Stiff clay c’ ϕ’ derivation through pressuremeter test data

T. L. Gouw
Universitas Katolik Parahyangan, Bandung, Indonesia, [email protected]

ABSTRACT: Due to its rather brittle nature, retrieving undisturbed samples of Jakarta cemented greyish stiff clay, often
found at a depth of 30 to 120m, is very difficult. Good and reliable shear strength parameters, i.e, c and ϕ values, obtained
from triaxial test are hardly available. In practice, many engineers are often forced to estimate these parameters through
SPT test data which are of course greatly varied from one engineer to another. It will be good if these parameters can be
derived by an in-situ testing device. Since pressuremeter is an in-situ soil testing device able to yield stress strain rela-
tionship of soil, a research is carried out to derive c and ϕ values from pressuremeter test data curves through cavity
expansion theory. Results prove that c and ϕ values of Jakarta stiff clay can be derived by matching pressuremeter test
data curve with values calculated through modified cavity expansion theory. The derived c and ϕ values are comparable
with CIU triaxial test strength parameter obtained from relatively good ‘undisturbed’ samples.

Keywords: Jakarta stiff clay; shear strength; pressuremeter; cavity expansion theory

the membrane. The actual pressure or stress working

1. Introduction against the borehole wall is corrected against the mem-
brane resistance and against the hydrostatic pressure
The first 30m depth of Jakarta alluvial coastal plain from the manometer to the centre of the membrane. The
generally consists of alternating soft to medium silty clay volume of expansion is corrected against the expansion
and loose to medium dense sands layers. Below 30m and of the hose to deliver the water from the control unit to
up to 60m depth, alternating old alluvium deposits of ce- the membrane (ASTM D4719-00 [4]). The corrected vol-
mented stiff clay and cemented sands are found. Below ume is then converted into radial strain of the borehole
60m depth, the soil generally consists of stiff cemented wall. The resulting corrected radial stress strain curve
old alluvium clay. The cemented stiff clay layer exhibits acting against the soil can then be plotted. Fig. 1 shows
a rather brittle nature. It needs a highly skillful technician the schematic pressuremeter test device and the typical
to get and to prepare good quality of undisturbed samples test data curve modified after Briaud [7].
for triaxial test. Many soil laboratory does not have such
qualified person, therefore, engineers are often ‘forced’
to estimate its c and ϕ values, as well as its stiffness value
through SPT data which is of course can greatly varied
from one engineer to another as SPT test itself is often far
from reliable when executed on clayey soils. It is good if
these parameters can be derived by means of another
more reliable in-situ test. Since pressuremeter test is so
far the only in-situ testing device able to generate a stress
strain curve of in-situ soils, it hypothesis that by simulat-
ing the pressuremeter test through cylindrical cavity ex-
pansion theory and matching the resulting stress strain
curve with the actual test data curve, it is possible to cal-
culate the shear strength parameters, i.e., c and ϕ values.
An in-depth research is the carried out to investigate this
possibility. The methodology and the result of the re-
search are presented in this paper.

2. Literature Reviews

2.1. Pressuremeter test

Pressuremeter test, originally developed by Louis Mé-
nard in 1957 (Baguelin et al [1-2]; Gambin [3]), is per-
formed by lowering a cylindrical probe into a carefully
prepared borehole to the required test depth where the cy-
lindrical membrane is then pressurized against the bore-
hole wall and the subsequent volume expansion of the
cylindrical membrane is measured.
Generally, the pressure is applied by pumping de-aired
water into the cylindrical membrane and its volume of Figure 1. Pressuremeter test and typical test graph modified after
Briaud [7]
expansion is measured through the volume pumped into
 From the installation point of view, there are three
types of pressuremeter, i.e.: preborehole pressuremeter,
push-in pressuremeter, and self-boring pressuremeter
(Baguelin et al [2]; Clayton et al [6]; Briaud [7]; Clarke
[8]).
Preborehole pressuremeter is performed on a carefully
prepared borehole prior to the insertion of the test probe.
Push-in pressuremeter probe is equipped with a conical
tip and the probe is pushed into the soil to the test depth
before the test is begun. Self-boring pressuremeter probe
is equipped with a drilling bit at its tip, the probe is
advanced into the ground following the drilling process.
Push-in pressuremeter induces more disturbances to the
borehole wall than pre-borehole pressuremeter, self- Figure 2. Extrapolation of test data to get PL (modified after Baguelin
et al [2] and Ghionna et al [10])
boring pressuremeter induces less disturbances.
Among those three types, pre-borehole pressuremeter
has been employed in Indonesia as early as 1983 (Gouw 2.2.4. Coefficient of horizontal subgrade
[9]) and up to now it is the most commonly adopted. reaction, Km
Push-in pressuremeter is never applied, and only one
government institution has self-boring pressuremeter but Coefficient of horizontal subgrade reaction, Km, is
rarely used. Therefore, the research is limited to pre- obtained through linear part of the test curve, i.e.:
borehole pressuremeter test data only.
∆P 𝑃𝑦 −𝑃𝑜
Km= ∆R
=
𝑅𝑃𝑦 −𝑅𝑃𝑜
(3)
2.2. Pressuremeter parameters
Where RPy and RPo is cavity radius at yield pressure Py
From the stress strain pressuremeter test curve, tradi-
and horizontal pressure, Po, respectively.
tionally six parameters are generated, those are: Po, Py,
PL, Km, Em, and G as elaborated below:
2.2.5. Stiffness or deformation modulus, Em
2.2.1. Horizontal pressure, Po Soil stiffness modulus, Em, is derived by the following
Horizontal pressure, Po, is defined as the pressure equation:
where the membrane first touches the borehole wall, i.e. R Po +R Py
first point at the beginning of linear or elastic part of Em = (1+υ) 2
Km (4)
pressuremeter stress strain curve, indicated as Po in Fig.1.
This pressure is also often interpreted as total horizontal Where v is Poisson ratio of the soil, usually taken as
pressure at rest as represented in the equation below, 0.33.

