Camber in Steel cantilever beam 1

[Tstruct]

I am designing a steel shed at entrance of a building. It consists of 13ft cantilever beams supporting glass floor over it. My question is:
1) To control the deflection of cantilever I want to provide camber, is it practical and common practice to provide camber in steel cantilever beam?
2) The glass resting on this cantilever may get damaged by deflection, the glass supplier doesn't have any technical data. There will be no live load on this glass as it is not accessible except some for occasional cleaning but still this glass can withstand the load of people walking over it. Are such glasses prone to damage due to deflection of the supporting members?

I have no experience with glass flooring.

Thanks.

 

[DrZoidberWoop]

It is not advisable to camber either 1) short beams (under 24' long per AISC recommendation

https://www.aisc.org/globalassets/modern-steel/steelwise/30755_steelwise_camber.pdf

) or 2) cantilevered beams.

The note you would add to your design documentation would be something along the lines of: "Tip of cantilevered beam needs to be preset to +1/4 inch." or something like that. It's a bit of a pain, but fabrication shops know how to account for the slight angle at the moment connection end of the beam (via shimming or conx plate alignment).

If you are talking about a cantilever at the end of a continuous beam, that goes out the window.

 

[271828]

For #1, AISC's DG36 has some text on cantilevers.

For #2, DG3 has some info on glass.

I'd check out both of those to see if they help.

 

[dik]

What is the backspan?

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[canwesteng]

The beam will deflect regardless of how you camber it.

 

[JoshPlumSE]

1) To control the deflection of cantilever I want to provide camber, is it practical and common practice to provide camber in steel cantilever beam?
2) The glass resting on this cantilever may get damaged by deflection, the glass supplier doesn't have any technical data. There will be no live load on this glass as it is not accessible except some for

1) It's certainly not common practice in my experience. However, we did do it for one job I worked on. It was practical for that job because other constraints basically make it a more efficient solution that other possibilities.

2) You make it sound like glass flooring. If so, then the supplier MUST provide some number for their system, or they'd be liable. If's it's a glass cladding system then there is a whole sub-industry built up to handle how these connections should work to accommodate movement.

 

[BridgeSmith]

According to the previous responses, it seems cambering a cantilever beam is not a good idea. If other methods don't work to mitigate/compensate for the deflection (shims, etc.), you could consider a welded plate beam.

 

[hokie66]

As canwesteng said, the beam will deflect the same amount whether it is cambered or not. So what would be the purpose of cambering?

 

[Tomfh]

Don't get too clever. Glass floors on steel beams often fail to fit together properly. Large sheets of trafficable glass, being rigid planar elements, don't tolerate much in the way of geometric error at their supports. If it doesn't go together properly your fancy precambered system will be in the firing line.

 

[DETstru]

canwesteng is right. Camber does nothing for you. Don't bother.

IBC 2403.3 covers vertical glazing supports but you can infer that the same applies to your horizontal glazing. Requires supports to deflect no more than L/175 or 3/4". I would use L/360 at the most though. Given that this is a pretty small structure, you can design it to an even tighter tolerance without much cost, which I recommend.

 

[Tstruct]

Thank you everyone for your very valuable and practical responses, they helped me a lot. I really appreciate it.
@271828 As per DG-3 chapter-3 "DESIGN CONSIDERATIONS RELATIVE TO SKYLIGHTS", it reads that "The movement (deflection) of individual supporting beams and girders should be limited to control movement of the skylight normal to the glass to span divided by 300, to a maximum of 1 in., where span is the span of the supporting beam. The loading for this case includes those loads occurring after the skylight is glazed."
I am a bit confused here about which load to consider? For the deflection of my steel frame, I am considering L/360 limit against dead load (self weight) deflections only, is it ok ? Now according to above qouted text should I consider only live and wind loads and check it against L/300?
The concept which is coming to my mind is deflection of cantilever due to self weight will occur as they are fabricated, now we have a support for skylight to rest on. Any deflection occurring post installation of skylight can only damage the glass. Prior deflection (due to self weight of structure) are not a problem.
Thank you everyone in advance.

 

[canwesteng]

In your case, camber won't serve any purpose. You should include live, wind and the dead load of the glass, plus any dead loads added after the glass

 

[trevorshiloh]

Cambering could actually leave you worse off in this case. Often with cantilevers, instead of natural camber, the steel fabricator wants to just cut the steel beam at an angle or shim it at the moment connection side instead. You might end up with the other edge being level, but the middle will be bowed up enough to affect whatever sensitive material you have on top. It's no different than just not cambering at all, but you'll be misled into thinking you have a lower deflection number than you really do. I agree that in this case cambering is irrelevant to the discussion since you want to limit movement of the floor.