Excellent question. I can't answer with any real authority, but I would think you would check it the same way.
Side-sway web buckling occurs when the top (compression) flange is restrained. But, the buckling occurs on the opposite flange. Therefore, I'm not sure the cap channel (which is at the compression flange) would really affect the behavior at the other flange.... after all, we already assumed that flange was braced, didn't we?
Josh-
"...Side-sway web buckling occurs when the top (compression) flange is restrained..."
I'm not sure this is exclusively the case.
In other words, what if you have a simply supported beam with a point load at mid span and the beam is unbraced for the full length, WSB can still occur, right?
Honestly, I don't know enough about this to answer definitively. I wrote that after reading through the AISC commentary on J10.4. That states that the sidesway web buckling provision were developed to explain away some failure that occured in test specimens which had their compression flanges restrained.
But, re-reading the section I think you're correct. The issue could probably happen at any time. For a wide flange beam with an un-restrained compression flange, I would then this failure state would never control (because the compression flange buckling should control).
Your case is interesting in that the compression flange is not restrained. But, it is reinforced. Therefore, I would think it would be possible for sidesway web buckling to still occur. In fact, since the neutral axis of the beam shifts up (and a larger portion of the web is in compression) it might even be more likely to occur for your case.
"....In fact, since the neutral axis of the beam shifts up (and a larger portion of the web is in compression) it might even be more likely to occur for your case...."
if the N.A. shifts up, there would be less web in compression, no?
I hesitate butting in some times, terminology changes from time to time for reasons I don't understand. I thought I knew what you were getting at but wasn't sure.
The NA does move up decreasing the length of the web in compression. This means that the tension stress is higher than the compression stress. The cap channel is laterally restrained as to its own lateral direction by the beam, it stabilizes the top flange of the beam in the beam's lateral direction.
Mostly, I've seen these used for crane runways where the channel is useful for resisting lateral forces from the equipment.
I think there was a formula that used the radius of gyration of the top flange and one sixth (or was it a third?) of the web to calculate lateral stability. Sorry, I don't keep up to date enough to be sure of my ground, I just remember what we used to do. I remember using that formula and including the effect of the channel.
Michael.
Timing has a lot to do with the outcome of a rain dance.
If the channel is sufficiently connected to the WF to consider it a composite section, then, I would assume you could include it...have never personally had the need to check this.
JoshPlum...how can a tension fla buckle?
Pad-
You're correct. We calculate the radius of gyration of a section comprising of the top flange of the girder + 1/3 of the compression web and call it "rT". This is used is calculating the allowable or available moment capacity of the beam.
Web sidesway is a different check altogether for concentrated loads on beams
You're right, the neutral axis moving up would reduce the portion of web in compression. That probably makes it LESS likely in your case. Ugh! It is so weird thinking about the tension flange buckling that I keep getting turned around.
Sail - What happens (or my layman's interpretation of what happens) is that there is a concentrated load at the top (compression flange). This causes the come localized compression forces to develop in the web of the beam. While the the top flange may be restrained against lateral translation, the bottom flange is not restrained. The compression in the web becomes sufficient to cause the web to want to kick out and take the tension flange with it....
The AISC commentary has a number of conditions which would indicate that it could neve happen. Then they also go onto say that, if you have stiffeners under the load for more than 50% of beam depth then it will prevent this type of failure. I don't think this is a very common failure mechanism.
ok, it looks like the tension fla could buckle, however counter-intuitive it may be to me.
I have looked at this in the past and analized possible failure
mechanisms and would always come to the same conclusion that the restoring mechanism from the tension in the fla would overcome any tendency for it to to buckle. I guess what I did not consider was the affect of the web on the tension fla and it's influence if ever it lost it's integrity.
Most of my concern was with deep slender trusses or bar joists where the tension fla may not be adequately secured at the supports.
It is always good practice to use stiffeners on the bm for any significant conc. load.
What the AISC code does not quantify is the magnitude of the load in repect to the capacity of the bm...one would expect that a conc load that would cause this type of failure would have required stiffeners to be added in the first place and in turn would have prevented this failure. There must have been an odd combination of bm dimensions and load for this type failure to occur.
