Braced frame layout - Torsion 1

[Robcat71]

Hello,

I am designing a single story braced frame building (110' long by 35' wide and 16' tall). Seismic design category B, Risk category III, 120 mph wind speed per ASCE 7-16. Wind loads govern base on the calculated loads.

Before my involvement with the project it was decided to remove the braced bays on one side of the longer building face (Architecturally driven), with the concept that the diaphragm (metal roof deck)will distribute the loads to the opposite side of the building, and resulting torsion will result in a force couple handled by the two braced bays on the short sides. To me this seems like it is asking a lot from the metal roof deck diaphragm, and the frame on the one side will go along for the ride without adequate stiffness. There are clerestory window panels on the upper 4 feet of the two long side walls that will be sensitive to too much drift/deflection.

This design has been started by another engineer and now I am responsible for making it work. Has anyone done a similar design? I would appreciate any guidance/advice. Going back to the architect to add bracing at this stage of the design is not going to be fun.

Thanks in advance,

 

[straub46]

Can you add moment frames to the open side? With a metal deck roof acting as a flexible diaphragm, the stiffness difference won't matter as much. That would eliminate your torsion and at least give some stiffness on that side.

Go Bucks!

 

[Robcat71]

Hi Straub, I have thought about doing so but have received some negative feedback on mixing systems from my coworkers. To me it seems like any added stiffness along that line will help the cause. Thanks for the input!

 

[DaveAtkins]

I designed a large three sided diaphragm once, but would not do it again. I mean, you can certainly justify it by calculation, but I kept second guessing myself. For example, is a metal deck diaphragm really a rigid diaphragm? Most engineers would say no, but that is the assumption if you design a three sided diaphragm.

I like the idea proposed by straub46. Add a moment frame on the open side of the diaphragm. Then you have the flexibility (no pun intended) to say the braced frame and the moment frame take equal lateral load, or the braced frame takes more lateral load.

DaveAtkins

 

[phamENG]

Do you have a copy of SDI's Diaphragm Design Manual?

It's very useful and has design examples for three sided diaphragm design.

For wind analysis, ASCE 7 allows you to idealize an untopped steel diaphragm as rigid. If you have drift sensitive windows, I'd be hesitant to do that and use the diaphragm design manual to estimate the semi-rigid stiffness to get a more accurate estimate of service level drifts.

 

[KootK]

Three sided "storefront" buildings are very common. Lots, and lots, and lots of retail and industrial park buildings rely on this this strategy.
Typically, your diaphragm will have a very favorable cantilevering aspect ratio that will help. This is what I see done for this kind of thing:

1) Calculate your diaphragm displacements at the corners explicitly to make sure that you feel good about those numbers which will include the impacts of both rigid body diaphragm rotation and diaphragm flexibility.

2) Many engineers like the approach of adding moment frames to the open side. The strategy is really one of redundancy. It's usually not realistic to assume that the moment frames participate significantly under an elastic analysis model. They're usually too flexible relative to the much stiffer lateral force resisting systems elsewhere in the building. I consider the moment frames somewhat optional in low seismic regions but pretty much mandatory in high seismic regions.

 

[dold]

I consider the moment frames somewhat optional in low seismic regions but pretty much mandatory in high seismic regions.

Another issue you'll run into in high seismic regions is plan reviewers (California....cough cough...) outright rejecting the "flexible diaphragm idealized as a rigid diaphragm" thing when trying to use a three-sided design. They'll insist that "flexible diaphragms are not able to transmit loads via torsion" and make you add some sort of resistance on the open side. Except in cases of light-framed wood buildings where SDPWS allows "open front diaphragms" when you meet geometric limitations and aspect ratios.

But in ASCE 7 there is a section for determination of whether a diaphragm can be considered rigid - I think it's something like "if average diaphragm deflection is less than two times the average deflection of the LFRS then it can be classified as rigid" (don't quote me on that one). Not sure if the plan review sticklers would buy off on that.

 

[phamENG]

dold - that's a seismic limitation for flexibility assumptions. For wind, which the OP indicated is the driving design requirement (OP - I'm assuming you're using R=3?), it just lets you assume it is rigid. It's in the definition of diaphragm in Chapter 26.

And flexible and rigid classifications are pure assumptions - neither are ever precisely accurate. Good enough, usually, but not right. There's always some stiffness to it, and it can always take some torsion. Just a matter of proving it, which takes more time than any of us have, usually.

 

[KootK]

Another issue you'll run into in high seismic regions is plan reviewers (California....cough cough...) outright rejecting the "flexible diaphragm idealized as a rigid diaphragm" thing when trying to use a three-sided design.

In truth, a 3-sided building becomes "rigid" by definition as a result of it being 3-sided. It's a non-redundant system wherein equilibrium requires that the diaphragm behave as a determinate cantilever thing. In this sense, a 3-sided building's diaphragm flexibility is irrelevant with respect to it's classification. Either a 3-sided building is prohibited outright... or it obeys a rigid diaphragm equilibrium model. Granted, I'd not have high hopes for convincing a reviewer of this. And, obviously, consideration of the the diaphragms flexibility is crucial for the windows and all that jazz.

 

[sticksandtriangles]

You can check out my blog post on 3-sided diaphragms if you want some more info on these interesting creatures.

[URL unfurl="true"]https://re-tug.com/post/the-humble-3-sided-diaphragm/4[/url]

Note that I no longer agree with the WL^2/8 moment diagram that is shown in the post. I should update that image.

There is also some fun reading on the counter intuitive diaphragm deflections of 3-sided diaphragms that Kootk and DaveAtkins helped me greatly with.



S&T -

http://www.re-tug.com

 

[DaveAtkins]

Yes, a three sided diaphragm is not a cantilever beam. Rather, it behaves like half of a simple span beam.

DaveAtkins