936
Proceedings of the 18
th
International Conference on Soil Mechanics and Geotechnical Engineering, Paris 2013
Table 1. Scaling factor for different physical units.
Physical unit
Prototype/Model
Length W, H, R
N
Frequency
1/N
Time
N
Velocity
1
Acceleration
1/N
Wavelength
N
Dimensionless length W/
λ
, H/
λ
, R/
λ
1
Since an unbounded half space soil medium is replaced by a
container with limited dimensions, radiation conditions at the
boundaries cannot be satisfied perfectly. It is well known that
body and surface waves lose the most of their energy after
traveling some cycles of motion or wavelengths through the soil
(after 3 to 4 wavelength) because of the geometric and material
damping. So appropriate container dimensions and the
excitation frequency must be selected to reduce these effects. In
principle, screen dimensions (width and length) are normalized
with respect to the shear wavelength to be comparable at
different frequencies.
2 TEST BENCH
The test bench consists of the following parts:
1) The container: a demountable box with a floor, and side
walls. The walls are made by deformed galvanized-steel plates
with 0.75 mm thickness. Interior of the container, side walls and
the floor covered by wooden plates to provide a proper smooth
surface.
2) The isolating screen: a concrete slab that can be covered by a
thin layer of resilient material. The isolating screen is
completely embedded in the soil medium.
3) The soil: a sieved, dried fine sand.
2.1
The soil treatment
The container is filled with Mol silica sand with an average
grain size (D
50
) of 0.26 mm. The sand is properly sieved,
washed and then dried.
The soil treatment should be (1) repeatable, (2) operator-
independent, and should results in (3) a tight tolerance in soil
conditions (uniformity of the soil density).
Investigation in different soil deposition methods (Miura and
Toki 1982, Vaid et al. 1999) have shown that the pluviation
method is less operator-dependent and more repeatable than the
other methods such as the moist tamping, dry tapping (the sand
being poured in layers) and pouring using a hand rotated flask.
The density of pluviated specimen depends on (1) the fall
height, (2) the depositional intensity, and (3) the uniformity of
the sand raining. To provide a uniform density, it has been
shown that the pluviation device should be raised continuously
with a constant low fall height and a constant drop energy.
Since an universal device does not exist for the soil deposition
by the pluviation, a pluviation device compatible with the
container dimensions has been designed and fabricated. The
pluviation device consists of three main parts:
1) A tank or reservoir in the upper level to deliver the sand
through a nuzzle, 2) A shutter that can be in open/close position
to control the deliverance of sand. Shutter consists of a fixed
perforated plate and a sliding plate. Opening the sliding plate let
the sand to pass through the holes in the perforated plate.
3) A diffuser consisting of a guide box with two grids (sieves).
The second grid is posed at the lower part of device and its
holes have different direction to polarize the drop.
The sand delivers from the top by its gravity through the
opened-shutter. The minimum fall height can be modified by
changing the position of the diffuser respect to the soil surface.
Figure 1. The sand deposition by the pluviation technique.
As the rate of the sand flow increases, because of the air
turbulence, a non-flat surface of the sand is generated. This
effect generally happens when the sand is raining with a high
flow rate and higher fall height. The air turbulence and non-
uniformity can be controlled by reducing the fall height as well
as by decreasing the depositional intensity.
The sand deposition has been performed using the pluviation
device that moves on two rails over the container with a speed
of approximately 0.18 to 0.2 m/s. The height of the sand drop is
varied from 15 to 20 cm, figure 1.
2.2
Investigation on sand properties
The sand density has been measured conventionally during
pluviation by posing small cylindrical receptacles in different
depth. To measure in-situ density, a total of 16 receptacles were
installed at different depths: 40 cm, 20 cm and at the surface of
the sand. The receptacles were distributed along the container
width at distances of 40 cm and 60 cm from the container
sidewalls. Results show an average density of 1640 kg/m3 near
to the surface, 1685 kg/m3 at 20 cm in depth and 1700 kg/m3 at
40 cm in depth.
In addition, upon completion of the soil pluviation, the
uniformity of the soil stiffness (at the top layer) is examined by
the impedance test. The test configuration consists of a small
steel foundation, two accelerometers installed on the
foundation, and a hammer, figure 2.
The foundation response due to several hammer impacts is
measured. A set of points on the sand surface has been selected
for the impedance test.
Figure 3 shows the mobility function of the foundation
measured due to the impact hammer test. Results show a
resonance frequency range from 120 to 130 Hz at different
measurement points.
Since the foundation is rigid, the dynamic foundation-soil
system can be modeled with a dynamic system with a single
degree of freedom with the foundation mass and the soil
stiffness.
