4
Each water chamber plate bordering the slope model
has perforations in the plexiglass to allow water to flow in
and out of the chamber. The perforations are covered with
a fine mesh smaller than the sand particles to prevent any
sand particles from entering the water chamber but rather
to allow free flow of water through to the soil and prevent
flushing of soil particles.
A plain water hose connected to a water container was
used to fill in the water chambers. The water container can
be elevated or lowered to any level to change the pressure
heads. Changes in water level inside the model slope were
controlled by injecting water into the sand layer or drawing
water out of the sand layer through inlets and outlet points.
These helped in the various simulations of constant water
rise and drawdown of water.
3.4 Finite Element Modeling
Slope stability programs such as Slide 2D can be used to
find safety factors in both soil and rock slopes. The com-
puter program can compute circular and non-circular fail-
ure surfaces while evaluating the probability of failure. The
software is developed by Rocscience in Canada and can
be used for soil and rock slope stability analysis. “Slide2
is simple to use, yet complex models can be created and
analyzed quickly and easily. External loading, groundwater,
and support can all be modeled in various ways. Individual
slip surfaces can be applied to locate the critical slip sur-
face for a given slope. Users can also carry out deterministic
(safety factor) or probabilistic (probability of failure) analy-
ses” (Rocscience Slide, 2019).
The two-dimensional finite element method Rocscience
was used to analyze various boundary elements. The geom-
etry and soil properties of the model were the same for the
soil used for the laboratory experiment. The modeling was
used to find the failure planes, pressure heads, flowlines, and
flow vectors. Modeling in Slide2 was possible because it can
analyze groundwater, pore-pressure, steady-state, and tran-
sient state. In all the analyses, the mesh consisted of 3-node
triangular elements and approximately 1500 elements. The
default unit weight of 9.81kN/m3 was used for the density
of water and good soil characteristic curve and water con-
tent provided to help in the analysis. The numerical results
are compared to that of the laboratory test conducted.
3.5 Initial Conditions.
The model slopes were constructed in the middle cham-
ber and compacted in layers of 50 mm. The first layer was
spread uniformly and compacted to form the base of the
slope. The first experiment was to simulate steady-state flow.
The dimensions of the slope were 220 mm high, 100 mm
wide, and a crest of 70 mm. The final slope angle was about
35° (Figure 5). After constructing the model slope, the
upstream tank was filled gradually with water to the level
of 190 mm. The water was kept constant and allowed to
flow through the soil and seep to the downstream chamber
exit points. To initiate failure, the water level is kept con-
stant until failure occurs. This condition is a resemblance to
a steady-state seepage in the real world where water accu-
mulates behind the walls of a slope for a period of time.
The condition is mainly seen in tailings dams in the mining
industries and natural slopes.
Experiment 2 was to raise the water level gradually
until failure occurs. At a slope height of 225 mm, the ini-
tial water level was set at 70 mm. The water level is kept
constant for 3 minutes and then raised a further 70 mm
to 140 mm slope level for another 3 minutes. The last
stage was raising and maintaining the water level steady at
215 mm high until failure. The final slope angle for the
model was 35°. The image shown in Figure 6. is the setup
for the experiment.
Figure 4. Final image of the box for experiments
Figure 5. Box of soil sample before experiment 1
Each water chamber plate bordering the slope model
has perforations in the plexiglass to allow water to flow in
and out of the chamber. The perforations are covered with
a fine mesh smaller than the sand particles to prevent any
sand particles from entering the water chamber but rather
to allow free flow of water through to the soil and prevent
flushing of soil particles.
A plain water hose connected to a water container was
used to fill in the water chambers. The water container can
be elevated or lowered to any level to change the pressure
heads. Changes in water level inside the model slope were
controlled by injecting water into the sand layer or drawing
water out of the sand layer through inlets and outlet points.
These helped in the various simulations of constant water
rise and drawdown of water.
3.4 Finite Element Modeling
Slope stability programs such as Slide 2D can be used to
find safety factors in both soil and rock slopes. The com-
puter program can compute circular and non-circular fail-
ure surfaces while evaluating the probability of failure. The
software is developed by Rocscience in Canada and can
be used for soil and rock slope stability analysis. “Slide2
is simple to use, yet complex models can be created and
analyzed quickly and easily. External loading, groundwater,
and support can all be modeled in various ways. Individual
slip surfaces can be applied to locate the critical slip sur-
face for a given slope. Users can also carry out deterministic
(safety factor) or probabilistic (probability of failure) analy-
ses” (Rocscience Slide, 2019).
The two-dimensional finite element method Rocscience
was used to analyze various boundary elements. The geom-
etry and soil properties of the model were the same for the
soil used for the laboratory experiment. The modeling was
used to find the failure planes, pressure heads, flowlines, and
flow vectors. Modeling in Slide2 was possible because it can
analyze groundwater, pore-pressure, steady-state, and tran-
sient state. In all the analyses, the mesh consisted of 3-node
triangular elements and approximately 1500 elements. The
default unit weight of 9.81kN/m3 was used for the density
of water and good soil characteristic curve and water con-
tent provided to help in the analysis. The numerical results
are compared to that of the laboratory test conducted.
3.5 Initial Conditions.
The model slopes were constructed in the middle cham-
ber and compacted in layers of 50 mm. The first layer was
spread uniformly and compacted to form the base of the
slope. The first experiment was to simulate steady-state flow.
The dimensions of the slope were 220 mm high, 100 mm
wide, and a crest of 70 mm. The final slope angle was about
35° (Figure 5). After constructing the model slope, the
upstream tank was filled gradually with water to the level
of 190 mm. The water was kept constant and allowed to
flow through the soil and seep to the downstream chamber
exit points. To initiate failure, the water level is kept con-
stant until failure occurs. This condition is a resemblance to
a steady-state seepage in the real world where water accu-
mulates behind the walls of a slope for a period of time.
The condition is mainly seen in tailings dams in the mining
industries and natural slopes.
Experiment 2 was to raise the water level gradually
until failure occurs. At a slope height of 225 mm, the ini-
tial water level was set at 70 mm. The water level is kept
constant for 3 minutes and then raised a further 70 mm
to 140 mm slope level for another 3 minutes. The last
stage was raising and maintaining the water level steady at
215 mm high until failure. The final slope angle for the
model was 35°. The image shown in Figure 6. is the setup
for the experiment.
Figure 4. Final image of the box for experiments
Figure 5. Box of soil sample before experiment 1