What is the source of the 0.4SDS max. diaphragm shear found in ASCE7?

[JAE]

In ASCE 7 there is a seismic design section for diaphragms. (12.10.1.1 in ASCE 7-05)

There it specifies a formula to determine the lateral seismic force to be used in diaphragm designs.
It also includes an upper limit on the shear (0.4SDS x I x wpx) and a lower limit (0.2SDS x I x wpx)

Does anyone know where the 0.4SDS value comes from or how it was developed - or what the philosophy is behind it?

The lateral force resisting system is designed based upon other derivations earlier in Chapter 12 and these don't apply to diaphragms.

However, it seems weird that you would design your vertical brace system and collectors, but then have cases where the diaphragm itself is designed for a lower (0.4SDS) load.

Thanks

 

[structSU10]

I don't know the exact answer but here are my observations:

I have never had a diaphragm design force lower than a vertical distribution force for SFRS design. Cs is Sds/(R/I), or some other variation, which should always be smaller than the 0.2 Sds or 0.4Sds.

The diaphragm is not supposed to behave plastically - if it were to create a plastic hinge a rigid diaphragm would lose the ability to act rotationally. That is why the diaphragm is designed for a high load. I think it also has to do with the notion that a diaphragm will see a higher load the ELF provides, due to the whipping action of the structure, but I am not 100% on that idea.

 

[JAE]

If you look at FEMA 450 and other documents like it the design examples they give show that the 0.4 limit does control quite often. Especially near the roof in multi-story buildings.

 

[structSU10]

Yes, I have observed that, but the actual diaphragm force, Fpx, is still generally higher than the ELF procedure produces, Fx. At the worse I think they have been equal. See attached for a comparision.

I can dig through my dynamics stuff to see if I can find a more solid explination / reference.

 

[jdgengineer]

Hi JAE. I tried to look into this but found no definitive answer. 0.4 SDS is the assumed PGA, but I doubt it is related. I think that is likely a coincidence as the peak horizontal floor accelerations would be significantly higher. More likely, I think that some moderate amount of ductility was assumed and the horizontal accelerations were reduced accordingly. Assuming that the peak floor accelerations are about 3-4 times the PGA that would imply an R of approximately the same. Just my best guess, not really helpful unfortunately. Although, this has been in the code for awhile under different forms (1.0 Ca in 97 UBC)

 

[JAE]

jdgengineer - your comments are right where we ended up - I know that a maximum cutoff diaphragm force was in the UBC many years ago (under a different numerical format) but in there none-the-less.

I saw that the typical spectral response curve always starts with 0.4SDS for T = 0 but this doesn't make sense for a diaphragm as they would have their own "period" higher than zero of course.