Need a book? Engineering books recommendations...

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

Re: Dynamic seismic design

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

Your questions are classic in that they point out some of the differences
between the mindset unsed in seismic  design and the way, most engineers
look at the problem.  Seismic design uses the concepts that you mention but
in many instances these issues are addressed in a much simpler manner that
allow designs to be finalized without sophisticated analyses.  Most seismic
design puts the emphasis on defining a mechanism and making sure that the
mechanism will be the final mechanism.



> On single storey buildings there are well documented types of collapse
mechanisms. I surpose that for multi-storey buildings under lateral loading
these are mainly sway mechanisms with pattern loading and 
restrictions placed on overall frame stability.

For a moment frame building the basic mechanism assumed is one in which the
hinges form in the beams and not in the columns with the exception of the
base of the column.   I think you will find that at ultimate collapse that
the pattern loadings are not very significant. and generally ignored.

Remember that my approach assumed that you had already sized the members to
comply with the code specified forces and drift limits.  This process
indirectly addresses the issue of stability by insuring that the structure
has a minimum strength and stiffness.  At the time you start to look at
collapse mechanisms you already have the member sizes which are used to
calculate the forces in the hinges which are then applied to the frame. 
The  forces in the hinges are used in place of the seismic forces on the
structure.

> My old "Structural steel design"  published by Ronald Press Company
(1964) covers plastic design for wind
>loading and I surpose that these mechanisms can still be used and applied
to seismic design. 
>The plastic design concept allows considerable redistribution of stress.
Permanent rotations will be left afterwards.

> What is eventually done about these permanent deformations left in the
buildings? 

Typically nothing.

Since the code provisions are aimed at protecting life safety, permanent
deformations are not a problem.  In this context life safety means people
can leave the building and nothing more.  There is no assumption in the
code that the building will be repairable after a big earthquake.

>Is it normal practice to keep hinges out of the bolts in moment
connections and axially loaded columns ?
>(Possibly this is covered in the design section of UBC). 

The use of bolted connections in moment frames is still a mater of debate. 
In short the answer is yes.  

Typically hinges are avoided in columns.

Look in the materials section of the UBC for provisions on connections. 
For more up to date information on connections visit the SAC web site and
read the FEMA reports generated by the SAC effort.

>The sequence in which hinges form at the ends of the beams, is effected by
the combination of sway moments
>and the beam's normal FEM's. Because of this hinges can develop at the
first hinge at a lower loads, before
> the full mechanism forms, and this can happen under working loads.

Remember you are starting with the assumed collapse mechanism and designing
the structure to make sure that it will be the actual mechanism.  Because
the emphasis is on the final collapse mechanism you will typically find
that the sequence of hinges is not critical.

>This probably means that the elastic stresses need to be checked after
completing the plastic design  and 
>sizing of members (maybe building lateral deflection restrictions prevent
this sort of problem). 
>Is it normal practice to do this extra check ? 

All that you need to do is to design the members so that they can resist
the hinge forces and the other loads such as dead and live loads.

>Are the base shear forces factored for plastic design and then distributed
linearly up the building, or are the
>floor masses concentrated at floor levels directly factored by ZIC/R ? 

The approach I suggested assumes you sized the members to resist the code
specified forces.  Then you define your desired mechanism.  The hinge
forces are then calculated based on the capacities of the members
previously selected.  These hinge forces are applied to the structure which
is checked to make shure only the assumed hinges will yield/fail.

Strictly speaking you do not perform an inelastic analysis you just make
sure that your structure can sustain the assumed inelastic mechanism.

>The storeys can then be individually plastically designed using
appropriate collapse mechanisms. 

You want to suppress all story collapse mechanisms.  This typically can be
accomplished by insuring that no hinging occurs in the columns.

>The post elastic P delta effect needs to be considered somehow ? Does this
mean that a computer non-linear
>analysis has to be done ?

You are trying to be too sophisticated.  The code provisions have
provisions for consideration of P delta effects.  While they do not
directly consider post elastic P delta effects they tend to insure that the
building has enough stiffness that the post elastic P delta effects are not
a major concern.



Mark Gilligan