Seismic Separation Joint

[lutein]

I have a question of designing seismic separation joint:

When sizing for seismic separation joint or expansion joint between 2 buildings, should we amplify the cracked deflection of a concrete building by Cd factor in Table 9.5.2.2? The building’s cracked deflection is produced by applying the crack factor prescribed in ACI 318 by using ETABS and Pdelta Analysis.

ASCE 7-02 Section 9.1, under “DISPLACEMENT” – “Total maximum displacement” specifies that: “The maximum considered earthquake lateral displacement, including additional displacement due to actual and accidental torsion, required for…design of building separations and vertical load testing of isolator unit prototype.”

Question:

1) It appears that for the design of building separation, the amplification factor, Cd, in Table 9.5.2.2 is not required. Is this correct?

2) For seismic displacement, are we allowed to use the stiffness increase as specified in ACI 318 Section R10.11? Which specifies that: “Analyses of deflections, vibrations, and building periods are needed at various service load levels…it is satisfactory to use 1/0.7 = 1.43 times the moments of inertia given in 10.11.1 for service load analyses.”

Thank you for your help.

 

[JAE]

1) Displacement for your purposes should include the Cd factor. See 9.5.5.7.1 in ASCE 7-02.

2) The commentary section you are referring to is dealing with service level analyses....such as second order analysis of a moment frame where the Pdelta effects are to be calculated based upon non-factored loads. The rough-estimate Ie values in 10.11.1 are meant to estimate the extent of cracking in these sorts of analyses.

This commentary section is not deailing with seismic deflection calculations (the only seismic reference is to the fact that building periods are based upon service level loads).

However, for seismic design, the extent of cracking under the "real" seismic event is much greater (Cd times the design seismic loads) and this consideration is demanded by Section 21.2.2.1.

Go read ACI 318-99 commentary section R21.2.2, second paragraph. There it states: [red]"For lateral displacement calculations, assuming all the horizontal structural members to be fully cracked is likely to lead to better estimates of the possible drift than using uncracked stiffness for all members."[/red]

Thus, adding to the stiffness with 1/0.7 is not proper for seismic deflection calculations as under the real event, the building enters a non-elastic condition anyway.