Need a book? Engineering books recommendations...

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

RE: Seismic Lateral Load

[Subject Prev][Subject Next][Thread Prev][Thread Next]
I think the most important thing is to apply loads the way they will be
applied to the structure.  It doesn't make sense to me to have your
computer design for heavy axial loads in a beam only because you applied
the seismic load to one end of a long frame.  It also doesn't make sense
to ignore forces in collectors when you have a short frame in a long
continuous framing line.  Lastly, it doesn't make sense to design the
second floor collectors for half the force of the third floor
collectors, when the UBC requires that the diaphragms be designed for
similar loads (1994 UBC section 1631).  In summary, you should
understand the building better than the computer, and design and detail
according to what you think is correct, not according to what the
computer spits out.  Skill of computer modeling is getting the computer
to agree with you - if indeed your assumptions are valid.

Patrick Nickel, SE
 ----------
From: Geoffrey G. Melchor
Thanks Christopher, that's also one of the things that I'm planning to
do.
Geoffrey

Christopher Wright wrote:
> >Do I have to apply
> >the Fx (determined node load)to all joints at the same level or divide
> >the Fx by the number of joints on that same level.
> This is a good rule for any finite element software when you're not
> absolutely certain how it will apply loads--
> Make a simple model with loads applied the way you suppose they should
> go. Check the reaction forces and moments (all six equilibrium equations)
> against what you think the input loads should be. If the reactions
> equilibrate the applied loads within 5 significant figures, you got it
> right. Checking within three significant figures is usually ok, but might
> want further checking (like the effect of arcs modelled as chords) to
> make sure it's not an input error. Less than three figures means bad
> software or input problems.
>
> I've found hundreds of my own (and other people's) mistakes, program bugs
> and documentation errors this way. And, 30 years after my last mechanics
> class, boy, can I draw free-body diagrams. ;->
>
> Christopher Wright P.E.