Total shear design method in lieu of tributary area method (flexible diaphragm) 1

[HofOblivion]

A fellow engineer told me that you could design wood shear walls using the total shear design method. Every shear walls designed with equal unit shears based on the story shear.

I always thought that wood shear walls need to be designed with a tributary method based on the flexible diaphragm method.

Has anyone used total shear design method? And if so, which code section allows this design approach?

 

[jayrod12]

I've heard this rumour too. In fact when I was a junior designer my supervising engineer told me to do the same thing. I don't necessarily agree with the approach. There are arguments to be made where you can account for the diaphragm to be considered rigid, but I don't agree with the Total shear method.

 

[KootK]

Whether your diaphragm should be treated as rigid or flexible should rightly be something that you determine rather than arbitrarily decide. That said, I arbitrarily decide flexible unless I suspect that rigid will be expedient for me for some reason.

I can think of two possible origins for the total shear method:

1) Low rise masonry and concrete shear wall design from the old days (long walls). Since squat walls are generally dominated by shear flexibility, it was common -- and reasonably accurate -- to assume that load would distribute according to wall length. That, because stiffness was more or less a linear function of wall length. In this case, the method would only work if the center of load aligned with the center of rigidity. Fat chance nowadays.

2) For walls that can be assumed to yield in a ductile fashion, there is a school of thought that says that you can ignore building torsion effects for seismic because the diaphragm's own rotational inertial will keep it from twisting under transient inertial loading. I've never actually employed this and would expect it to have a pretty limited range of applicability. None the less, I believe that one of the SEAOC seismic manuals contains an example of this method.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[atrizzy]

As usual, this depends.

I have justified to myself such a load determination on a long wood building in the long dimension of the building since the diaphragm is unlikely to deflect much at all. In those cases the walls are fairly uniformly and tightly spaced.

Now looking at the same structure with the load perpendicular to the length of the building, I considered it much more likely for the diaphragm to behave traditionally flexibly and distributed the forces per tributary area.

 

[JoshPlumSE]

This reminds me of a concrete class I took in college. Where the professor said that it didn't matter much how you split the reinforcement between top (support) and bottom (mid-span) of the the beam. As long as the total capacity added up to WL^2/8. Because the beam would crack / hinge at the under reinforced area and redistribute additional moment to the over reinforced areas.

Conceptually, I can understand how one might apply this same idea with wood shear walls to come up with the total shear method. However, I don't personally feel comfortable doing so.

My concerns are the following:
1) This requires the "failing walls" to soften up enough to allow re-distribution of load to the other walls. As long as you have a truly ductile failure of the wall then this would be okay. But, you have to be willing to live with a lot of damage to allow this redistribution (the same as you do with the concrete example).

2) For more rigid diaphragms, you have a torsion issue that can lead to an instability if you're not careful. So, I'd think you'd have to have to modify the procedure to make sure that the order of failure doesn't result in a torsional instability.

3) For more flexible diaphragms, you have a similar issue. Where if an outer wall line fails at 50% of it's trib load, then the additional 50% would induce a torsion in the structure that would have to be considered.

 

[rapt]

JoshPlum,

Agree with your concerns. That logic only works if the structure can survive the redistribution!

If the initial overstress in one area causes failure rather than simply ductile yield in that area, then you probably end up with overall failure/collapse.

 

[NorthCivil]

I've heard some compelling arguments, that in wood frame, diaghrams should be considered more rigid than you think, due to the typical detail of glueing the board down (at least around these parts) that gives it some still unkown stiffness contribution, that will likely keep it much stiffer than a shearwall.

That said, I'll confess that when designing for seismic in a fairly regular shaped wood frame house, I will try to break up the forces to approximately 40% of the story mass on either end wall, and 60% down the middle. Might seem a bit fast and loose (especially to kootk, who knows how anal I am about glass , But I think it mostly works.

Getting into 4-6 story wood frame, I might start worrying more about stiffnesses, diaghram torsion, etc, all that fancy stuff - haven't had the opportunity to cut my teeth on one of those projects yet though.

 

[KootK]

I'm hoping my jurisdiction's new energy code will just make windows obsolete.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.