23-02-2013, 12:45 PM
FATIGUE PROPERTIES OF HEAVYWEIGHT, HIGH STRENGTH
CONCRETE
FATIGUE PROPERTIES.pdf (Size: 401.86 KB / Downloads: 56)
ABSTRACT:
Heavyweight high strength concrete can be used when particular properties, such as
combined high strength and good radiation shielding, are required. Such concrete, using magnetite
aggregates, can have a density in the range 3.2 to 4 t/m3, significantly higher than the density of concretes
made with normal aggregate. The high strength can be achieved with low water/cement ratios with
superplasticiser and the incorporation of silicafume.
In the present work 12 mixes were selected to achieve a slump above 100mm and strength up to 140
MPa at 180 days. To achieve this a 0.24 water/cement ratio and 3.5% superplasticiser were used. The
investigation used three levels of silicafume (0, 10%, 20% by weight of cement), two coarse aggregate
proportions of total aggregate (0.48, 0.65 by volume) and both magnetite and natural sand fine aggregate
were used. This paper discusses the results of an experimental investigation into the effect of the various
parameters for HSHWC on the fatigue life.
INTRODUCTION
The mechanical properties of high strength concrete have been studied under cyclic loading (1). However,
no information appears to exist for fatigue loading. Such loading is very important for bridges, offshore
structures, and structures subjected to heavy wind loads or machinery; these may be made from special
high strength concrete. Cyclic loading causes cracks to grow, which result in a growth of deflection and,
after certain number of cycles, may cause failure. The purpose of this test is describing the fatigue
strength of special high strength concrete.
Rosseland (2) studied the fatigue of high strength lightweight and normal weight concrete. It can be shown
that moisture effects on fatigue have be found to be scale dependent. Dried specimens of small
dimensions give generally longer fatigue lives. The sealed condition was found to give results closer to the
water curing than to air curing condition. The cyclic stress-strain curve appears to be more nearly linear
for lightweight concrete than normal weight concrete.
Research by Sparks (3) has shown however that for lightweight concrete made with sintered PFA
aggregate, the response against fatigue was much less impressive than ordinary concrete.
EXPERIMENTAL PROGRAMME
Materials
Detailed information about the materials used and their characteristics are given in this section.
Cement Portland cement has been found to be adequate for production of high strength concrete.
Experience has indicated that PC is, in most cases, entirely satisfactory and adequate for constructing
concrete shielding (5).
Mineral admixtures. Silicafume slurry was used in this programme and particular care was taken over
curing since the pozzolanic reaction takes place over a longer time (6).
Aggregate Magnetite was used as both coarse aggregate and, in some mixes, as fine aggregate as well.
This is in line with ASTM C638 and the grading was chosen to conform with the requirements of ASTM
C637-84 for radiation shielding concrete. The coarse aggregate had a maximum size of 16 mm and the
typical chemical analysis was Fe 62 %, SiO2 3.6 %, MgO 1.2 %, Al2O3 0.5 %. The specific gravity was 4.9
and the absorption 0.3 %. The fine magnetite aggregates typically have a slightly higher iron content and
lower aluminum and magnesium oxide content.
Chemical Admixtures (Superplasticisers). A range of superplasticiser ( SP ) was investigated and SP6
was chosen for this programme were. SP6 is supplied as a liquid instantly dispersible in water and
complies with BS 5057: Part 3, 1985.
Mixed design
In order to establish a basic range of mixes a number of trial mixes were investigated. From the results of
trial mixes, eight mixes were selected for the main programme. These mixes involved a range of coarse
aggregate ratio ( A ), silicafume content and type of fine aggregate (magnetite or sand) as variables. In
addition, two reference mixes in the normal strength range with cement contents of 350 kg/m3 and
water/cement ratio of 0.5 were used. Table 1 gives the composition of the selected mixes.
Fatigue test:
Prismatic specimens, 75 × 75 × 250 mm, were chosen to be compatible with the load capacity of the
frame. Three strain gauges of 60 mm length were fitted to each specimen. Strain was monitored directly
using a SPECTRA-ms system.
Before the cyclic test, two specimens from each mix were tested under static load to define the stress to
be used for the fatigue test. In the fatigue tests the shape of the loading was sinusoidal, as is common for
experiments of this type. The variable amplitude of load is shown in Figure 4, the frequency being 2.5 Hz.
These cyclic load characteristics have been chosen to study the behavior of special high strength
concrete, observing the fatigue life at a peak stress level of 0.7fc and the effect of cyclic load on stressstrain
relationship.