Proceeding of PILE 2017

September 26 – 27, 2017

SIGNIFICANCE OF PRESSUREMETER TEST FOR MODELING SECANT

PILE REINFORCED BY SOIL NAILING IN EXCAVATION

Aswin Lim Paulus Pramono Rahardjo

Lecturer Professor
Universitas Katolik Parahyangan Universitas Katolik Parahyangan
aswinlim@unpar.ac.id rahardjo.paulus@gmail.com

Adityaputera Wirawan
Geotechnical Engineer
Geotechnical Engineering Consultant
adityaputera.wirawan@gmail.com

ABSTRACT: This paper highlights the importance of pressuremeter test in determining the soil modulus for
modeling an excavation project which is located in South Jakarta. The excavation covered around ± 3673 m2
and it adopted secant piles reinforced by soil nailing as the retaining wall system. The average excavation depth
is 12.0 m. Several soil investigations have been conducted, especially the pressuremeter test, and the wall
deformation were well monitored with the inclinometer. Pressuremeter was used to measure the in-situ soil
modulus accurately. Later, based on the measured soil modulus, a 2D finite element analysis was conducted to
investigate the performance of the retaining wall system. The results show the measured soil modulus obtained
from the pressuremeter could be used a basis for determining soil modulus input parameters, rather than using
empirical correlations. The computed wall deflection was agreed with the measured data. Moreover, the
distribution of friction along the soil nailing was investigated. It was concluded that frictional resistance along
° inclination was mobilized more than the soil nailing with 30° inclination.
the soil nailing with 45
Keywords: Pressuremeter, Soil modulus, Secant pile, soil nailing, FEM.

INTRODUCTION

Excavations in South Jakarta area are usually constructed with a combination of secant piles and
soil anchors or soil nailing. Despite this retaining wall system is relatively cheaper than braced
excavation methods, soil conditions in South Jakarta also relatively good compared with soil
conditions in North Jakarta. For an excavation analysis, besides the soil shear strength, the soil
modulus is a predominant parameter that controlling the deformations induced by excavation (Lim
and Ou, 2017). In common practices, the soil modulus for designing an excavation was obtained
from empirical correlation, for example, the correlation to the N-SPT values ( E = β × N ). Although
the standard penetration test (SPT) procedure has been standardized (ASTM D1586), but the
accuracy is operator dependent. Since the value of β is empirical and the N value is inconsistent,
the determination of soil modulus becomes challenging. Most of the time, different consultants
have different opinions about the value of β, and yields different design of excavation methods. In
project owner point of view, the cheapest and safest excavation method is a priority to be selected.

The pressuremeter is one of an in-situ testing for soil where a cylindrical flexible membrane is
expanded inside a borehole. The pressuremeter enters the soil by pre-boring a hole into which the
probe is placed. Once in the soil, increments of pressure are applied to the inside of the membrane

1
2

forcing it to press against the soil and so loading a cylindrical cavity. The output of this test is a
radial stress-strain curve. One of the advantages of pressuremeter test is the avoidance of
empiricism in deriving soil properties. The obtained fundamental soil parameters if the test reaches
plastic zone are the ground pressure at rest (P0), the yield pressure (Py), the elastic modulus (EM),
the shear modulus (GM) and the undrained shear strength (Su). Pressuremeter and the schematic
test is illustrated in Fig 1. The main objective of this paper is to demonstrate the significance of
pressuremeter test in determining soil modulus, in which it is an important soil parameter in an
excavation modeling. Later, an excavation case was analyzed thoroughly via PLAXIS 2D and the
performance of the retaining wall system was investigated.

Figure 1. Pressuremeter device and the testing scheme

PROJECT DESCRIPTION

An excavation project was located in South Jakarta, the capital city of Indonesia, where it was
bordered by Kali Baru River at East and with the main road at West. This project consisted of 3 levels
of basement and 19 stories of upper structures. The average excavation depth was 12.0 m. The
earth retaining wall system was contiguous bored-pile with 800 mm diameter and 1200 mm spacing
(center to center) with the effective length was 23.5 m. In addition, soil nailing with 12 m length
was installed behind the retaining wall with an inclination of 30° and 45°, as illustrated in Fig 2, and
they were connected by a waler beam. Moreover, Figures 3a and 3b show the excavation location
and geometry, respectively.
 3

