Angle of twist - Torsion

[StrP88]

I have a Channel section and I am trying to check it for torsion. It is simply supported.
I noticed the spread sheets out there calculate the angle of twist but dont know this number has to be check about what number? Where can I find allowable angle of twist?
If is not needed then why these spread sheets calculate it?

If more infor is need please ask and I will provide

Thanks

 

[KootK]

How much twist is acceptable is a decision that you, as the designer, get to make based on your assessment of what is important about the situation being considered. It's rare that anyone actually cares about twist itself. However, if something is attached to the twisting member, that twist may well result in transnational movement of the attached member that could cause serviceability issues.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[StrP88]

KootK (Structural),
May I ask this angle of twist we measure for a channel section is about the ceneter of gravity or it is amount of twist about shear center?

 

[jimstructures]

I have used the angle of twist around the centroid of the cross section. Be aware that the number you receive may not be a number that you can translate easily (without experience) into real world effects. You also need to calculate the physical displacements at the extreme fiber, i.e. top and bottom of the cross section. If you don't have an understandable amount of displacement you may not be aware of the real world effect of the twist.

Jim

 

[rb1957]

if the beam is truly simply supported, then you're got a problem 'cause there's nothing to react the torsion ... unless you have applied equal and opposite torques to the beam.

can you close the channel with a over plate (and make a tube out of it) ?

consider deflection ... the allowable deflection is based on how springy the people in the building will perceive it to be. torsional deflection is bounded by the same limits ... what is perceptible ? what messes with the operation of your installation ?

another day in paradise, or is paradise one day closer ?

 

[BAretired]

I agree with rb1957. The beam must be fixed against torsional rotation at not less than one point along the length of the beam; it cannot be simply supported.

Also, it makes no difference about which point you measure the angle of twist. It is the amount the cross section has rotated from its original position and has nothing to do with center of gravity, shear center or any other point on the cross section.

BA

 

[XR250]

Channels are very un-stiff in torsion. Boxing it would make it significantly better, if needed.

 

[Teguci]

I have a Channel section and I am trying to check it for torsion.

The best way to check a channel for torsion is to redistribute the loads so that the torsion on the channel = 0 < the torsion allowed on the channel.

It is simply supported.

When dealing with torsion, this typically means that the pinned connection can take the twisting motion but is allowing the channel flanges to warp (top flange rotates opposite direction from the bottom flange). Verify that the shear plate, or whatever is providing your connection, can take the torsion reaction.

I noticed the spread sheets out there calculate the angle of twist but dont know this number has to be check about what number? Where can I find allowable angle of twist?

If you can't deal with torsion on a channel the proper way (see above), then you will need to do a stress check on your section for the added normal stresses due to warping and shear stresses due to pure torsion. Sharpen your pencil and get your difEQ book out because the distribution of pure torsion to warping is fairly involved. Of course, you could grab a copy of AISC Steel Design Guide 9 and match up your conditions with one of the lookup charts in the back. I would recommend that you not use an excel spreadsheet unless you develop it yourself.

As for allowable twist, if this a serviceability condition, don't use a channel. I calculate twist on my torsion members as part of the process of getting to my pure torsion stresses.