Torsion in Steel Beam 1

[CANeng11]

I'm looking for ideas for negating torsion in a steel I-Beam. What we are dealing with is a beam supporting floor joists on one side with the other side being an exterior wall. Have any of you dealt with this type of situation before? The floor joists are a few inches shorter than the steel beam will be, so they aren't flush on the bottom flange.

 

[BAretired]

The question is too general. You need to show a cross section. Is the exterior wall being supported by the steel beam?

BA

 

[ENGINEER92]

When torsion has been a problem for me in the past I have basically added a knee brace framing into the beams that my beam experiencing torsion are supporting. When that was not an option I had to switch to closed cross section beam, which is stronger in torsion.

 

[FLCraneBuilder]

We often plate one side of a beam to resist torsion
plate approx 3/4 of span
stitch welded
1/4" thick plate up to 24" deep beam
You may need access holes to install bolts for your joists (I assumed the bolt on?) - adds a bit more complexity to the detailing
Note: I assume it will be shop welded(?)... the plate welds will pull a sweep into the beam. We pre-sweep the beam in opp direction to allow it to pull back to reasonably straight.

 

[Hokie93]

Assuming the wall is not supported by the steel beam, you could specify that the centroid of the joist seat reaction be coincident with the beam centerline, thus eliminating the eccentricity. There are two possible downsides: some added cost and the depth of the joist seat potentially increasing beyond that of a standard joist seat. I recommend consulting with a joist manufacturer if you cannot otherwise minimize the torsion.

 

[JedClampett]

Do you want to negate torsion or negate torsional warping?

 

[thoughtofthis]

This seems like a pretty common problem (i.e. - joist seats putting an eccentric load on a beam). Worst case, bearing length is 2.5 inches and the reaction is approximately 1.25 inches from edge of flange. There are a few options that I know of, but each has it's own issues.

1) require the joist reaction to be centered on the beam, by extending the joist seat across the flange of the beam. Depending on the flange width of the beam, this may cause the seat to be deeper and push the top of steel lower. $
2) brace the bottom flange with a strut from the joist. If the bottom of beam is less than the bottom of joist, extend the brace from the nearest top chord panel point. $

I think most people ignore the torsion and call it good. I'm not sure exactly what happens when it's ignored, but I guess the the beam rotates and induces a moment back into the joist? How do you all handle this?

 

[JedClampett]

thoughtofthis, or you could run the numbers for torsion, increase your beam size if necessary, and sleep soundly at night.
I've seen torsional issues due to shear plate connections all on the same side of a beam. The beam (a W12 x 14; a horrible section for any torsion) was perfectly fine for the vertical loads but it twisted noticeably when loaded. We didn't recognize the torsional mode. Luckily it was an easy fix and not a change order.