Aggregates for concrete - Concrete Mix Design, Quality Control and Specification - page 47

made with them. The sands were sieved to provide size fractions to comply
with two grading criteria and used in mortars of set composition for each
of the two gradings. Workability of the mortars was assessed using a flow
table. It was concluded that “the orifice test appears to be a satisfactory
means of determining the shape and surface texture, and hence the water
requirement, of fine aggregate”.
The test has been further developed in New Zealand (Clelland, 1968;
Hopkins, 1971) and independently in the United States (Gaynor, 1968;
Tobin, 1978). The voids result depends little, if at all, on the dimensions
of the equipment or the sample size, but different flow times will result
from differences in the equipment and size of sample. It was found, for
example, that even the sharpness of the transition from conical to cylindri-
cal profile at the orifice has a marked effect on flow time (Kerrigan, 1972).
Kerrigan (1972) and Elek (1973) describe a standardised test with defined
sample size and dimensions of the test equipment, including the size and
profile of the orifice. The specification also includes removing any particles
of size greater than 4.75 mm from the test sample, as these interfere with
the flow. Flow time results reported in this account of the test have all been
obtained using the equipment and procedure developed and standardised
by Kerrigan and Elek.
Correlation of voids in fine aggregate and corresponding water demand
of concrete is acknowledged in the ACI publication “Guide for Selecting
Proportions for High-Strength Concrete with Portland Cement and Fly
Ash” (1998), which advises a factor of (percent voids - 35) × 8 lb/cu yd
(approximately 5 kg/cu metre) amounting to approximately 15% increase
in water demand per 5% increase in voids, for fine aggregates having the
same grading. As the voids property of commonly used fine aggregates
ranges from below 40% to approaching 48% this represents a very signifi-
cant change (more than 20%) in water demand, and corresponding cement
content to obtain the same performance from the concrete.
Harrison (1988) analysed data from 37 examples of concrete mixes
for which both the flow test parameters of the fine aggregate and water
demand of the mixes were known. The latter was expressed as a dimen-
sionless parameter, relative water demand (RWD), being the factor between
water demand of a mix made with the fine aggregate in question and a
corresponding mix having fine aggregate for which voids and time plot at
a particular location on a chart with axes as shown in Figure 3.6. Using
linear functions, correlations were found between RWD and both percent
voids and the flow time.
The results shown in Figure 3.4 and Harrison's data have subsequently
been analysed further to find the positions and orientations of plane sur-
faces that best represent the dependence of specific surface and relative
water demand separately on the flow test parameters. The outcome of
this analysis is shown in Figure 3.6. Following a line of constant specific