2092
Proceedings of the 18
th
International Conference on Soil Mechanics and Geotechnical Engineering, Paris 2013
facing with a 1.2 cm annular rod and a small steel pin.
The monitoring system includes:
- A load cell between to the vertical jack and the plate to
measure the load;
- Three vertical displacement transducers recording the plate
settlements;
- A digital camera taking lateral images of the model during
the entire test. By applying the Particle Image Velocimetry
(PIV) technique (White et al. 2003) to the image sequences
it is possible to reconstruct the evolution of displacement
during the test;
- A laser scanner for monitoring the frontal displacement of
the face at some significant load steps. Since the scanner
takes about 1 min to complete the scansion, the loading
sequence must be temporarily stopped. A small load
reduction, due to the occurrence of the soil viscous strains,
was observed in this short time interval;
- Eight strain gauges, glued pair by pair, at 2.3, 10.4, 18.5 and
26.7 cm from the face, on the nails located along the central
vertical section at 15.3 and 25.4 cm from the top (the central
ones). They permit to reconstruct the distribution of axial
strain and, consequently, of axial stress along nails.
In order to evaluate which stiffness – i.e. the axial or the
flexional ones - mostly influences the soil nailing behaviour
during excavation and subsequent plate loading, six tests were
performed with different facing types (Table 1 summarises the
geometrical and mechanical properties of the various
coverings). Four facings, covering the entire excavated front,
were:
a
) 4 mm-thick plate of Polymethyl methacrylate
(PMMA);
b
) a 0.25 mm-thick sheet of brass (BRASS);
c
) a steel
mesh formed by 1-mm wires, perpendicularly welded at 6 mm
spacing (MESH);
d
) a steel net formed by 0.24 mm-diameter
wires, perpendicularly woven (NET). Three of these continuous
coverings have an axial stiffness with the same order of
magnitude but a flexional stiffness decreasing about one order
from one to another facing, while the fourth is very deformable
both in axial and flexional sense.
The other ones are two discontinuous facing constituted
by rectangular tiles in PMMA (obtained by cutting a PMMA
cover like that used in test
a
): in these cases the covering ratio,
defined as the ratio between the total covered area and the total
extension of facing, are respectively equal to 95% (PMMA95)
and 25% (PMMA25). Due to this discontinuity the axial
stiffness vanishes, so these covers are flexional stiff (like the
test
a
) but without any axial stiffness.
Since the soil forming the model is dry sand without
cohesion, to avoid the collapse of sand among the tiles or across
the mesh holes, a very light and low-resistant geo-synthetic
behind them was set up. The same geo-synthetic was also
inserted at the rear of the other facings to reach homogeneous
test conditions.
After the models being completely set up, they were
driven to failure in three steps: 1) application of a uniform load
q
of 24 kPa on the plate; 2) removing one by one of four
wooden blocks, simulating the front excavation; 3) application
of an increasing uniform load on the plate up to failure.
3 MODEL RESPONSE DURING EXCAVATION
Figure 2 depicts the vertical displacements of the plate,
y
p
,
during the 4 steps of excavations in the all the tests. Even if the
plate horizontal displacements,
x
p
, (obtained from the PIV
analysis of digital images) are not reported here for brevity, they
show a similar trend with the same magnitude order of vertical
settlements – i.e.
y
p
/
x
p
≈1. Moreover, since the plate is located
just at the rear of facing, the horizontal plate displacement may
be considered equal to the horizontal displacement of the front
tip.
Note that the vertical displacement does not exceed 0.5
mm (equivalent to 0.13% of the slope height) in tests
a
and
b
with very rigid facings (PMMA and MESH), while the
maximum displacement, equal to about 2 mm and equivalent to
0.5% of the height, is observed in the tests
d
and
e
with very
deformable facing (NET and PMMA25).
It is also interesting to observe that in the test with
PMMA95, with discontinuous covering, the displacement does
not exceed the 0.27% of the height: this means that the high
flexional stiffness of PMMA tiles prevents the soil near to the
face to move laterally.
Figure 3 reports the tensile force distribution along the
monitored nails at the end of the excavation. Even if the tensile
force is determined in few points, it is possible to recognize the
typical bell-shaped distribution observed in many applications
and described in the international practical guides (i.e. FWHA
2003, Geoguide7 2008). As known, the slope of the lateral
segments depends on the shear stresses mobilized at the
interface soil-nail in the active and passive zone respectively.
Table 1. Mechanical characteristics of facings adopted in the physical model.
Model
Covering
ratio
Thickness/
Facing
Wire Diam.
(mm)
(%)
Wire
spacing
(mm)
Young
modulus
E (GPa)
Axial stiffness
EA/m
(N/mm)
Flexional
stiffness EJ/m
(Nmm
2
/mm)
a
PMMA
100
4
-
3.2
12800
17066.67
b
MESH
100
1
6
210
26180
3318.06
c
BRASS
100
0.25
-
126
31500
236.25
d
NET
100
0.24
1.02
70
3105*
22.66
e
PMMA95
95
4
-
3.2
-
-
f
PMMA25
25
4
-
3.2
-
-
* The axial stiffness of canvas is the mean values obtained from to traction tests performed on two 178mm x 25mm samples.
Figure 1 Perspective view of 1g physical model.
