collector design

[broncosfan]

I have a steel framed building with braced frames. On the frame lines, I have wide flange girders supporting joists spaced at 6' o.c. The wide flange girders are acting as collectors and support an axial force. Looking at the weak axis of these wide flange collectors in compression, what should I use as their unbraced length... the length of the beam or the spacing of the joists? The joists are only attached at the top flange. Do they indirectly also brace the whole beam to some extent. Is it too conservative to say that the beam is totally unbraced in it's weak axis because the joists only brace the top flange?

 

[jike]

I have used the entire length of the beam for unbraced length. If I needed to add a joist bottom chord extension for economy, that is what I would do.

By Yura's info. on LTB, a reduced effective length can be calculated. As you suspect, the welded joist seats have a tendency to hold the top flange in such a way as to reduce the overall slenderness ratio of the beam.

 

[shin25]

As jike has mentioned, add a bottom chord extension so to brace the bottom flange as well. Also, you need to consider the eccentricity of the force as the joists are connected at the top flange only.

 

[broncosfan]

Thanks. I also have to look at the other direction, where the collector beam is parallel to the joists and the top flange is braced only by the metal deck. In this direction, I can't add an extension.

 

[shin25]

Why not let floor joists to act as collectors in the other direction?

 

[jike]

In the other direction, add a brace between the top chord of the adjacent joist and the bottom flange of the beam.

Alternately, you can always increase the beam size!

 

[broncosfan]

Ok, there are rows of joists parallel to the frame line, but they are not on the frame line. How would that work? Wouldn't the load ultimately have to get to the frame line? The rows of joists could collect the load and transfer the load to the joist girders. But then the joist girder would need to somehow bring the force to the frame line. Not sure if I am seeing this correctly.

 

[shin25]

broncosfan,

Now it makes sense to me. You have these collector beams at the frame lines, which I think is a correct way to approach the problem.

 

[jike]

broncosfan:

The lateral load gets transfered thru the diaphragm until it gets "collected" by the collector beam on the (braced) column lines.

 

[wreckage]

The old school guys always taught me to double the axial load and use the unbraced length of the top flange as the overall unbraced length in the weak direction. (Or to use the entire load and 1/2 the section area of the beam in question).

Anyone else learn it this way?

 

[DaveAtkins]

wreckage,

I think that might be reasonable. You are assuming that the collector load stays in the top flange.

But each collector beam would need to be designed for the moment induced by the eccentricity between the top flange of the beam and the mid depth of the end connection. The collector load can't just "jump" from top flange to top flange--it has to go through the beam end connections.

DaveAtkins