𝑃𝑜 = 𝜎′𝑣𝑜 𝑘𝑜 + 𝑢𝑜 (1) 2.2.6. Shear modulus, G

𝜎′ℎ𝑜 = 𝜎′𝑣𝑜 𝑘𝑜 = 𝑃𝑜 − 𝑢𝑜 (2) Shear modulus, G, is calculated as follows:

Where σ’vo is vertical effective pressure, σ’ho is G=

Em
(5)
2(1+υ)
horizontal effective pressure, ko is at rest horizontal earth
pressure coefficient, uo is hydrostatic groundwater
pressure. 2.2.7. Shear strength parameters
Other than those traditional pressuremeter parameters,
2.2.2. Yield pressure, Py some researchers derive undrained shear strength of clays
The end of the linear part and the beginning of the non- and angle of internal friction of sands. Gibson and
linear (plastic) part of the pressuremeter test curve, Anderson [11], formulized undrained shear strength of
indicated as Py in Fig. 1. cohesive soils, Su, through Tresca constitutive model as
follows:
2.2.3. Limit pressure, PL G ∆V
c = σho +Su [1+ln (S ) +ln ( V )] (6)
u
Limit pressure, indicated as PL in Fig. 1, is defined as
the ultimate horizontal pressure of pressuremeter test Where σc is pressuremeter cavity wall pressure, σho is
curve where soil start to ‘flow’, i.e. radial strain keeps on total horizontal stress of the soil, ∆𝑉 ⁄𝑉 = 𝐿𝑛[(𝑅𝑐 2 −
increasing while the pressure is hardly increases.
Generally, in real test, this limit pressure is hardly 𝑅𝑃𝑜 2 )/𝑅𝑐 ]= volumetric strain, Rc is the cavity wall radius
achieved, and the test curve must be extrapolated in a at σc.
logaritmic plot as shown in Fig. 2. With Eq. (6), Gibson and Anderson indicated that if σc
is plotted against 𝑙𝑛(∆𝑉/𝑉), the plastic part of the test
data will form a straight line, and the slope of this straight analysing deep foundation, tunnels, borehole stability,
line is the undrained shear strength of the soil as show in stone columns and other geotechnical problems (Ladanyi
Fig.3. [35]; Baguelin [1]; Vesic [36]; Vesic [37]; Baligh [38];
Wroth and Windle [39]; Datye and Nagraju [40]; Pandit
et al [41]; Yu 2000 [42]).
Vesic [36] divided the expansion of cylindrical cavity
into elastic and plastic zone as illustrated in Fig. 4. Based
on this Vesic work and Mohr Coulomb failure criterion,
Mecsi [34] derived equations to calculate the c and ϕ val-
ues of soil from pressuremeter test data. His equations are
elaborated below:

Figure 3. Derivation of undrained shear strength, Su (modified after

Gibson and Anderson [11])

Marshland and Randolph [12] derived a simpler

formula to derived the undrained shear strength, Su:

Su = (𝑃𝐿 − 𝑃𝑜 ) ∕ 𝑁𝑃 (7)

Where Np is a constant taken as 6.2. Mair and Wood

[13] reported that Eq. (7) is less sensitive against the
effect of pressuremeter installation method. Mair and
Wood [13], Windle and Wroth [14-15], reported that
generally undrained shear strengths derived from
pressuremeter test were higher than laboratory test data.
Some researchers have also tried to derive angle of
internal friction of sands through pressuremeter test data ro = initial cavity radius; ru = pressurised radius;  = radius of
(Hughes et al [16]; Fahey and Randolph [17]; Mair and plastic zone; uro = cavity radial displacement; u = radial
Wood [13]; Schnaid [18]). As in undrained shear displacement of plastic zone; r = radial stress; t = tangential
stress;  = radial stress at boundary of plastic Zone
strength case, the resulting friction angles were also Figure 4. Cylindrical cavity expansion zone (modified after Mecsi
higher than the ones obtained from triaxial tests (Bruzzi [34])
et al [19]). However, from literatures published up to 2.𝑐
2013, it seems that no attempt has been tried to derive the 𝜎𝑢 = (8)
√𝜉
effective shear strength c‘ and ϕ‘ of clayey soils by using
pressuremeter test data. In France, where pressuremeter 1−𝑠𝑖𝑛𝜙
is widely used, the engineers directly adopted the 𝜉= (9)
1+𝑠𝑖𝑛𝜙
pressuremeter parameters Em, G, Py and PL to calculate
foundation capacity and settlement (Baguelin et al [2]; σu and ξ are derived from Mohr circle as illustrated in
Gambin [20]; Boumedi et al [21]; Bustamante et al [22]; Fig. 5.
Gambin and Frank [23]; Reiffsteck [24]; Schlosser et al
[25]; Hamidi et al [26]).
Outside French, the research generally directed to
derive and to use undrained shear strength (Briaud et al
[27]; Jefferies [28]; Ferreira and Robertson [29]; Briaud
[30]; Bullock [31]; Silvestri and Ghassan [32]; Ramdane
et al [33]). Only one literature is found trying to estimate
c‘ and ϕ‘ values of clayey soils by using pressuremeter
test data through cavity expansion theory by Mecsi [34].
The cavity expansion theory and Mecsi model is
elaborated below.

