Need a book? Engineering books recommendations...

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

Fw: Shear Walls

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Regarding ETABS:

I have found ETABS to be the most efficient program for building DESIGN as
long as the stucture is not to funky.  Other programs such as SAP2KN, and
RISA, are oftentimes better for EVALUATIONS, since I can model the floors
(extracting diaphragm forces using ETABS is a pain) and in general have more
flexibility modelling wise so I make fewer assumptions.

If I am using a general FEM program (e.g. SAP, RISA) finding wall forces can
be trying.  I usually resort to checking corner forces, or in a pinch, to
plotting the shell forces graphically, then basically eyeballing it to find
moments.  I wouldn't recommend that for anything but a ballpark figure -->
go to the corner forces in the end before you are done.

Regarding ETABS and "L", "C", "T" or "I" shaped flanged walls - the great
thing about ETABS is that it automatically integrates the shell element
forces to give you moments, shears and axial loads.  As far as modelling
these types of walls, I almost always model them as a single wall.  Remember
that ETABS does the job of computing the moments, about the major and minor
axes which are centered at the COM of the wall panel assemblage.  I have
found it to be problematic to model a C wall using separate wall numbers for
the flanges, because ETABS will induce high axial loads as well as moments
in the flanges.  If WALLER is used, such a case will often result in lots of
long. steel being required.  As we all know, in the design of wall which are
expected to go inelastic, over designing the long. steel can have very
undesirable results such as diagonal tension (shear) governing the behavior
of the wall, or at best, very large quantities of transverse steel being
required since Mo has been increased to such a high level.

On the other hand, I try to be careful when modelling a large, say, "L" wall
as a single element.  The UBC won't let you use too much of the flange, and
this makes sense as a very very long "L" wall will probably act more like
two separate panels than one continuous one.  But as long as the dimensions
aren't absurdly large, I like to model flanged walls as a single unit, then
use a basic column analysis program (e.g. PCACOL) to derive the PM diagram
(or even a biaxial PM), after which checking the wall for design level
forces is pretty straightforward.  Checking the inelastic behavior can be
more complicated, but is simplified if you do a moment curvature analysis
(using BIAX or even a spreadsheet) so you can calc Mo and the associated Vo
for hinge region detailing (confinement and Av, that is).

On that note, I would like to know if anyone out there has any interest in
having a moment curvature program.  Up until now I have used either BIAX
(painful!) or RcSection by Pikaso, or SEQMC.  Problem is that most of the mc
stuff I do is for T beams, which SEQMC won't do, BIAX and RcSECTION will do
but only with a lot of work.  I was thinking of doing a little program to
handle cases like T beams, but just for our office.  I already did an Excel
spreadsheet which works well, except that the user has to do the iterations
(varying c until equilibrium is attained).  Would anyone be interested in
such a program, and if so, what would they want it to do?  Any particular
section shapes, what type of concrete stress strain models etc?

I can't say that I will be doing this very soon, but maybe in the next month
or so.  Let me know if you have any comments.

T. Eric Gillham PE
GK2 Inc.
PO Box 3207  Agana, Guam  96932
Email - gk2(--nospam--at)kuentos.guam.net
Ph:  (671) 477-9224
Fax: (671) 477-3456
-----Original Message-----
From: Mlcse(--nospam--at)aol.com <Mlcse(--nospam--at)aol.com>
To: seaint(--nospam--at)seaint.org <seaint(--nospam--at)seaint.org>
Date: Monday, March 29, 1999 4:03 PM
Subject: Re: Shear Walls


>In a message dated 3/28/99 10:54:35 PM EST, Francis.Ang(--nospam--at)toyota.com.ph
writes:
>
><< Sir,
> Thank you very much for your suggestions. I intend to use a dual system to
> provide sufficient rigidity for lateral deformations and forces.
> With regards to the forces, I use STAAD for 3-D analysis but the results
are
> in stresses (kPa and psi) and not forces (KN and lbs). Previously, I
> determine the wall forces manually (assuming that the walls are cantilever
> beams with each story forces) with the resulting conservative design. I
> intend to use STAAD but I need a confirmation on the results. I presume
that
> the results are to be multiplied to the wall dimensions to obtain the
> moments and forces.
> Any suggestions and comments are most welcome. Thanks.
> By the way, do you know any other structural engineering programs, aside
> from STAAD and ETABS, which are user friendly and efficient? Thanks again
> !!!!
>  >>
>
>I have found ETABS to be rather friendly to use, but I have used it for
>several years.   In reviewing your analysis to include the elevator core,
are
>all sides of the elevator core attached to the diaphragms so the forces can
>actually get to all walls.  This can be a problem if the elevator is
located
>at the perimeter of the building sometimes.
>
>Also be aware if you are modeling the elevator as a "C" shape or box shape
>that you can develop large axial forces in wall due to overturning forces
and
>flange affects of the walls.  In some cases, the axial forces may be large
>enough that according to code (1921.6.5.3) you can not use the wall as part
of
>the lateral resisting system.
>
>To eliminate the axial loads, you can model the "C" shape as three separate
>walls  with no connections to the cross walls to eliminate the seismic
axial
>loads on the walls.  In the real world, the slab still acts to tie the
walls
>together, so there will be some axial forces.  I am curious to know how
others
>have been addressing this issue.
>
>Michael Cochran
>
>
>
>