-
1
- #1
saburo
Structural
- Mar 19, 2004
- 13
I have read the old thread(802-169700)regarding the usage of "Semi Rigid Diaphragm" that became available from ver 9?(Im not sure from what version to be honest).
The thread basically states that the only usage of semi rigid diaphragms is to define the extent of the diaphragm(building width etc dimensions necessary for code specific wind) for automatic wind generation with in ETABS, but I just wanted to add a bit more to it and share my opinion.
I have been using semi rigid diaphragms with earthquake response spectrum case. With the semi rigid diaphragms, we can put accidental eccentricity(most code states 0.05B -0.1B,B=building width) without having rigid diaphragms.So that's the beauty of having semi rigid diaphragms(same thing can be said to the user wind forces too, we donot need assign rigid diaphragms any more to put forces).With the former version of ETABS, we had to assign rigid diaphragms to activate the ecc addition, which would always end up with very unrealistic force distribution,ie. if you have stiff elements in the model, forces just shoot off to stiff elements one to the other and often ends up with reverse shear forces between adjacent storeys. The term of Semi rigid diaphragms however is very confusing. I understand, in ETABS, analytically, no difference at all to be made in the model(no reduction/condensation to the matrix) than flexible diaphragms case which is by default what they are.
Here is one thing tho. The way ETABS puts eccentricity in spectrum case seems a bit odd. I have checked ecc cases and found that ETABS actually superimpose ecc forces as series of modal external forces(static equivalent) x ecc dist to make up torsion. so if you have a look at member forces you can see the difference due to ecc. and also in the global forces(storey shear). But in the reaction forces for some reason, they seem to stay as if there is no ecc. Just wondered if anybody here can agree with me?(i am unsure if that's a bug or I am dreaming LOL)
I think CSI should in any case elaborate the theory behind it in the manual but..At present it's a black box.
Sorry for my English.
The thread basically states that the only usage of semi rigid diaphragms is to define the extent of the diaphragm(building width etc dimensions necessary for code specific wind) for automatic wind generation with in ETABS, but I just wanted to add a bit more to it and share my opinion.
I have been using semi rigid diaphragms with earthquake response spectrum case. With the semi rigid diaphragms, we can put accidental eccentricity(most code states 0.05B -0.1B,B=building width) without having rigid diaphragms.So that's the beauty of having semi rigid diaphragms(same thing can be said to the user wind forces too, we donot need assign rigid diaphragms any more to put forces).With the former version of ETABS, we had to assign rigid diaphragms to activate the ecc addition, which would always end up with very unrealistic force distribution,ie. if you have stiff elements in the model, forces just shoot off to stiff elements one to the other and often ends up with reverse shear forces between adjacent storeys. The term of Semi rigid diaphragms however is very confusing. I understand, in ETABS, analytically, no difference at all to be made in the model(no reduction/condensation to the matrix) than flexible diaphragms case which is by default what they are.
Here is one thing tho. The way ETABS puts eccentricity in spectrum case seems a bit odd. I have checked ecc cases and found that ETABS actually superimpose ecc forces as series of modal external forces(static equivalent) x ecc dist to make up torsion. so if you have a look at member forces you can see the difference due to ecc. and also in the global forces(storey shear). But in the reaction forces for some reason, they seem to stay as if there is no ecc. Just wondered if anybody here can agree with me?(i am unsure if that's a bug or I am dreaming LOL)
I think CSI should in any case elaborate the theory behind it in the manual but..At present it's a black box.
Sorry for my English.
