Overstrength Factor for Elements Supporting Discontinuous Walls

[sward1971]

I have several three-story & two-story wood-framed buildings with plywood shear walls. Where we have a beam supporting a discontinuous shear wall above, this is classified as a vertical irregularity Type 4 per ASCE 7-05 Table 12.3-2. According to ASCE 7-05 section 12.3.3.3 - Elements Supporting Discontinuous Walls or Frames, this beam shall be designed with the overstrength factor (Omega sub zero). In addition, "the connections of such discontinuous elements to the supporting members shall be adequate to transmit the forces for which the discontinuous elements were required to be designed."

I understand that the beam has to be designed with the overstrength factor, but here's my question: Aside from the beam, is the overstrength factor implemented throughout the rest of the design until reaching the foundation? In other words, do you have to design the beam-post connection, the post, and the post base connection using the beam reactions that have been increased by the overstrength factor? One could conservatively use the amplified loads throughout, but the connections can become quite expensive, especially for residential buildings. Therefore, it is more cost-effective not to use the increased load in the design unless it is warranted by the Code. The quote above is the only Code reference to using the 'amplified' loads for the connection design, but it refers to the wall-beam connection, i.e., the hold-down.

I have not yet found a definitive answer on any of these discussion threads and each engineer I talk to seems to have a different opinion on this all thanks to the ambiguity of the Code. Your help is appreciated.

Sincerely,

Stephen R. Ward, S.E.

 

[JAE]

The overstrength factor is only used on particular elements of the structure where specifically mandated by the code.

The code up-strengthens particular elements that are susceptible to brittle, non-ductile failures.
You do not have to extend that overstrengthened load through the entire load path of the structure.

 

[jittles]

As JAE said, only the elements that the code specifically states. In this case, the beam gets designed for overstrength. Then, "the connections of such discontinuous elements (the shear wall) to the supporting members (the beam)" comes in and makes you design the holdown connection to that beam. Since the beam is designed with overstrength, then the connection is also. That is as far as it goes.

 

[wannabeSE]

JAE and jittles,

The code says "Columns, beams, trusses, or slabs supporting discontinuous walls or frames . . ." Why wouldn't you apply the overstrength factor down to the foundation? I can't say that you are wrong. Rather, I just don't see it the same way.

This is not the case of a frame assembly with certain elements designed to yield while others elements need to remain ductile. The case in point is a shear wall supported by a beam. I would be equally concerned with the beam and the columns supporting the beam. I might even be more concerned with the columns since a column failure would jeopardize four bays and may lead to progressive collapse. When in doubt, I interpret the code literally and the columns are also supporting the discontinuous shear wall.

However, I do see your point. Particularly since the section of code also states "The connections of such discontinuous elements to the supporting members . . ." which only deals with the shear wall connection to the beam.

 

[jittles]

wannabeSE,

I can see your point.

First, the code specifically says the connections of the discontinuous element to the supporting members. It says nothing about the connections of the supporting members, or their respective supporting members.

I would be pretty sure of the interpretation for the connections, but it is a bit gray on the supporting member side. Technically, if a column is supporting the beam that is supporting the shear wall... well then the column is supporting the shear wall also, I suppose. If they used the language "directly supports" then that would clear it up.

 

[jittles]

Upon further review, the last time I ran into this situation I designed all the supporting elements for the overstrength load down to the foundation. I had a shearwall landing on a beam. Beam, columns, footings were all designed for the overstrength load. None of the connections were, as the "discontinuous element" aka the shear wall was not required to be designed to overstrength.

 

[MikeSFBay]

"the connections of such discontinuous elements to the supporting members shall be adequate to transmit the forces for which the discontinuous elements were required to be designed." means that the hold-down supporting the tension chord of the shear wall needs to be sized based on the required shear loading on the top of the shear wall, which in most cased is just rho x Eh.

As far as how far down the load path the omega needs to be applied, definitely apply over-strength to the supporting beam and its connection to the supporting column. I've seen several references (Malone) that interpret the code as mandating the over-strength design of the column and the column base cap, if any. I think this make sense since one can easily size an ECC, for example, for overstrength. But what would be the point if the column was not adequately sized to support the load for which the ECC was sized? Additionally, if you have to design the column for over-strength, what about the column connection at its base? If the column were designed using over-strength, does it make sense for this too to be penalized? If it does, then what about the foundation, etc?

My interpretation is that the code wants these elements to remain elastic during a design basis earthquake. If you're using ELF procedure, you don't expect the column base cap to actually be stressed to an over-strength level if you're only designing the lateral system based on the distributed shear from Eq 12.8-11, but you don't want it to be the weak-link either.