Civil Engineering Reference
In-Depth Information
and of settlement of the grout down the inclined stay, both eventualities leaving voids
and risking future corrosion.
Sheathed strands are generally installed as for bare strands. Clearly the sheaths must
be removed adjacent to the anchorage where the strands are secured by wedges. This
may be the weak point in the corrosion protection and must be carefully designed
and built. The windshield/duct consists of half shells that have to be erected using a
climbing rig or crane and is installed after the stay is complete.
There is absolutely no consensus on the best approach to the design of stays. This is
underlined by two of the longest cable-stayed spans. The Normandie Bridge in France,
with a main span of 856 m, uses individually sheathed galvanised 15.7 mm strand,
assembled on site and then protected by half shell windshields. The Tatara Bridge in
Japan with a main span of 890 m uses galvanised 7 mm parallel wires in an extruded
sheath, with the stay prefabricated off site.
It is generally accepted that stays should work under full live loading at a direct
tensile stress corresponding to 45 per cent of their breaking load. Although used as the
basis for the design of most bridges, this does not appear to be a completely rational
limit. One would expect the acceptable working stress to be dependent on a variety of
factors including:
the redundancy of the structural system;
the fatigue performance of the stays and of their anchors;
the spectrum of live loads.
The tension force in the stays under dead loads alone should not be too low; long
lightly stressed cables will sag appreciably, and their effective Young's modulus will
fall signifi cantly below the theoretical value [7]. For instance, for a cable that is 200 m
long in the horizontal projection, with a stress of 0.3 of its breaking load, the effective
E will be about 97 per cent of the theoretical value, while for a 400 m long cable the
fi gure would be about 88 per cent. At 0.2 of the breaking load, the two fi gures would
become approximately 90 per cent and 70 per cent. This effect is most signifi cant for
long backstays, as their stress will be reduced below the dead load value by live loads
applied to the side span, which pull the tower head backwards.
The protection of stays from corrosion is a continually evolving science. Not only
are they very exposed to wind and rain, but the solar heating and subsequent cooling of
the ducts causes a pumping action, tending to suck in damp air through any defects.
The angle at which the anchors are installed must be truly co-axial with the stay.
This must take into account the sag of the stay under dead loads, as well as its precise
inclination in both planes. The design of the anchorage housing must make it possible
to achieve these very close tolerances, either by adjustment on site or by precise
prefabrication.
The stay sag will vary as live loads are applied, and the stay will defl ect under
the effect of wind. To avoid these defl ections applying bending moments to the stay
anchors, which would compromise their fatigue resistance, the anchor is prolonged by
a rigid tube providing a point of support to the stay about a metre from its forward
end. Usually, the stay is separated from the tube by a neoprene ring that protects it from
crippling, and also provides some damping, reducing the risk of harmonic vibrations.
The tube end may also incorporate a deviator that allows the individual wires or strand
to splay to their spacing at the anchor head. The length of this supporting tube may be
Search WWH ::




Custom Search