Need a book? Engineering books recommendations...

# Re: Column Design

• To: seaint(--nospam--at)seaint.org
• Subject: Re: Column Design
• From: Scott Maxwell <smaxwell(--nospam--at)engin.umich.edu>
• Date: Fri, 30 Dec 2005 09:55:03 -0500 (EST)

```I would be a little careful.  The worst case for a member may not occur at
a situation/location where one of the "primary member forces" (i.e. axial
load, x-x moment, or y-y moment) occurs.  You could have a situation where
the axial load of a column is maybe at something like 70% (purely a "for
example" number) of what the maximum axial load along the length is in
combinations with a bending moment that is less than the maximum, but
that still produces the worst combination than where either the axial
load or bending moment are at their maximum.  It should be rare that such
a thing happens but it could.  Thus, ideally one should be checking at all
points along the length of a member.  This obviously not a practical thing
to do in general.  Thus, typically one checks at some descrete interval
(which is what a lot of structural programs do when they do code checks)
along the length of the member OR one would need to use some engineering
judgement to determine where the "critical" locations appear to be and
check things there.  Or one could just take the conservative route and
just check a "hypothetical" situation by using the maximums of all the
primary members forces (i.e. maximum axial load plus maximum moments) even
if those maximums occur at different locations along the length of hte
member.

Regards,

Scott

On Fri, 30 Dec 2005, Santhosh Kumar Yedidi wrote:

> Dear David Maynard,
>
> Thank you, your answer is quite satisfying. I wanted to know how
> exactly things are done. But one thing I could understand from this
> discussion is taking maximum values of the end reactions may not be
> correct. Since there is a chance of maximum values occuring in between
> the column length.
>
> Thanks
> Santhosh
>
>
> On 12/29/05, David Maynard <davemaynard(--nospam--at)vcn.com> wrote:
> > OK, if I get this correctly, what you are saying is that you are getting
> > a different location, and your weak axis bending at another location.  So,
> > three loading considerations all at different points along the length of the
> > member, and you want to know which one to design for???
> >
> > That's simple.  Check them all.  At each location, one loading situation is
> > at a max and the other two are there, but to a lesser degree then somewhere
> > else.  So, what you will end up with is three separate loading combinations
> > that will look something like this:
> > Location A:
> > P = Max
> > Mx = magnitude at location
> > My = magnitude at location
> >
> > Location B:
> > P = magnitude at location
> > Mx = Max
> > My = magnitude at location
> >
> > Location C:
> > P = magnitude at location
> > Mx = magnitude at location
> > My = Max
> >
> > You check all of these load scenarios for your member with the appropriate
> > design parameters (unbraced lengths, K values, and what have you), and look
> > at them all as beam-column interaction.  If all combinations are acceptable,
> > then you have successfully designed your member.  If one combination fails,
> > then you obviously have to do a little more looking around.  Implementation
> > of a spreadsheet would work very well in a case like this.
> >
> > I have used STAAD in the past for analysis and applied my own design using
> > the results.  Within STAAD, and arguably one of the most powerful tools for
> > me, is an input called "Beam Force Summary."  This output will give the user
> > all of the maximums and minimums along the beam length, or a group of beams,
> > at the specific location.  Along with that maximum, it will give the
> > corresponding loads at that point as well.  It makes getting the loads from
> > my analysis a lot more efficient with way less headaches.
> >
> > Now, if you want to take a conservative approach to it, you could take the
> > maximum axial, strong axis and weak axis bending, and design a section based
> > on those, with design parameters indicating the maximum unbraced length,
> > conservative K-values, and Lx and Ly at a maximum for the length.  With
> > doing this, your section will definitely be on the conservative side and can
> > likely be standardized where you feel applicable.
> >
> >
> > David Maynard, PE
> > Gillette, Wyoming
> >
> > -----Original Message-----
> > From: Santhosh Kumar Yedidi [mailto:sant527(--nospam--at)gmail.com]
> > Sent: Thursday, December 29, 2005 4:27 AM
> > To: seaint(--nospam--at)seaint.org
> > Subject: Re: Column Design
> >
> > ) Yes it means one set of forces at each end. There are two ends of
> > the column for which we get a total of two sets of forces.
> >
> > ) We can get the bending/SF/axial diagrams for each column from the
> > forces at their ends.
