IBC Handrail Loads - Torsion on supporting Channel

[gmoney731]

Hi guys,

I am designing a cantilevered service platform - to support 15 psf DL and 60 psf LL. See attached image for a 2D sketch of the platform.

Per the IBC, we should be able to resist a 50 plf lateral or vertical load (or the 200 lb concentrated equivalent).

I am applying a 0.175 k-ft/ft moment to the channel (based on 50 plf * 3.5' handrail height). The channel support beam for the handrail (also supporting grating above) is laterally braced every 5'.

If I model this channel either as a 10' beam with brace points every 5', or as a full platform, with the unbraced length defined, RAM Elements fails the channel when checking against AISC DG-9.

The results are weird too - for other shapes, reducing the unbraced length clearly reduces torsion. However, for the channel, using the full beam length yields less failure in the model as opposed to specifying a smaller unbraced length.

Can someone give me insight on handrail torsion on channel beams?

 

[dik]

Can the channel web and flange interface accommodate the load in flexure? With an 8" deep channel, you will likely develop 1.5' of the material. Torson from the loading is another issue. A little C4 (not a C8) at the post, back to the interior channel will accommodate this, I suspect if it's an issue. Another heads up, by connecting to the face of the channel, the guard height is increased, and sometimes normal tubing used for guards is a little shy for strength. I often have to increase the tube thickness by a size or two.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Do you feel any better?

-Dik

 

[human909]

Can someone give me insight on handrail torsion on channel beams

I do this sort of thing all the time and I don't have the challenges you describe. I'd normally expect deflection and not strength to govern. The member sizes you are using seem adequate for the design.

RAM Elements fails the channel when checking against AISC DG-9

You don't adequately explain HOW it fails which is the key part of your question. Is the strength or deflection? Is this under full LL? What degree of rotation are you experiencing of the beam and is it realistic? Your computer model might be outputting a failure code but does that actually match reality.

Also if I was designing this I'd be having the channel flush with the beam and not continuous across it. It will provide better twist restraint on the channel by supporting it from the web and not the flange.

If I point a point moment load into some of my computer models I get some crazy torsional results too. But most computer models don't adequate account for all elements offering torsional restraint.