2.3. Cylindrical cavity expansion theory

Cylindrical cavity expansion theory discussed the
change of stresses, pore water pressure and deformation
of soils caused by expansion or contraction of cylindrical Figure 5. Definition of σu and ξ from Mohr failure criterion (after
cavity. This theory has been widely employed in Mecsi [34])
 The relationship of soil stiffness vs deformation
modulus (soil sttiffness) is assumed to have a power The induced radial displacement ur:
function as follows.
∆𝜀𝑟(𝑖−1) +∆𝜀𝑟(𝑖)
∆𝑢𝑟 = (𝑟(𝑖) − 𝑟(𝑖−1) ) (16)
σ a 2
Es = Eo (σ 𝑐 ) (10)
ref
With the above formulas, it is supposed to be able to
Where Es is the deformation modulus at a cavity derive the c and ϕ of clayey soils by matching the
pressure of σc, and Eo is the deformation modulus at a pressuremeter test data curve the calculated radial strain
reference pressure σref = 100kPa as shown in Fig. 6, a is or radial displacement, i.e. matching σc vs εr plot from
named as rigidity index. pressuremeter against σc vs εr plot from the above cavity
expansion formulas.

3. Field and laboratory test performed

The research was carried out at a project site at central
Jakarta area, where many high-rise buildings are located.
The following field and laboratory testings were carried
out:
• 21 deep borings carried out between 90 to 120
m depths. SPT tests were taken at every 2 to
3.5 m intervals.
• 20 pre-borehole pressuremeter tests conducted
at cemented stiff clay layers.
• A total of 123 undisturbed samples for labora-
tory index properties tests, triaxial UU, triaxial
Figure 6. Deformation modulus vs cavity pressure (after Mecsi [34]) CIU and consolidation tests.
When a cylindrical cavity is subjected to a cavity
Fig. 7 to 15 show index and engineering properties of
pressure σc, the radial stress reduces from σc at the cavity
the soil tested. The subsoil exhibits an increasing trend of
wall to at rest horizontal earth pressure of σ‘ho at a certain
SPT blow counts against depth, and the water contents
distance. At the same time, the induced tangential stress
below 20m falls near the plastic limits, an indication of
σc first reduces until radius ρ, where it then increases to
stiff clays (Fig. 7). Plasticity indices of the stiff clay,
finally reaches σ‘ho. The radius where the soil is still in
found below 20m depth, are mostly within 20 to 60%
compression and the tangential stresses still reduces is
with liquidity indices less than 0.25 (Fig. 8). Other index
defined as radius of compression (plastic) zone, ρ, and
properties are shown in Fig. 9 to 11.
formulated as:
Fig. 12 shows the compression and re-compression in-
1+𝑠𝑖𝑛𝜙 dices, Fig. 13 shows the pre-consolidation pressure and
𝜎𝑐 +𝑐.𝑐𝑜𝑡𝜙 2𝑠𝑖𝑛𝜙 the undrained shear strength of the stiff clay from triaxial
𝜌 = 𝑟𝑐 ( ) (11)
𝜎𝜌ℎ +𝑐.𝑐𝑜𝑡𝜙 UU tests. The pre-consolidation pressures appear to be
increasing with depth. Comparing with the correspond-
Where rc = cavity radius at cavity pressure σc and σρh ing effective stresses, the over consolidation ratio of the
is horizontal or radial stress at boundary of compression stiff clay layers if found to be around 2.0. The undrained
zone which is defined as: shear strengths of the stiff clay layer are found to be in-
creasing with depth, from around 90 kPa at 24 m depth
σ'ho σ to about 300 kPa at 100 m depth. The effective and total
σρh ≈ [1+ξ-√(1+ξ)2 -2a(1-ξ)2aξ 'u ] +σ'ho (12) shear strength obtained from triaxial CIU tests are shown
a σ ho
in Fig. 14 and 15. Fig. 16 shows typical graphs of pres-
The radial stress inside the compression zone (at suremeter test performed. Fig. 17 shows a bad quality
radius r ≤ ρ): pressuremeter test data where the hole is too large.
Note that the notation of PMT DB-xx/yy in the pres-
2sinϕ suremeter graphs mean pressuremeter test (PMT) con-
c ρ c
σr = (σρh + tanϕ) . ( r )1+sinϕ - tanϕ (13) ducted at borehole no xx at depth of yy meter.
Fig. 18 shows the parameters derived from the pres-
suremeter tests. The notations on the graphs are as de-
The radial stress outside the compression zone (at
fined before, whereas Eur means unloading-reloading de-
radius r > ρ):
formation modulus derived from the test. Fig. 19 shows
the soil effective horizontal stress σ’ho is obtained by sub-
′ ′ ρ 2
σr =(σρh -𝜎ℎ𝑜 ). ( ) +𝜎ℎ𝑜 (14) tracting pressuremeter total horizontal pressure Po, with
r
its corresponding hydrostatic groundwater pressure, as
The induced radial strain, εr: formulated in Eq. (1). It is important to observe the
𝜎𝑟𝑒𝑓 𝜎𝑟
1−𝑎 ′
𝜎ℎ𝑜
1−𝑎 profile of effective horizontal stress as it needs to be im-
∆𝜀𝑟 = [( ) −( ) ] (15) plemented in Eq. (12), Eq. (14), and Eq. (15).
(1−𝑎)𝐸𝑜 𝜎𝑟𝑒𝑓 𝜎𝑟𝑒𝑓
 Wp Wn WL Wp, Wn, WL (%)
SPT N
0 10 20 30 40 50 60 0 10 20 30 40 50 60 70 80 90 100
Specific Gravity, Gs Water Content, Wn (%)
0 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 10 20 30 40 50 60 70 80 90 100
0 0
-10 -10
-10 -10
-20 -20
-20 -20
-30 -30
-30 -30

-40 -40
-40 -40

-50 -50
-50 -50

-60 -60
-60 -60

Depth (m)

Depth (m)
Depth (m)

Depth (m)
-70 -70 -70 -70

-80 -80 -80 -80

-90 -90 -90 -90

-100 -100 -100 -100

-110 -110
-110 -110

-120 -120
-120 -120

-130 -130
-130 -130

Figure 7. SPT blow counts and Atterberg Limits Figure 10. Specific gravity and water content

Void Ratio, e Degree of Saturation, Sr

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0 10 20 30 40 50 60 70 80 90 100
Plasticity Index, PI (%) Liquidity Index, LI 0 0
0 20 40 60 80 100 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
0 0 -10 -10