S o il
na ilin g
3 0° S o il
n a ilin g
3 0°
S oil
S o il n a ilin g
n a ilin g 4 5° 1
4 5° 2
m

dia =
Nail
20 0 mm
S p ace r

Con tig u ou s
bo re d p ile
dia = 80 0 mm

Wa le r be a m 1 20 0 mm 120 0 mm

Figure 2. Arrangement of secant piles and soil nailing

Figure 3. Excavation location and geometry

Soil stratification
According to two bore holes, the soil stratifications could be classified as follow: First layer (0 –
5 m) was silty clay with a medium consistency. This layer might be an overly consolidated layer due
to desiccation process. The second layer (5 – 20 m) was silty clay with soft to medium consistency.
Starting from GL -20 m, it was a silty sand layer. As indicated by the site investigation, the soil
parameters obtained via SPT and laboratory tests are summarized in Fig. 4. In addition, the
groundwater level was observed at GL -10 m.

Result of pressuremeter test

The result of pressuremeter test in BH-01 and BH-02 is plotted in Fig 5. In total, four tests were
conducted, that were at GL -6 m, GL -12 m, GL -18 m, and GL -24 m. All of the tests indicated soils
had reached plastic zone. The highest limit pressure was 30 kg/cm2 which was measured at the
cemented silty sand layer (GL -24 m). Meanwhile, for silty clay with medium consistency (GL -12 m
and GL -18 m), the limit pressure was in the range of 9 to 10 kg/cm2. In addition, the limit pressure
measured at GL -8 m was 4.5 kg/cm2. Table 1 summarizes all of the measured and interpreted
values from the pressuremeter tests. Furthermore, Fig 6a and 6b depict the value of OCR and Soil
4

Modulus along with depth, respectively. The original OCR data were calculated from 4 sets of
Oedometer test, and the rest were estimated as OCR = k × pa × (σv' / N SPT ) (Mayne and Kemper,
1988), where ks is constant from 1.0 to 0.2, and pa is atmospheric pressure. For the soil modulus
(EM), for data were measured directly from the pressuremeter test, meanwhile, the rest were
evaluated from an empirical correlation which is EM = (1500 to 2800 NSPT) with unit kPa. It should
be noted that the EM used for analysis was carefully evaluated according to the measured and
empirical data as shown in Fig 6a and 6b, respectively. In such a condition, the accuracy of soil
modulus could be optimized.
N-SPT t (kN/m3)
0 10 20 30 40 50 60 16 16.4 16.8 17.2

Silty
Clay

Silty
Clay

Sandy
Silt

Figure 4. Typical soil profiles and its parameters

35.00

30.00

25.00
Depth = 24 m
)
2
m
c/
20.00
gk
(
re 15.00
us
s Depth = 12 m
re 10.00
P Depth = 18 m

5.00

Depth = 6 m
0.00
35 36 37 38 39 40 41 42 43 44 45

R adius (mm)

Figure 5. Stress-strain curve obtained from pressuremeter test in BH-01 and BH-02

Table 1. Summary of the measured and interpreted values from the pressuremeter tests.
Depth PO PY PL EM su K0
No NSPT
(m) (kg/cm2) (kg/cm2) (kg/cm2) (kg/cm2) (kg/cm2)
BH-01 6.5-7.5 2 0.75 1.74 4.6 66.73 0.7 0.72
12.0-13.0 7 1.83 5.69 10.5 181.07 1.58 0.89
BH-02 18.0-19.0 6 2.13 5.56 9.25 348.68 1.29 0.57
24.0-25.0 50 4.25 18.2 39 1686.51 6.13 1.14
FINITE ELEMENT MODELING
 5

The Plaxis 2D finite element program was used to model the excavation project. Fig 7 depicts
the finite element mesh and the model boundary. The excavation depth was 12 m, and the
excavation width was 44 m. The analysis was followed the cross section in which the inclinometer
casing was located. Only a half of the excavation geometry was adopted in the analyses due to
symmetry. The soil movements were fixed at the bottom of boundary and restrained at the vertical
direction for the both sides. In addition, the distance between the retaining system and the outer
boundary of mesh was ensured to be larger than 2He (final excavation depth) to minimize boundary
effects. Moreover, fifteen-node triangular elements were used to simulate the soil cluster, 5-node
plate elements were used to model the retaining system and 10-node interface element was
applied to model the soil-plate element interaction behavior. The behavior of interface friction
(Rinter) between the structural elements and adjacent soils follows the Mohr-Coulomb model
Moreover, fifteen-node triangular elements were used to simulate the soil cluster, 5-node plate
elements were used to model the retaining system and 10-node interface element was applied to
model the soil-plate element interaction behavior.
OCR
0 2 4 6 8
0
EM (MPa)
4 0 10 20 30 40 50
8 0
12 For analysis
Depth (m)

