The rigidity referred to in Section 2227 refers to connection
consistent with your frame analysis. If in your frame
software you assign joint and member boundary conditions as
pinned, your connection design must match. I believe the
with engineers who treat the connections as partially
analyze them as fixed or pinned. If you consider the rack
to be partially restrained, you must somehow provide an
for the partial restraint, something that may not be
Thank you for your reply.
The situation is not "totally" equivalent to classify a
connection as fixed or pinned. Even for a pinned connection, how do you
make sure a design has adequate rigidity for axial force? Most truss
analysis assumes pinned connection, not a spring connection, is perfect
rigidity for axial forces and has the infinite stiffness for axial
force. If a pinned connection design does not provide a sufficient large
stiffness for axial force, the internal forces must be redistributed.
Safe or not? I personally believe the problem is not on how to classify
a connection as fixed or pinned, but how to analyze the minimal
(required) connector stiffness so as to comply with the design code.
Connector stiffness and connection loads (forces/moments) are
coupled. Different connector stiffness must have different connection
loads. It would be better to fully satisfy the requirements of adequate
rigidity and forces (moments) for a connection design.
Partial restraint (partially rigid) analyses and software are
well developed, including static and dynamics. It is not a point to
discuss structural software and analysis methods. If you have design
experiences, is it possible to share your experience about how to make
sure a connection has an adequate rigidity. Thank you.