Civil Engineering Reference
In-Depth Information
yield of the main flexural reinforcement while ductility is governed by
failure of the concrete.
For the tests which have been conducted, the main reinforcement ratios
(typical of these found in practice) were low enough for the main
reinforcement to yield. It is theoretically possible to increase the amount of
reinforcement to the point where it does not reach yield before the concrete
crushes. As in the case of normal or shallow beams, such over-reinforced
members are to be avoided in practice.
4.2.2
Continuous deep beams vs continuous shallow beams
The two test series noted in section 4.2.1. revealed the following major
behavioural differences between deep and shallow continuous deep beams,
i) Deep beams develop a marked truss or tied arch action while shallow
beams do not.
Figure 4.3
presents a comparison of deep beam and a shallow
beam. In the shallow beam the shear is transferred through a fairly uniform
diagonal compression field with compression fans under the point load and
over supports. In the deep beam most of the force is transferred to the
supports through distinct direct compression struts (zones of predominately
uniaxial compression).
ii) After cracking, stresses in deep beams deviate significantly from those
predicted by an elastic analysis.
Figure 4.4
presents the stresses in the main
flexural reinforcement of a deep beam immediately before and after
diagonal cracking.
iii) The initial diagonal cracks in a deep beam do not cross the major
compression strut. In some instances they outline the strut. After diagonal
cracking, the concrete contribution to shear strength increases for a deep
beam because a stable truss is formed. In a shallow beam there is little if any
increase in shear capacity.
iv) The bending moments over supports are smaller and the midspan
bending moments are correspondingly larger than predicted by elastic theory
for shallow beams. The crack patterns, support reactions, and strain
measurements all indicated that the negative moment over the interior
support was smaller than the positive moment at midspan. The ratio between
experimental and elastic interior support moment was typically 60-70%
prior to yielding of the bottom flexural reinforcement. For several of the
beams without heavy stirrup reinforcement, the top flexural reinforcement
did not reach yield before the specimen failed.
v) The deep beams were found to be very sensitive to differential support
settlements. Even small differences in support settlements lead to large
redistribution of moments for deep beams which must be considered in
design. In the laboratory under ideal conditions, differential support
settlements (elastic shortening of load cells and so on were hard to control.
The laboratory differential settlements ranged from about
L
/2000 to
L
/10
000. In real structures, differential support settlements can be an order of
