Need a book? Engineering books recommendations...

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

FW: Joist Girder Deflection at CMU wall

[Subject Prev][Subject Next][Thread Prev][Thread Next]
(Re-sent with html turned off.)
I too have the same concerns with joist girders so close to a wall.  I
try to keep the girders at least a couple of feet from the wall to allow
for some rotation and deflection at the joist girder. 
I can't argue that what you describe is not often done nor that there
are known failures.  But it doesn't make sense to me to build something
that can't structurally behave as designed.  The wall connection will
carry load as the joist deflects and something will have to "give" for
the system to work.  
Bill Sherman


From: Greg Mosier [mailto:gmosier(--nospam--at)]
Sent: Mon 7/3/2006 11:25 AM
To: seaint(--nospam--at)
Subject: Joist Girder Deflection at CMU wall

Fellow list server subscribers,

I am currently working with a large retail client who has a set of
prototypical drawings that I am supposed to adapt to the specifics of
each project.  The general building system is a light metal deck and bar
joist/joist girder roof with an approximate 50 foot by 50 foot bay.  The
lateral system for the building is CMU shearwalls and the building is in
seismic design category D per the IBC 2003/ASCE 7.  I have an issue with
one of the details included in the set and was hoping to get additional
thoughts on it.

The issue I am having is that there are 50 foot long steel joist girders
that are approximately 8" from a CMU wall and steel bar joists bear on
the joist girders.  There are embedded plates in the CMU wall which the
heel and toe of an angle gets welded to.  The other toe of the angle
gets welded to the top of each bar joist.  The roof deck gets welded to
the top of the angle and the angle serves to anchor the walls out of
plane to the joists and to transfer the building lateral loads to the
CMU wall.  The problem I am having is justifying what happens when the
joist girder deflects.  The joist girder is supposed to be designed to
an L/360 live load criteria which would result in a 1.67" deflection at
midspan.  When the joist girder starts to deflect, it will begin to pull
down on the toe of the angle that is welded to the tops of the joists.
In the design load condition, I calculate loads that will either cause
failure of the weld to the top of the joists (or failure of the angle)
or an overload condition in the CMU walls themselves due to the moment
caused by the eccentricity of the load.  If this connection fails, there
is potential loss of out of plane wall anchorage and loss of capacity to
transfer lateral loads to the CMU walls.  

I would think that some sort of vertical slip connection would be needed
to allow the joist girder to deflect, but still allow out of plane wall
loads and in plane shear loads to be transferred.  I have spoken with
the joist manufacturer that typically supplies the joists and joist
girders and they have led me to believe that this is a very standard
method of accomplishing the connection for this condition.  They added
that there is no history of failures that would indicate a problem.  I
would prefer to avoid anything that is a potential failure issue.  Has
anyone seen this before and been able to justify that what I am talking
about is not a problem?  Are there any studies or testing to show that
this is ok?  Any thoughts would be appreciated.

Thank you,

Greg M. PE, SE

E N G I N E E R I N G 
T (253) 627.4367    F (253) 627.4395 
Civil Engineering  *  Structural Engineering 

******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at:
*   This email was sent to you via Structural Engineers 
*   Association of Southern California (SEAOSC) server. To 
*   subscribe (no fee) or UnSubscribe, please go to:
*   Questions to seaint-ad(--nospam--at) Remember, any email you 
*   send to the list is public domain and may be re-posted 
*   without your permission. Make sure you visit our web 
*   site at: 
******* ****** ****** ****** ******* ****** ****** ********