Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Re: Response Spectrum Analysis

[Subject Prev][Subject Next][Thread Prev][Thread Next]

Your approach seems reasonable.  In your step 3, you describe the analysis using a scale factor of "g" as elastic;  this would actually correspond to an R value of 1, at which point your building would probably not be elastic.  The "elastic" response would correspond to a response spectrum analysis with a scale factor of g/R.  The maximum inelastic response would then be estimated by multiplying the displacements by the Cd factor (IBC) or 0.7R (UBC).  The base shears with the scale factor of g/R should be compared with the ELF base shear.  There is a code minimum for the response spectrum base shear, specified as a percentage of the ELF base shear (you referred to 0.85).  I assume that this is because the response spectrum base shear is very sensitive to how cracked your structure is;  decreasing the cracking factors on your lateral elements will decrease the apparent force.

A caveat:  the combination of modal responses in a response spectrum analysis creates positive values for all responses.  The code checks must therefore be done carefully, assuming the worst combination of the signs of the forces.  ETABS seems to handle this well in their design post-processors, but it should still be verified.  Also, note that it is possible to have only one Response Spectrum load case that results in design forces for all directions of motion, if two orthogonal directions are specified for the functions.  The ETABS help and documentation contains useful information on this topic.

Daniel Popp

----- Original Message ----
From: Anantha Narayan C.K. <anant_27(--nospam--at)>
To: seaint(--nospam--at)
Sent: Tuesday, September 26, 2006 2:47:45 AM
Subject: Response Spectrum Analysis

I am using ETABS to do a Response Spectrum Analysis for a parking garage. I have never done an RSA on ETABS before and hence, want to make sure
that I am doing things the way they are supposed to be.
1. I have done an ELF analysis on the building and have the base shears from it. I have used an R value of 6 for my system (combination of shear walls
and braced frames)
2. For the response spectrum analysis, I am using the IBC 2003 curves generated by ETABS with Ss = 1.71 g and S1=0.68g. When I use the curve to set up
my response spectrum cases, there is a scale factor term on the program. For the first run, I would scale the accelerations in "g" to the corresponding units "k-in" or "k-ft" using the factor 386.1 or 32.2. Is this correct?
3. The generated response spectrum is an elastic response spectrum and the base shears are for an elastic response. In the end, would I divide the base shear by my R value to get the inelastic response or use a scale factor of 386.1 / 6 or 32.2/6 to get the inelastic response?
4. Or should I use 0.85 Base shear (elf) / Base shear (modal) for the scale factor?
I would appreciate a response for carrying out a RSA analysis in ETABS.

Structural Engineer
Bliss and Nyitray Inc.
Miami, FL - 33134

Now you can scan emails quickly with a reading pane. Get the new Yahoo! Mail.