Need a book? Engineering books recommendations...

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

Re: Dynamic Analysis

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Taking that into account, if doing modal analysis, you will need to account for a percentage of  your mode shapes based on the static analysis base shear.  So it is appropriate to have a static model set up well.  I agree that these CSIBerkeley Papers are some of the best I have found.

I took a course in graduate structural dynamics at UNM, but I did not find it a whole lot more comprehensive than my undergraduate vibrations course at UCSD which boggled my brain.  However, I sat-in on a course in structural dynamics graduate level course at UCSD, and it went into fairly advanced topics in programming Time history analyses and such, which you probably will not need.  I did later purchase the book that it used, and I do find it a very good read, that can help one understand the levels of computation involved and which walks through hand analyses.
http://oed.berkeley.edu/dta99/chopra.html   - Dynamics of Structures.

I do recommend familiarizing yourself with a 1-D model (sticks and masses with dampers and springs),  so that you can interpret the results of a computer model.  Chopras book goes into this.

I have not run a complete dynamic analysis on an actual building, however, I ran into a lot of problems with my model on a fairly complex building because I was unfamiliar with the FE tool I was using.  Make sure you get all of your connectivities in line for the tool you use, so you don't have mode shapes of flapping members or something obscure.  Perhaps run some 1-D tests to evaluate your program, and then for your particular model to make sure you are getting some ballpark numbers, and then go for it.  I think it is straight forward.  Just back yourself up, so you feel comfortable with your numbers.  Look through your mode shapes, in the visual feedback, and see if they are in line with your 1-D model.  If not, look to understand why not.

In effect, while your overall base shear should resemble your static analysis, keep in mind, the direction of action of the dynamic base shear will be unique for the building.  It will not be standard X and Y.  This finding of the "principle direction" is considered to be one of the best reasons for doing dynamic analysis, as it will accurately resemble how the building will respond.  The papers in CSIBerkeley discuss this as I recall (if I am recalling my source correctly).  It is also a good reason to do it in 3-D, and another reason why it will be very difficult to take on by hand, unless you want to create a book or the building is fairly regular shaped, symmetrical and with few stories.

You may also want to read up a bit on resonance and damping if approaching from a Time-History perspective.  This is the more complicated method to go, which requires more knowledge about your particular location, and perhaps creating your own earthquake input from an earthquake generator (Berkeley has one somewhere that I found once upon a time, but would take a while to find).  You may need this approach if you are on a more complicated soil strata like Mexico City !

Hope this helps.

On 11/21/06, Jake Watson <jake.watson1(--nospam--at)gmail.com> wrote:
I picked it up under the supervision of two other senior engineers.  The basic concepts are relatively easy.  CSIBerkely has a some great white papers on their website.  If you have never done this before, I strongly suggest your read them.  You can find them at http://www.csiberkeley.com/support_technical_papers.html.

My two cents for dynamic are the same as for static analysis.  Garbage in, garbage out.  In addition to needing a realistic stiffness model (similar to a static analysis) you also need a realistic mass model.  The few times I have done dynamic models, I started with a static model.  Get that model working then work towards the dynamic model.  The final dynamic results should be confirmed with static results for a gut check.

Best of luck,
Jake Watson, P.E.
Salt Lake City, UT


On 11/21/06, Will Haynes <gtg740p(--nospam--at)gmail.com> wrote:
How many of you have done a dynamic analysis due to earthquakes?  Did you pick up doing this on your own or did you take a dynamics class? I have not ever had to do structural dynamics but I know it can be very difficult and I am just wondering if it is something you can pick up on your own if you are required to do it. I figure that most people are using software (SAP etc.) to do the majority of the work but I would like to be able to do it all by hand if I had to. Do you think that most engineers that doing earthquake dynamics really know what they are doing or are they just relying on the computer? 
 
 
Will