-10 -10 -20 -20

-20 -20
-30 -30

-30 -30
-40 -40

-40 -40
-50 -50

-50 -50
-60 -60
Depth (m)

Depth (m)
-60 -60
Depth (m)

Depth (m)

-70 -70
-70 -70
-80 -80
-80 -80
-90 -90
-90 -90

-100 -100
-100 -100

-110 -110
-110 -110

-120 -120
-120 -120

-130 -130 -130 -130

Figure 8. Plasticity and liquidity indices Figure 11. Void ratio and degree of saturation

Dry Unit Weight, g d (kN/m3) Bulk Unit Weight, g (kN/m3)

Compression index, Cc Recompression index, Cr
0.0 0.5 1.0 1.5 2.0 0.00 0.05 0.10 0.15 0.20
0 2 4 6 8 10 12 14 16 18 20 22 0 2 4 6 8 10 12 14 16 18 20 22
0 0 0 0

-10 -10 -10 -10

-20 -20 -20 -20

-30 -30 -30 -30

-40 -40
-40 -40

-50 -50
-50 -50

-60 -60
Depth (m)

Depth (m)

Depth (m)
Depth (m)

-60 -60
-70 -70
-70 -70
-80 -80
-80 -80
-90 -90
-90 -90
-100 -100

-100 -100
-110 -110

-110 -110
-120 -120

-130 -130 -120 -120

Figure 9. Dry and bulk unit weight -130 -130

Figure 12. Compression and recompression indices

 P'c (kPa) SuTXUU (kPa) 2500
0 500 1000 1500 2000 2500 3000 0 100 200 300 400 500
0 0 PMT DB-07/27 (a)
-10 P'c_Oedometer -10

Linear
-20 (P'c_Oedometer) -20 2000
Linear (Eff.Ver.Stress)

-30 -30

Corrected Pressure , kPA

-40 -40

-50 -50 1500

Depth (m)
Depth (m)

-60 -60

-70 -70
Py = 1043 kPa
-80 -80 1000

-90 -90 PMT_Data

-100 -100 Po-Py Linear Line

y = -0.0557x
500
-110 R² = 0.6334 -110 Po = 425 kPa
-120 -120

y = -0.1261x
-130 -130
R² = 0.9324
Figure 13. Pre-consolidation pressure and undrained strength 0
34 36 38 40 42 44 46 48
Borehole Radius, R (mm)
C'TXCU (kPa) f'TXCU (degree)
2500
0 50 100 150 200
0 10 20 30 40 50
0 0 PMT DB-01/35 (b)
-10 -10

-20 -20 2000

-30 -30

-40 -40
Corrected Pressure , kPA

-50 -50
1500
Depth (m)

Depth (m)

-60 -60

-70 -70

-80 -80 Py = 1046 kPa

1000
-90 -90

-100 -100

-110 -110
500
PMT_Data
-120 -120 TX_CU_Eff.Stress
Po = 471 kPa
TX_CU_Eff.Stress
Po-Py Linear Line
-130 -130

Figure 14. c’ and ϕ’ from triaxial CIU tests

0
34 36 38 40 42 44 46 48
CuTXCU (kPa) f uTXCU (degree) Borehole Radius, R (mm)
0 50 100 150 200 250 0 10 20 30 40 50
0 4000
0

-10
PMT DB-04/56 ((c)
k)
-10
3500
-20 -20

-30 -30
3000
-40 -40
Corrected Pressure , kPA

-50 -50 2500

Depth (m)

Depth (m)

-60 -60
Py = 2121 kPa
-70 -70 2000

-80 -80
1500 PMT_Data
-90 -90

Po-Py Linear Line

-100 -100
1000 Po = 892 kPa
-110 -110

-120 -120
TX_CU_Total_Stress TX_CU_Total_Stress 500
-130 -130

Figure 15. cu and ϕu from triaxial CIU tests 0

34 36 38 40 42 44 46 48
Borehole Radius, R (mm)

Figure 16. Good pressuremeter test graphs

 4000
PMT DB-09/66 ((d)
o) (i)
3500

3000
Corrected Pressure , kPA

2500
Py = 2364 kPa

2000

1500 PMT_Data
Po-Py Linear Line

1000 Po = 893 kPa

500

0
34 36 38 40 42 44 46 48
Borehole Radius, R (mm)

4000
Figure 17. Bad pressuremeter test graph
PMT DB-03/86 ((e)
r) Po, Py, PL (kPa) Em, Eur (kPa)
3500 0 1000 2000 3000 4000 5000 0 50000 100000 150000 200000
0 0
Po Em
-10 -10
3000 Py
Eur
-20 PL -20
Corrected Pressure , kPA

2500 -30 -30

-40 -40

2000 Py = 1987 kPa -50 -50

-60
Depth (m)

Depth (m)
-60

1500 -70 -70

-80 -80

1000 -90 -90

Po = 1212 kPa
PMT_Data -100 -100
500 Po-Py Linear Line -110 -110

-120 -120
0
-130 -130
34 36 38 40 42 44 46 48
Borehole Radius, R (mm) Figure 18. Pressuremeter parameters

4000 'vo(kPa) 'ho= Po - uo(kPa)

PMT DB-03/97 ((f)
t) 0 200 400 600 800 1000 0 200 400 600 800 1000
0 0
3500
-10 -10

3000 -20 -20

-30 -30
Corrected Pressure , kPA

2500 -40 -40

Py = 2017 kPa -50 -50

2000
Depth (m)

Depth (m)

