Civil Engineering Reference
In-Depth Information
with column-sway modes exhibit non-ductile response. This demonstrates that buildings with SCWBs
possess high action redistribution, while those with WCSBs are characterized by poor ductility and
limited action redistribution. The reason for such structural performance is threefold. In frames with
SCWB, the total number of plastic hinges is generally higher than in frames with WCSB. In weak-
column structures, plastic deformations are often concentrated only in certain storeys along the height.
For the same level of roof translation ductility, a relatively high storey ductility factor is required com-
pared to the beam-sway mechanism. In both fl exural and shear failure of columns, degradation is higher
than in beam yield. Axial forces erode the ductility available in columns, while the levels of axial force
in beams are negligible. Furthermore, systems with WCSB may experience severe damage in columns
(Schneider
et al
., 1993). Column failure leads to the collapse of the entire building.
The number of possible plastic mechanisms increases with the increase in the number of elements,
and plastic hinges are likely to form at different locations in different earthquakes. Global frame
response is often characterized by mixed-mode mechanisms, with hinges in beams and columns as
shown in Figure 2.41. Mixed mechanisms are generally caused by material randomness, material strain
hardening and overstrength due to the presence of slabs or other geometric characteristics.
To ensure that plastic hinges occur in beams rather than in columns, the latter are capacity- designed
and hence they exhibit high strength. Beam-to-column connections can be designed to withstand actions
higher than the capacity of the members framing into them, as required by the capacity design
approach.
2.3.3.3 Structural Collapse Prevention
Prevention of structural collapse is a fundamental objective of seismic design. The defi nition of collapse
may be expressed in terms of different response quantities, at local (e.g. strains, curvatures, rotations)
and global (e.g. inter-storey and/or roof drifts) levels. Collapse implies that horizontal and vertical
systems utilized to withstand effects of gravity and earthquake ground motions are incapable of carrying
safely gravity loads. Generally, structural collapse occurs if vertical load-carrying elements fail in
compression and if shear transfer is lost between horizontal and vertical elements, such as shear failure
between fl at slabs and columns. Collapse may also be caused by global instability. Individual storeys
may exhibit excessive lateral displacements and second- order
P
- Δ effects signifi cantly increase over-
turning moments, especially in columns at lower storeys.
Brittle structures, such as unreinforced masonry, fail when the maximum applied actions exceed the
strength of the system. When failing in shear, masonry walls exhibit limited energy dissipation capacity,
especially when subjected to high compression stresses (e.g. Tomazevic, 1999). In order to increase the
lateral resistance and to improve the horizontal translational ductility, masonry walls can effi ciently be
reinforced with longitudinal and transverse steel bars. Reinforced masonry walls may exhibit adequate
local and global ductility. The extent of inelastic excursions in reinforced masonry walls depends on
the detailing adopted in the design. Global ductile response imposes high inelastic deformation demands
at fi bre level, as shown for example in equation (2.18) .
Ductile steel structures, RC and composite systems do not collapse at the onset of the maximum
strength. They sustain inelastic deformations and dissipate the input energy. They are safe as long as
the required ductility capacity is available, i.e.
μ
a
>
μ
d
. Alternating actions may cause stiffness and
strength deterioration, especially in RC members. The net effect is the erosion of the available ductility
μ
a
and hence the energy dissipation capacity is lowered. Experimental simulations have shown that
collapse depends on the maximum displacement demand for well-detailed RC elements and structures
without bond or shear failure (e.g. CEB, 1996). Similar response is observed for steel structures in
which local buckling is inhibited.
Structural collapse prevention can be achieved through failure mode control. The latter is the basis
for the capacity design of structures. In the capacity design approach, the designer dictates where the
damage should occur in the system. The designer imposes a ductile failure mode on the structure as a
whole. In so doing, the parts of the structure that yield in the selected failure mode are detailed for high
