Need a book? Engineering books recommendations...

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


[Subject Prev][Subject Next][Thread Prev][Thread Next]
Thank you for your comments Bruce.  I don't doubt that you have formulated 
the warping torsion problem correctly.  However, it is virtually impossible 
to restrain warping (bimoment) at a normal structural steel connection.  A 
wide falnge beam would have to be connected to develop both Mx and My into 
something that has not only strength but global and local stiffnesses for 
both types of moments.  A connection that resists "torsion" does not 
necessarily resist "warping torsion."  When you do not release the degrees 
of freedom at the ends of members in a RISA3D model, the program presumes 
that warping torsion is restrained.  The default should be that it is not 
restrained.  A distinction must be made between simple torsion which can be 
resisted rather easily with a wide variety of end connections and warping 
torsion which requires that the member end be welded to a monolithic block 
that restrains all deformations both global and local.

As regards the master-slave relation in STAAD.  If you model a table with a 
"master" node at the center and a "slave" node at each leg, when you impose 
a rotation theta on the node at the center, STAAD interprets the 
master-slave relation to mean that a rotation theta must be imposed at all 
nodes.  I've always thought that the master-slave relationship implied 
geometric consistency, requiring that a rotation at the master would result 
in rotations and translations at the slaves based on the geometric 
relationship between the ,master and slave.  Is this wrong?

Gregory P. Luth, Ph.D., S.E.
Krawinkler, Luth, & Associates, Inc
160 Jefferson Drive
Menlo Park, CA  94025
PH:	650 688 4710
FAX:	650 688 4712