-60 -60

-70 -70
1500
-80 -80

1000 Po = 1450 kPa -90 -90

PMT_Data -100 -100

500 Po-Py Linear Line -110 -110

y = -0.1261x
-120 -120 y = -0.1988x
0
-130 -130
34 36 38 40 42 44 46 48
Borehole Radius, R (mm) Figure 19. Effective vertical and horizontal pressure (horizontal pres-
sure calculated from Eq. (2))
Figure 16. Good pressuremeter test graph (con’t)
4. c and ϕ derivation from pressuremeter test While the test data curve is the same, each of the dia-
gram in Fig. 20 shows different combination of rigidity
index and c – ϕ values. The same results were obtained
4.1. Mecsi model from other test data. It is clear that Mecsi model needs to
Mecsi model as presented by Eq. (8) to (16) was tried be investigated further.
to derive the c and ϕ from pressuremeter test data. It was
found that Mecsi model could not give a unique values c 4.2. Modified E function model
and ϕ, neither match the test data curve, especially on the
plastic phase of the curve, i.e. the part after yield pressure Through many cycles of numerical study applying the
Py. The typical results are shown in Fig. 20. pressuremeter data to cylindrical cavity expansion model
4000 proposed by Mecsi, it was found that the key factor lies
( i) on changes of deformation modulus with its
3500
corresponding stress level as expressed in Eq. (10).
Among the parameters in Eq. (10), it is obvious that
further investigation need to be carried out on the
3000
reference pressure σref and rigidity index a. As explained
above, Mecsi proposed to use reference pressure of 100
Radial Stress, kPA

2500
kPa, considering that Mecsi method can only approach
the linear part of the pressuremeter curve and far from
2000
approaching the plastic part, it looks like the reference
pressure σref, should be adjusted as follows:
1500

PMT DB-09/66
• From Po to Py, where the pressuremeter stress-
1000 strain or stress-deformation curve still linear,
Mecsi_a=0.5_c=0_phi=32deg
the reference pressure of 100 kPa and a
500 constant value of rigidity index a shall be
39 42 44 47 49
Cavity Wall Radius, R (mm)
taken.
4000 • At and above yield pressure Py, where the
( ii ) curve starts to show non-linear characteristic,
3500
the reference pressure shall be taken equal to
the yield pressure, i.e. σref = Py, and rigidity
3000
index values needs to be adjusted in
accordance with their stress-strain level.
Radial Stress, kPA

2500
A further trial and error parametric studies was carried
out, it was found that at the linear or elastic part of the
2000
pressuremeter curve, the rigidity index, a, indeed
constant, and the values lies within 0.25 to 0.80 with an
1500 average 0.5. However, when entering the non-linear
plastic part, apart from changing the reference pressure
PMT DB-09/66
1000 from 100 kPa to Py, it appears that Eq. (10) needs to be
Mecsi_a=0.9_c=0_phi=21deg
modified. After many rounds of investigation, rather than
500 doing trial and error, intuitively an idea came to mind to
39 42 44 47 49
Cavity Wall Radius, R (mm)
search the changes of plastic deformation modulus
through its corresponding strain from within the test data
4000
( iii )
itself. Subsequently, Eq. (10) is then modified to:
3500
• When pressuremeter stress level is still within
the linear range, i.e. within Po to Py, Eq. (10)
3000
becomes:
Radial Stress, kPA

2500 σ 0.5
c
Es = Eo (100 ) (10a)
2000

• When pressuremeter stress level is above yield

1500
pressure Py,
aye aye
σ σ
Esy = Eyo ( Pcy) →Esy = my Eo ( Pcy) (10b)
PMT DB-09/66 y y
1000
Mecsi_a=0.81_c=0_phi=18deg
Where :
500
39 42 44 47 49 Es = elastic soil deformation modulus at cavity
Cavity Wall Radius, R (mm) pressure of σc
Figure 20. No unique c – ϕ values by Mecsi model Eo = Em = pressuremeter modulus as defined in Eq. (4)
Esy = plastic deformation modulus = σcy/εy = cavity It is possible to calculate parameter c and ϕ in a more
pressure at plastic part divided by its corresponding consistent way by using Eq. (8) to (16) but replacing Eq.
strain (from pressuremeter test data) (10) with Eq. (10a) and (10b). Fig. 22 shows one of the
Eyo = my.Eo = my.Em results of pressuremeter test curve matching with curve
my = yield factor calculated from the modified E function or modified
σcy = cavity pressure at and above yield pressure cavity expansion model. The result shows that when the
aye = rigidity factor after yield pressure stiff clay is still in linear “elastic“ range, the shear
strength consists both cohesion and angle of internal
To find both my and aye, Eq. (10b) then is normalized friction (since the parameters are derived from
to: pressuremeter, it is notated as cPMT and ϕPMT). However,
once the soil entering non-linear plastic part, the stiff clay
aye
Esy σ lost its cohesion (cyPMT = 0), and only the angle of internal
= 𝑚𝑦 ( Pcy) (10c)
Em y friction remains constant throughout the elastic and
plastic phase, i.e. ϕPMT = ϕyPMT. The same outcomes are
Then from the pressuremeter data calculate and plot found from all the pressuremeter test.
Esy/Em vs σcy/Py, the parameter my and aye can then be
obtained by running the power function regression
analysis. Fig. 21 shows one of the plotted test data. In this 4.3. Refining with linear hyperbolic model
case, my = 0.6151 and aye = -2.06. Once parameter my and
aye are found, substitute these parameters to Eq. (10b). Due to the manual reading process, the rather fast
execution may yield some degreeof erratic readings.
Since between Po and Py pressuremeter test curve shows
1.0 linear behaviour, while after Py the test curve shows
hyperbolic characteristics, a rather erratic test curve can
0.9
PMT DB-09/66 be smoothened out with a mathematical approach as
0.8 follows:
0.7
• When pressuremeter stress level is still linear,
i.e.within Po to Py, the test data can be
Esy / Em

0.6
linearized by linear equation:
0.5 y = 0.6151x-2.06
R² = 0.9249 𝜎𝑐 = 𝜆 𝜀𝑐 + 𝜅 (17)
0.4

0.3 where σc is cavity wall pressure, εc is cavity

0.2 wall strain, λ and κ is linear line coefficients.

