Understanding multi-story braced frame structures.. 3

[GalileoG]

This is a very fundamental question, but I need to make sure that my understanding of this topic is correct.

In a multi-story braced frame structure, lateral loads are applied to the chords of the diaphragm, which then transfers these loads to each side where the bracing can resist these loads (kind of like how a beam would transfer shear at each end.)

1. How would you detail the chords and the connection to the diaphragm?

2. The parameter beams will take axial as well as bending, whereas interior beams will only take bending?

3. I know that a metal deck slab is a flexible diaphragm, what then constitutes a rigid diaphragm?

Thanks.

 

[Lion06]

I think a concrete slab over metal deck is a rigid diaphragm. A metal roof deck only may be a flexible diaphragm. Someone else will have to weigh in on that - I believe it has something to do with the deflection of the diaphragm compared to the story drift.
I am pretty certain that a concrete slab over metal deck is a rigid diaphragm.

 

[frv]

StructuralEIT is correct. If the diaphragm deflects a certain amount more than the bracing or shear walls, then it's flexible. I'm not in my office, so I can't point you in the right direction with certainty, but I think this is defined in the IBC. If not, surely the Diaphragm Design Manual published by the SDI would have something on this.

I'm fairly certain there are no circumstances under which a concrete diaphragm could be considered flexible.

Metal Deck and Plywood diaphragms are pretty much always considered flexible.

The main difference is what percentage of the shear is transferred to the resisting elements. In a flexible diaphragm span, it pretty much boils down to assigning half the shear to each resisting element. In a rigid diaphragm you have to assign the shear in proportion to the stiffness of the resisting elements; if they are not equally rigid, you get some torsion in the diaphragm which then must be added to to the shear the resisting elements must resist. (wow! that was convoluted, but at this time I can't think of an easier way to phrase it)

 

[slickdeals]

Diaphragm is considered flexible when the computed maximum in-plane deflection of the diaphragm is more than 2 times the average drift of adjoining vertical elements.

The connection at the chords will have to designed for axial and shear components simultaneously.

Let me try and explain my transfer mechanism. Other please feel free to chime in. I am assuming a transfer to a shear wall.
1. Check if the diaphragm has enough strength along its length to carry the forces. If yes, then you have to figure out how to transfer it into the lateral load carrying elements.

2. Calculate how much force can be transferred into the wall by the interface of the slab-wall connection. The transfer at this interface occurs by the concrete action (?2?fc’bd) and the shear-friction dowels (?Asfy). The sum of the two gives the force that you are allowing to be transferred into the wall.

3. Since the collectors are being designed for the ? forces, you multiply story shear x ? and take the difference of that from step 2.

4. This gives the force you have to transfer into the wall through the collectors.

5. Divide Step 4 / (length of collector beams) to get line shear. The reason only the length of the collector beams is taken is because you have already relied on your diaphragm to transfer the forces into the wall and relying on that length again would be double-dipping.

6. Figure out how much shear each wall can take based on its rigidity.

7. Draw your collector force diagram. Then design the connections.

HTH

 

[slickdeals]

Step 2 : 2 sqrt fc' b d and phi * As fy
Step 3 : designed for the OMEGA (overstrength) level forces. Multiply story shear by OMEGA

 

[GalileoG]

How would one go about properly modeling a rigid diaphragm in a typical FEA software?

 

[Lion06]

clansman-
I believe you specify either rigid or flexible to the diaphragm, there is nothing that you do in the modelling that will make it behave one way or the other.
One thing to keep in mind ismaking sure the proper nodes are or are not connected to the rigid diaphragm. Having nodes connected to the diaphragm (i.e. beam end nodes at a braced frame) will cause the entire member to move as a rigid body and will not reflect the axial load that the member will actually see.

 

[slickdeals]

One thing to keep in mind when modeling diaphragms is whether in-plane shear deformations occur or not.
Rigid Diaphragm:
Load distribution is based on rigidity of lateral force resisting elements. There are no in-plane deformations of the diaphragm.
Flexible Diaphragm:
Load distribution is based on tributary width. A good analogy is assuming your diaphragm to be a series of simply supported beams between the lateral load resisting elements. In-plane stiffness is accounted for.
Semi-Rigid:
In-plane stiffness of diaphragm and stiffness of the slab-column joint is considered.

In FEA programs, a rigid diaphragm definition is more of a rigid body constraint (similar to a master and slave node). ETABS has a semi-rigid diaphragm assignment which considers the inplane stiffness component. No flexible diaphragm modeling is possible in ETABS.