> >
> > For example, a column of height 10 m is there. We get the bending
> > moment diagram in both the directions (say X and Y). Let the maximum
> > bending moment in X-dir occur at  2 m from the bottom and max. BM in
> > Y-dir occur at 5 m from bottom.
> >
> > We get the axial force daigram, Let the maximum axial force be at the
> > bottom of the column.
> >
> > But all these are not at the same location from the bottom, so which
> > section of the column is critical for design.
> >
> > And another thing, let us suppose we have 500 columns in a structure.
> > Then we cant do bending/shear/axial diagrams to find max
> > bending/shear/axial for each column.
> >
> > The general procedure I saw is one takes maximum of the column
> > reactions for design. But I feel one should also consider the
> > direction of the moments at both the ends. If they are in same dir how
> > to take their effect, and if they are in opposite direction how to
> > consider.
> >
> > Santhosh
> >
> > On 12/29/05, Polhemus, Bill <BPolhemus(--nospam--at)wje.com> wrote:
> > > It isn't clear to me what you mean be "not possible to get the
> > bending/SF/axial diagrams." If you know the forces at each end of the
> > column, and you know the lateral loading on the column if any, of COURSE you
> > can get the diagram. Unless this is a response spectrum analysis?
> >
> > >
> > > When you say "two sets of forces at each end," do you mean "one set in
> > each of two orthogonal directions" or should this read "ONE set of forces at
> > each (of two) ends"?
> > >
> > >
> > > ________________________________
> > >
> > > From: Santhosh Kumar Yedidi [mailto:sant527(--nospam--at)gmail.com]
> > >
> > > After frame analysis we get two sets of forces at each end of a
> > > vertical member (column). But while doing design we generally require
> > > only one set of forces. So can any one help which set of forces have
> > > to be considered.
> > >
> > >
> >
> > ******* ****** ******* ******** ******* ******* ******* ***
> > *   Read list FAQ at: http://www.seaint.org/list_FAQ.asp
> > *
> > *   This email was sent to you via Structural Engineers
> > *   Association of Southern California (SEAOSC) server. To
> > *   subscribe (no fee) or UnSubscribe, please go to:
> > *
> > *   http://www.seaint.org/sealist1.asp
> > *
> > *   Questions to seaint-ad(--nospam--at)seaint.org. Remember, any email you
> > *   send to the list is public domain and may be re-posted
> > *   site at: http://www.seaint.org
> > ******* ****** ****** ****** ******* ****** ****** ********
> >
> >
> >
> >
> > ******* ****** ******* ******** ******* ******* ******* ***
> > *   Read list FAQ at: http://www.seaint.org/list_FAQ.asp
> > *
> > *   This email was sent to you via Structural Engineers
> > *   Association of Southern California (SEAOSC) server. To
> > *   subscribe (no fee) or UnSubscribe, please go to:
> > *
> > *   http://www.seaint.org/sealist1.asp
> > *
> > *   Questions to seaint-ad(--nospam--at)seaint.org. Remember, any email you
> > *   send to the list is public domain and may be re-posted
> > *   site at: http://www.seaint.org
> > ******* ****** ****** ****** ******* ****** ****** ********
> >
>
> ******* ****** ******* ******** ******* ******* ******* ***
> *   Read list FAQ at: http://www.seaint.org/list_FAQ.asp
> *
> *   This email was sent to you via Structural Engineers
> *   Association of Southern California (SEAOSC) server. To
> *   subscribe (no fee) or UnSubscribe, please go to:
> *
> *   http://www.seaint.org/sealist1.asp
> *
> *   Questions to seaint-ad(--nospam--at)seaint.org. Remember, any email you
> *   send to the list is public domain and may be re-posted
> *   site at: http://www.seaint.org
> ******* ****** ****** ****** ******* ****** ****** ********
>
>

******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at: http://www.seaint.org/list_FAQ.asp
*
*   This email was sent to you via Structural Engineers
*   Association of Southern California (SEAOSC) server. To
*   subscribe (no fee) or UnSubscribe, please go to:
*
*   http://www.seaint.org/sealist1.asp
*
*   Questions to seaint-ad(--nospam--at)seaint.org. Remember, any email you
*   send to the list is public domain and may be re-posted