0.1 • When pressuremeter stress level is above yield

0.0
pressure Py, hyperbolic equation is used:
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0
𝜀𝑐𝑦 𝜀𝑐𝑦
 cy /Py cy = or = 𝛼 + 𝛽𝜀𝑐𝑦 (18)
𝛼+𝛽𝜀𝑐𝑦 cy
Figure 21. Finding my and aye from pressuremeter test data
Where σcy is after yield cavity wall pressure εcy
is after yield cavity wall strain, α and β is
hyperbolic coefficients. With this hyperbolic
equation, the limit pressure PL can also be
calculated:

𝑃𝐿 = 1⁄𝛼 + 𝑃𝑜 (19)

Fig. 23 shows the example of smoothing the test data

with linear hyperbolic mathematical model.
By smoothing out pressuremeter test data point with
linear-hyperbolic model, it is then possible to make a
simple program by use of Microsoft Excel spreadsheet to
calculate cPMT and ϕPMT in semi-automatic way. With this
method, all the pressuremeter data were re-analysed and
the results are shown below.
In Fig. 24 cPMT and ϕPMT are plotted against the
undrained shear strength of the stiff clay obtained from
triaxial CIU test. While in Fig. 25 the cPMT and ϕPMT are
plotted against the drained shear strength of the stiff clay
Figure 22. Test data vs modified cavity expansion theory obtained from triaxial CIU test.
 5. Undrained shear strength
By employing Gibson and Anderson [11], Marshland
and Randolph [12] methods presented in section 2.2.7,
undrained shear strength of the stiff clay is derived from
the pressuremeter test. Fig. 26 shows example of the
derivation with both methods. Results obtained from all
the data are plotted in Fig. 27 together with the UU
triaxial test data. Gibson and Anderson method undrained
shear is 2 – 3 times higher than UU triaxial, while
Marshland and Randolph method gives 1.5 times higher
undrained shear strength.
4500

4000 Marshland & Randolph (1977)

Cu = (PL - Po) / Np
3500 Cu = (4210-893)/6.2 =535 kPa

Corrected Pressure, rc (kPA)

3000

2500
Gibson &
Figure 23. Smoothen out test data using linear hyperbolic mathemati- Anderson (1961)
cal model 2000

1500
CPMT , CuTXCU (kPa) fPMT, fuTXCU (degree)
0 50 100 150 200 250 300 350 400 450 0 10 20 30 40 50 1000
0 0
500 PMT DB-09/66
-10 -10

-20 -20 0
0.01 0.02 0.05 0.14 0.37 1.00
-30 -30 Ln (DV/Vo) = Ln [(Ri2-Ro2)/Ro2]
-40 -40 Figure 26. Undrained shear strength derivation from pressuremeter
-50 -50 test
Depth (m)

Depth (m)

-60 -60 Gibson & Anderson method Marshland & Randolph

-70 -70
Cu (kPa) method Cu (kPa)
0 200 400 600 800 1000 0 200 400 600 800 1000
0 0
-80 -80

-10 -10
-90 -90
y = -0.3826x
R² = 0.8136 -20 -20
-100 -100

-30 -30
-110 PMT_LinHyp -110 PMT_LinHyp
-40 -40
-120 TX_Tot.Stress -120
TX_Tot.Stress
-50 -50
-130 -130
Depth (m)

Depth (m)

Figure 24. cPMT and ϕPMT vs triaxial CIU undrained strength -60 -60

-70 -70

-80
CPMT , C'TXCU (kPa) fPMT, f'TXCU (degree) -80

0 50 100 150 200 250 300 350 400 450 0 10 20 30 40 50 -90 -90
0 0
-100 -100
-10 -10
-110 -110
-20 PMT PMT
-20
-120 -120 TX_UU
-30 TX_UU
-30
-130 -130
-40 -40
Figure 27. cuPMT vs cu triaxial UU
-50 -50
Depth (m)

