rigid vrs flexible diaphragm

huangyhg

rigid vrs flexible diaphragm
i have always designed a metal roof deck to be a flexible diaphragm where the lateral loads are distributed by tributary areas.  the ibc 2000 uses the definition for a flexible diaphragm is that the lateral deformation of the diaphragm is more than two times the average story drift of the associated story, determined by comparing the computed maximum in-plane deflection of the diaphragm itself under lateral load with the story drift of adjoining vertical-resisting elements under equivalent tributary lateral load.  from this definition a steel deck could be qualified as a rigid diaphragm.  (or actually plywood for that matter.)
how do others design their metal deck diaphragms?  are there any special provisions that people make to insure that their metal diaphragms behave rigidly?  with the ibc definition it seems to me that you could have the same deck behaving rigidly in one direction but not the other.  how could the torsional forces be transferred if this was the case?  
any thoughts on this subject would be appreciated.
craigory,
i have had opportunity to study flexible/rigid/semirigid diaphragms in building sap 2000 fe models.  here is what i can tell you for a quick and dirty design.
if you have a relatively small buliding, say 10-20,000 sq ft then your steel deck will likely act as a rigid diaphragm.  designing it as such is conservative.  
if you have a larger building, such as a warehouse or big-box retail, here is my suggestion.  design the diaphragm as if it were rigid, as that will give you the most conservative deck fastening.  redistribute the load between braces and shear walls torsionally to give added shear in the diaphragm.  design the supports, especially braces, as if the diaphragm were flexible.
here is what i found;  if you have a large warehouse with an expansion joint in the middle, and braced frames along that expansion joint, then the shear walls - being 100-500 times stiffer than the braces - will absorb up to 70% of the total shear, even if the ibc says you have a flexible diaphragm.
i hope this helps.  i could go on to explain it more, but i don't know who would want to read all that much.
-doug
dougantholz
i understand easily what you do when you say yoo model for some situations the diaphragm as rigid.
my question is about your second kind of model...
when say you model the deck as flexible, what do you do?
say...
do you model the actual deck through say membrane plates and some form of pinned connection to the beams, joists or purlins?
or you use some kind of further simplification?
thanks beforehand for your attention.
ishvaag,
you can reduce the stiffness in a model so that the diaphragm is flexible, or you can use the tributary width method.  either works - and more importantly gives the same result.
dougantholz or anyone else,
i was wondering more how do people make their flexible diaphragms into rigid ones.  is it by pouring a layer of concrete (how thick?), add steel straps, or use a lot of sidelap connectors.  
thanks,
craig
ps.  does sap analyze diaphragm deflections?
craig,
it depends on the building, if it is existing or new.  how big?
diaphragm deflection is an easy hand calc when the diaphragm is regular in shape: see the sdi handbook in the us for equations.
sap can do the deflections, once it is calibrated; but why go through that much trouble when it is an easy calc.
ok, dougantholz, thanks.
craigory28
i had an oppurtunity recently to design a cold-formed joist floor building with concrete shear walls. i too was wondering how to consider diaphram behaviour.
finally i found out the shear wall forces by considering both the diaphram is flexible and rigid and i considered the worst shear in the wall for the design. we can call this as envelope method using which we will always be in the conservative side, since the codes are not clear.