Allowable beam rotation

[stl001]

Can anyone provide some guidance on allowable rotations of W-sections for servicability considerations?

I have an architecturally finished wall section being supported by the W-sections. However, the wall is two feet off the centerline of the beam, and is creating a substantial torsional moment. The wall finish is sensitive to deformation and the beams cannot be laterally braced.

Any guidance on this issue would be greatly appreciated.

 

[csd72]

Need more information. Is the wall parallel to the beams, what is it supported by? Why is the beam in torsion, is there a cantilever?

 

[jike]

Perhaps more description might help us!

 

[stl001]

The wall is running parallel to the beam. The wall is supported by a canilever system attached to the top flange of the W-section. The distance between the centerline of the beam and the center of mass of the wall/wall support system is two feet.

 

[jike]

What type of wall? What type of deck? Weight of the wall on the cantilever? If you can't kick back to the deck or another beam can you add framing to kick back to?

 

[stl001]

The structure is a large four story atrium-like entrance, but with minimal windows. There is no decking, ect. to tie it back to and no additional framing can be added. The cantilever-wall system is to be cold form shapes with an architectural finish on the cold-form wall frame. The finish weighs 18 psf on beam spacings of ten feet. Beam spans are twenty feet and column/wall placements cannot be altered due to the retrofit nature of the project.

 

[Lion06]

Do you HAVE to use a WF? I would try to use an HSS.

 

[jike]

I would either use an HSS or reinforce the WF with side plates to make a closed section (tube).

 

[DaveAtkins]

I agree that a tube would be a better solution here.

But, back to your original question--the serviceability issue is with items ATTACHED to the beam, and not the beam itself. For example, what does the beam rotation do to the wall system? A small rotation in the beam could result in a large deflection in the wall system. The wall system will rotate with the beam, where it is attached to the beam.

DaveAtkins

 

[stl001]

Right. I was informed that the wall finish would be very rigid, but could crack due to any substantial deformations.
I have taken care of local deflections on the wall.
I was worried about the beam rotations because any differential in rotation from beam to beam could cause deformations. I had hoped to counteract this by limiting beam sizes, but I am limited with how large a beam I can use.

 

[stl001]

I actually meant to write limiting beam rotations, not limiting beam sizes.

 

[csd72]

I am still not 100% sure how the two beams fit in, are the outriggers propped cantilevers(i.e a cantilever with a backspan)?

Anyhow with this limited understanding in mind I have two comments:

1. If you are concerned about the rotation of the beam then the beam probably doesnt work strength wise anyhow. Double check your numbers.

2. how are the beams restrained at the ends against torsion?

In this type of situation, I would usually weld an angle to the opposite side from the cantilever to take out the torsion from the cantilever. If the beam is not restrained against torsion at the ends then it will rotate a little to the point that there is negligible torsion in the beam and the full moment goes into the backspan.

Also be aware that this type of detail will amplify the effective load on the nearest beam.

As you have correctly realised, it is the deflection of the line of support directly under the wall that matters. Calculate the maximum allowable deflection at this point as if this line was a beam, then compare your actual deflection from combined cantilever, beam and rotation deflection to this number.

csd

 

[shin25]

If you use HSS, can you bolt it down some how to eliminate rotation?

 

[msquared48]

If you cannot add purlins to laterally brace the beam, why not upset the beam in question and create a moment connection at the juncture of the steel beams that support the wall? These beams would then serve to help take out the torsion imparted to the beam in question. Having the beams the same size at the intersection would help too.

Is this possible?

Mike McCann
McCann Engineering

 

[UcfSE]

I would calculate the amount of deflection the wall could tolerate if it were supported by a beam under its center. Use this to calculate the amount of rotation in the beam that would cause this maximum deflection.

I would suggest something like Mr. McCann. Instead of a cantilever hanging off of one side of the beam, provide beams on the opposite side of the main support to turn the cantilever beams into overhangs. If you can't have those bear on the supporting wide flange, then frame them into the wide flange each side and provide a continuous splice connection.

 

[JAE]

Yes - WF sections have very little stiffness rotationally.

Very little.

I think they have very little stiffness....

(hint hint) Do what these excellent posters say above and use an HSS. If your wall is brittle as you say, you'll regret using a WF.

 

[stl001]

I would like to thank everyone for their advice and comments. You guys have been very helpful.

We obtained allowable local deflections from the architect and from there determined allowable rotations of the W-shape.

W-shapes were used because the owner wished that the new additional structural system (which would be partially exposed) match the existing system (W-shapes).

We had beam to column moment connections, but were still getting minor (<2degree) rotations at mid-span which were still above the allowable rotation determined above. This was counter-acted by selecting the appropriate (if slightly larger) shape and adding stiffeners at quarterpoints along the beam.

 

[UcfSE]

What kind of stiffeners?

 

[hokie66]

The stiffeners will just go along for the ride. You have been given good advice that you MUST use a closed section, and if you do not, nobody is going to be happy with the end result.

 

[Lion06]

hokie66 said it. The stiffeners will not help you. Why would they? Unless the beam is bolted to some sort of support at the stiffener locations, they aren't going to do much.