6. Discussion
Depth (m)

-60 -60

-70 -70

-80 -80 Comparing Fig. 24 and Fig. 25, it can be seen that from
-90 -90
27m to 97m depth the ϕPMT values are within 20° – 30°.
y = -0.3826x
-100 R² = 0.8136 -100
These values fall within the drained angle of internal
friction rather than the undrained angle of internal
-110 PMT_LinHyp -110 PMT_LinHyp
friction. As for the cohesion, the cPMT values have a clear
-120 -120
TX_CU_Eff.Stress TX_CU_Eff.Stress trend increasing with depths, starting from around 50 kPa
-130 -130 at 27 m to around 250 kPa at 97m depth, and it is clearly
Figure 25. cPMT and ϕPMT vs triaxial CIU drained strength higher than the values obtained from CIU triaxial test, be
the undrained or drained cohesion. The lesser values of
cohesion from triaxial tests are generally attributed to the
brittle nature of Jakarta cemented stiff clay which tends Acknowledgement
to have thin hair cracks resulted from the sampling
process by thin wall tube sampler and during the The author would like to thank Prof. Paulus P.
preparation of the samples in the laboratory. The higher Rahardjo and Prof. A. Aziz Djajaputra for their
values of cPMT is due to the cemented nature of the Jakarta continuing and valuable guidance during the research and
stiff clay. Comparing Fig. 25 and Fig. 27, the cPMT values the preparation of this manuscript. A special
against the corresponding values of cuPMT, it is clear that acknowledgement also covered to Prof. H. Moeno, R.
cPMT values are lower by around 1.5 to 4.0 times the Karlinasari PhD and S. Herina for their feedbacks.
cuPMT. This is consistent with the nature that drained
cohesion is lower than undrained cohesion. It can be said References
that the cohesion values derived from pressuremeter test
by modified cavity expansion theory are drained [1] Baguelin, F., Jezeqel, J.F., Lemee, E., and Le Mehaute, A. (1972)
“Expansion of Cylindrical Probes in Cohesive Soils”, JSMFE,
cohesion or at least partially drained cohesion. ASCE, Vol. 98; SM11. Proc. Paper 9377, pp1129-1142
Another fact need to mention is that the derivation of [2] Baguelin, F., Jezequel, J.F., and Shields, D.H. (1978) “The Pres-
shear strength parameters from pressuremeter data by suremeter and Foundation Engineering”, Trans Tech Publication,
using modified cavity expansion give a clear existence of Switzerland
[3] Gambin, M. (1980) “A Review of the Menard Pressuremeter over
soil cohesion when the stress strain of the stiff clay is still the Last Twenty Years in Europe”, Sol Soils, No. 32, Paris
within the linear “elastic“ range, and the stiff clay losses [4] ASTM D4719-00. (2000) “Standard Test Method for Prebored
the cohesion once the stress level reaching and above its Pressuremeter Testing in Soils”, ASTM, 100 Barr Harbor Drive,
yield stress level, what remain thereafter is the angle of West Conshohocken, PA 19428-2959, USA
[5] Briaud, J.L. (2013) “Geotechnical Engineering: Unsaturated and
internal friction which remain constant throughout all the Saturated Soils”, John Wiley & Sons, New Jersey, USA
stress level. [6] Clayton, R.I., Simons, N.E., and Matthews, M.C. (1982) “Site In-
From all the above phenomena, it can be concluded or vestigation A Handbook for Engineers”, Granada Publishing,
at least postulated that for Jakarta stiff clay, at the initial London
[7] Briaud, J.L. (1992) “The Pressuremeter”, A.A. Balkema, Rotter-
stage of pressuremeter test the soil is in partially or near dam
drained cohesion, as the radial stress and strain increases [8] Clarke, B.G. (1995) “Pressuremeters in Geotechnical Design”,
and reaches its yield pressure, Py, the stiff clay is already Blackie Academic and Professional, London
in fully drained cohesion. The explanation is: at the initial [9] Gouw, Tjie-Liong (1984) “A Study on Ménard dan Oyo Pressure-
meter and Its Application in Estimating Bearing Capacity of
stage, while the radial stress tends to reduce the soil Soils”, an undergraduate thesis, Civil Engineering Division, Uni-
volume, the concurrent induced tangential strain will versitas Katolik Parahyangan, Bandung
expand the soil radially, therefore the soil is not in a fully [10] Ghionna, v., et al. (1981), “Performance of Self-boring Pressure-
compressive nature, but rather in a radial and tangential meter Tests in Cohesive Deposits”, Report FHWA/RD-
81/173/1981, MIT, Boston
ring like shearing nature. Consequently, at this stage the [11] Gibson, R.E., and Anderson, W.F. (1961) “In Situ Measurement
soil at least is in a partially drained condition. At and of Soil Properties with the Pressuremeter”, Civil Engineering and
beyond yield pressure, the induced tangential strain will Public Works Review, Vol 56, pp615-618
be large enough to cause the distance within the clay [12] Marsland, A. and Randolph, M.F. (1977) “Comparisons of the Re-
sults from Pressuremeter Tests and Large in Situ Plate Tests in
particles move to a larger distance on another to lose its London Clay”, Geotechnique, June 1977 27(2), pp.217-243
cohesion and left only with its angle of internal friction, [13] Mair, R.J. and Wood, D.M. (1987), Pressuremeter Testing Meth-
at this stage the stiff clay is already in a fully drained ods and Interpretation, Butterworths, London
condition. [14] Windle, D. and Wroth, C.P. (1977a) “The Use of Self-boring Pres-
suremeter to determine the Undrained Properties of Clays”,
Ground Engineering, Sept 1977 10(6), pp.37-46, London
7. Conclusion [15] Windle, D. and Wroth, C.P. (1977b), “In-situ Measurement of the
Properties of Stiff Clays”, Proceeding of 9th International Confer-
From the research, it can be concluded that Mecsi ence on Soil Mechanics and Foundation Engineering, Tokyo, Vol.
1, pp.347-352
model cannot be directly used to derive the c and ϕ values [16] Hughes, J.M.O, Worth, C.P., and Windler, D. (1977) “Pressure-
of Jakarta stiff clay. Its formulation of deformation meter tests in Sands”, Geotechnique, Vol 27 (4): p. 455-477
modulus need to be modified into two parts as written in [17] Fahey, M. and Randolph, M.F. (1984) “Effect of Disturbance on