4
16
20 For analysis 8 For analysis

24 12
28
16
32
36 20

(a) (b)

Figure 6. (a). The OCR and (b). soil modulus values for analysis

Figure 7.Typical finite element mesh and boundary used for analysis

Soil constitutive model, model parameters, and structural parameters

6

The Hardening Soil (HS) model (Schanz et al., 1999), an advanced hyperbolic soil model
formulated in the framework of isotropic hardening double surface plasticity, was adopted to
simulate the soil behavior. It required eleven material parameters
ref ref ref ref NC
( c ', φ ',ψ , E 50 , E oed ,E ur , m ,υur , p ,K o ,Rf ) and is summarized in Table 2. The strength parameters
were directly obtained from the Triaxial CU test. Meanwhile, the stiffness parameters of the HS
ref
model, such as the secant referential stiffness ( E 50 ), the tangent referential stiffness for primary
ref
oedometer loading ( E oed ), the unloading/reloading referential stiffness ( E urref ) were calculated from
the EM values. Indeed, the EM value was equivalent to the E50. In order to be used as an input
parameter in the HS model, E50 should be converted to the as proposed by Schanz et al. (1999),
as shown in Eq. (1).
m
 c '. cos φ '− σ 3' sinφ ' 
E 50 = E 50ref   (1)
 c '. cos φ '+ p 'sinφ ' 
When is determined, then and can be estimated as suggested by Calvello and Finno (2004).
The coefficient of the at-rest earth pressure for the NC layer was calculated from Jaky’s (1944)which
was and the coefficient of the at-rest earth pressure for the OC layer was estimated based on
Ladd’s (1977) which was K 0,oc = (1 − sinφ ')(OCR )sin φ ' .
The structural members, such as the contiguous bored pile and soil nailing were assumed to
behave as linear-elastic. The input parameters are listed in Table 3. The equivalent Young’s modulus
of soil nailing was calculated according to the fundamentals of material strength as shown in Eq.
(2).
A  A 
E eq = E nail  nail  + E grout  grout  (2)
A
 total  A
 total 
Table 2. Soil input parameters for analysis
φ' c' E 50ref ref
Depth Analysis γt E oed E urref
OCR m υur
(m) type (kN/m3) (°) (kPa) (kPa) (kPa) (kPa)
0-2 16.5 27 0 5 18689 13082 56067 1 0.2
2-5 16.5 27 0 3 8658 6061 25974 1 0.2
5-8 16.5 30 0 2 14345 10041 43034 1 0.2
8-11 Undrained 16.5 30 0 1.5 15949 11164 47847 1 0.2
11-14 16.5 30 0 1.1 19308 13516 57924 1 0.2
14-17 16.5 30 0 1 24778 17344 74333 1 0.2
17-20 16.5 30 0 1 27043 18930 81130 1 0.2
20-23 18 35 0 1 97590 97590 292771 0.5 0.2
23-26 Drained 18 35 0 1 139534 139534 418601 0.5 0.2
26-35 18 35 0 1 171062 171062 513186 0.5 0.2

Table 3. Structural input parameters for analysis

 7

Model
diameter (mm) spacing (m) EA (kN/m) EI (kNm2/m)

Contiguous Borpile 0.8 1.2 13210000 528300

Soil nailing 0.2 2.4 472500 1181

RESULT AND DISCUSSION

Wall deflections
Fig 8 shows the comparison between measured and computed wall deflections for this project.
It was clearly shown that the computed wall deflection yield a close result with the measured wall
deflection. The maximum measured and computed wall deflections were 45 mm and 40 mm,
respectively. Both wall deflections yielded a cantilever shape of deflection. It implies that the
selection of input parameters and the modeling procedure could reflect the field condition. One
thing should be emphasized is the soil modulus parameters which were obtained from
pressuremeter test is quite accurate to °capture the deformation characteristic of the secant pile.
Meanwhile, the soil strength parameters are directly obtained from the Triaxial CU test. Furthermore,
an analysis was performed to check the effectiveness of the installed soil nailing. If the soil nailing
were not installed, the maximum wall deflection increased double to around 80 mm. This indicates
the installed soil nailing worked properly and has a significant effect in reducing the wall deflection.
Fig 8 shows the comparison between measured and computed wall deflections for this project. It
was clearly shown that the computed wall deflection yield a close result with the measured wall
deflection. The maximum measured and computed wall deflections were 45 mm and 40 mm,
respectively. Both wall deflections yielded a cantilever shape of deflection. It implies that the
selection of input parameters and the modeling procedure could reflect the field condition. One
thing should be emphasized is the soil modulus parameters which were obtained from
pressuremeter test is quite accurate to °capture the deformation characteristic of the secant pile.
Meanwhile, the soil strength parameters are directly obtained from the Triaxial CU test. Furthermore,
an analysis was performed to check the effectiveness of the installed soil nailing. If the soil nailing
were not installed, the maximum wall deflection increased double to around 80 mm. This indicates
the installed soil nailing worked properly and has a significant effect in reducing the wall deflection.

