. The question is that at that moment the roof sheets are not acting estabilizing the flange, but destabilizing it. Assymmetry of the loading pulls the flanges out of shape to cause LTB ... hence in my view the more substantial the braces and far from this kind of potential behavior, the better.