Eq. (10a) and (10b), with this modified E function, cavity Parameters Derived from Self-boring Pressuremeter Tests in
Sand”, Geotechnique, March 1984, 34 (1) pp. 81-87
expansion theory can then be applied to derive the shear [18] Schnaid, F. (2009) “In Situ Testing in Geomechanics”, Taylor and
strength parameters. Pressuremeter test in Jakarta stiff Francis, London
clay initially exhibits partially drained condition and the [19] Bruzzi, D. et al (1986) “Self-boring Pressuremeter in Po River
gradually become fully drained condition when reaching Sand”, in The Pressuremeter and its Marine Applications, Ameri-
can Society for Testing and Materials, Philadelphia, Special Pub-
and beyond its yield pressure. The c and ϕ values lication STP 950, 1986
obtained from pressuremeter test are effective stress [20] Gambin, M. (1995) “Reasons for the Success of Menard Pressure-
parameters. The pressuremeter test can reveal the effect meter”, Proceedings of Fourth International Symposium on Pres-
of cementation of Jakarta stiff clay which appear in a suremeters, May 17-19, 1995, Sherbrooke, Quebec, Canada
[21] Boumedi, J.Y., Baud, J.P. and Radiquet, B. (2009) “LNG Tanks
higher value of cohesion which cannot be captured by at Damietta on Drilled Shafts Designed and Tested Using Menard
triaxial test due to the difficulty in obtaining a good PMT”, Foundation Design with Menard Pressuremeter Test,
‘really’ undisturbed Jakarta stiff clay samples by normal French Contributions to International Foundation Conggress &
thin wall tube sampler. Equipment, Expo ’09, p. 35-42; also in ASCE Geotechnical Spe-
cial Pu Bustamante, M., Gambin, M., and Gianeselli, L. (2009)
“Pile Design at failure Using the Menard Pressuremeter: an Up-
Date”, Foundation Design with Menard Pressuremeter Test,
French Contributions to International Foundation Conggress &
 Equipment, Expo ’09, p. 11-18; also in ASCE Geotechnical Spe-
cial Publication no. 186, pp.127-134blication no. 186, pp.103-110
[22] Bustamante, M., Gambin, M., and Gianeselli, L. (2009) “Pile De-
sign at failure Using the Menard Pressuremeter: an Up-Date”,
Foundation Design with Menard Pressuremeter Test, French Con-
tributions to International Foundation Conggress & Equipment,
Expo ’09, p. 11-18; also in ASCE Geotechnical Special Publica-
tion no. 186, pp.127-134
[23] Gambin, M. and Frank, R. (2009) “Direct Design Rules for Piles
using Menard Pressuremeter Test”, Foundation Design with
Menard Pressuremeter Test, French Contributions to International
Foundation Conggress & Equipment, Expo ’09, p. 3-10; also in
ASCE Geotechnical Special Publication no. 186, p.111-118
[24] Reiffsteck, P. (2009) “ISP5 Pile Prediction Revisited”, Founda-
tion Design with Menard Pressuremeter Test, French Contribu-
tions to International Foundation Conggress & Equipment, Expo
’09, p. 11-18; also in ASCE Geotechnical Special Publication no.
186, pp.50-57
[25] Schlosser, F., Guilloux, A., Zaghouani, K. and Berthelot, P.
(2009) “Rades Bridge Drilled Shafts Designed and Tested Using
Menard Pressuremeter”, Foundation Design with Menard Pres-
suremeter Test, French Contributions to International Foundation
Conggress & Equipment, Expo ’09, p. 27-34; also in ASCE Ge-
otechnical Special Publication no. 186, pp.42-49
[26] Hamidi, B., Debats, J.M., Nikraz, H., and Varaksin, S. (2013)
“Offshore Ground Improvement Records”, Australian Geome-
chanics Journal, Vo. 48 No. 4 Dec 2013, p.111-122
[27] Briaud, J.L., Pacal, A.J, dan Shively, A.W. (1984) “Power Line
Foundation Design Using the Pressuremeter”, Proceeding of First
International Conference on Case Histories in Geotechnical Engi-
neering, p.279-283, Missouri, USA
[28] Jefferies, M.G. (1988) “Determination of Horizontal Geostatic
Stress in Clay with Self Bored Pressuremeter”, Canadian Geotech-
nical Journal, Vol 25, pp. 559-573
[29] Ferreira, R.S. and Robertson, P.K. (1991), “Interpretation of Un-
drained Self-boring Pressuremeter Test Results Incorporating Un-
loading”, Canadian Geotechnical Journal, Oct. 1991
[30] Briaud, J.L. (1997) “SALLOP: Simple Approach for Lateral
Loads on Piles”, Journal of Geotechnical and Geoenvironmental
Engineering, p.958-964
[31] Bullock, P.J. (2004) Insitu Rock Modulus Apparatus, Report on
Contract No. BC-354-RPWO No. 13, submitted to Florida Depart-
ment of Transportation, 23 August 2004
[32] Silvestri, D. and Ghassan, A.S. (2012) “Analytical Solution for
Undrained Plane Strain Expansion of a Cylindrical Cavity in Mod-
ified CamClay”, Geomechanics and Engineering, Vol. 4, No. 1
(2012), p. 19-31
[33] Ramdane, B., Nassima, A. and Quarda, B. (2016) “Interpretation
of a Pressuremeter Test in Cohesive Soils”, Proceedings of Inter-
national Conference on Geotechnical Engineering, 21-23 Feb
2013, Tunisia
[34] Mecsi, J. (2013) “Geotechnical Engineering Examples and Solu-
tions Using the Cavity Expanding Theory”, Hungarian Geotech-
nical Society, Hungary
[35] Ladanyi, B. (1961) “Etude Theorique at Experimentale de l’ex-
pansion daus un Sol Pulverulent d’une cavite Presintant une
Synietrie Spherique ou Cylindrique”, Annaes de Travaux Publics
de Belgique; Buxelles No. 2 et 4
[36] Vesic, A.S. (1972, “Expansion of Cavities in an Infinite Soil
Mass”, JSMFE, ASCE, Vol. 98, SM3, Proc. Paper 8790, pp.265-
290
[37] Vesic, A.S. (1975) Principles of Pile Foundation Design, Soil Me-
chanics Series No 38, Duke University, N.C
[38] Baligh, M.M. (1976) “Cavity Expansion in Sands with Curved En-
velopes”, JSMFE, ASCE: Poc. Paper 12536, p.1131-1146
[39] Wroth, C.P. and Windle, D. (1975) “Analysis of Pressuremeter
Test Allowing for Volume Change”, Geotechnique, Vol 25, No.3,
Technical Notes, pp.598-604
[40] Datye, K.R. and Nagraju, S.S. (1977) “Reinforce Granular Col-
umns – A New Design Approach”, Proceeding of IX ICSMFE,
Tokyo, Special Session No. 10
[41] Pandit, N.S., Chaney, R.C. and Fang H.Y. (1983) Review of Cav-
ity Expansion Models in Soil and Its Applications, Fitz Engineer-
ing Laboratory, Department of Civil Engineering, Lehigh Univer-
sity, Bethlehem, Pennsylvania, USA
[42] Yu, H.S. (2000), Cavity Expansion Methods in Geomechanics,
Springer Science Business Media B.V., NewCastle, Australia