Friction transfer mechanism along soil nailing

Figs 9a and 9b show the friction transfer mechanism along the soil nailing with 45° inclination
and 30° inclination, respectively. The plotted friction was the mobilized shear stress ratio along the
pile, and it follows Eq. (3) as:
τ mob
τ rel = (3)
τ ult
Where τult is the maximum shear stress of soils. When the mobilized shear stress ratio is equal
to unity, it implies that the frictional resistance along soil nailing has been fully mobilized.
As shown in Fig 9, the mobilized friction along the soil nailing was larger at the bottom side
compared with the top side. The possible reason was the soil right behind the secant pile and near
the soil nailing moved together when the wall deformed and causing the relative movement near
the top of the soil nailing became smaller than the bottom side. Meanwhile, the relative movement
8

at the bottom of the soil nailing moved freely without trapped by the secant pile. In this case, the
maximum capacity of friction was fully mobilized near the soil nailing bottom. In addition, the soil
nailing with 45° inclination experienced larger friction than the soil nailing with 30° inclination. It
implies that the soil nailing with 45° inclination might be more effective than the soil nailing with
30° inclination in controlling the deflection of the secant pile.
Wall deflections (mm)
0 10 20 30 40 50 60 70 80
0

10

12

14

16

18

20
Measured
22 Computed
Computed without soil nails
24
Figure 8. Comparison of measured and computed wall deflections

Top
Top

Exc to -5m
Exc to -5m
Exc to -8m
Exc to -12m
Maximum friction Exc to -8m
τ =1
re l
Exc to - Exc to -
Bottom
11m 12m
Maximum friction Exc to -
Bottom τ =1
re l 11m

(a) (b)

Figure 9. The mobilized frictional resistance along soil nailing with (a). 45° inclination and (b).
30° inclination
 9

CONCLUSIONS

Based on the performed analyses, the following conclusions can be drawn:

1. The soil modulus obtained from the pressuremeter test is reasonably accurate to be used for
modeling the deformation of secant pile induced by excavation. With combination with the
Hardening Soil model, the computed wall deflection yielded a close result with the field
measurement.
2. The soil nailing with 45° inclination might be more effective than the soil nailing with 30°
inclination because the larger mobilized frictional resistance along the soil nailing was detected
at the soil nailing with 45° inclination.
3. The mobilized friction along the soil nailing was larger at the bottom side compared with the
top side. The possible reason was the soil right behind the secant pile and near the soil nailing
moved together when the wall deformed and causing the relative movement near the top of
the soil nailing became smaller than the bottom side. Meanwhile, the relative movement at the
bottom of the soil nailing moved freely without trapped by the secant pile.

REFERENCES

ASTM D1586. Standard Test Method for Standard Penetration Test (SPT) and Split-Barrel Sampling
of Soils
Calvello ,M. and Finno, R. (2004). Selecting parameters to optimize in model calibration by inverse
analysis. Computer and Geotechnics 31(5):410-424
Jaky, J. (1944). The coefficient of earth pressure at rest. Journal for Society of Hungarian Architects
and Engineers, 78(22):355–358 (In Hungarian).
Ladd, C.C., Foott, R,, Ishihara, K., Schlosser, F. and Poulos, H.G. (1977). Stress-deformation and
strength characteristics. Proceedings of the 9th international conference on soil mechanics and
foundation engineering, (2):421–494:Tokyo.
Lim, A. and Ou, C.Y. (2017). Stress paths in deep excavations under undrained conditions and its
influence on deformation analysis. Journal of Tunneling and Underground Space Technology.
63:118-132.
Mayne, P.W.. and Kemper., J.B. (1988). Profiling OCR in Stiff Clays by CPT and SPT. Geotechnical
Testing Journal. DOI:10.1520/GTJ10960J
Schanz, T., Vermeer, P.A. and Bonnier, P.G. (1999). Formulation and verification of the Hardening-
Soil model. Beyond 2000 in Computational Geotechnics: Brinkgreve ed,Rotterdam Balkema:281-
290