See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/278770823

Determination of elastic modulus from stress controlled cone penetration test

Article · April 2008

CITATIONS READS

0 802

5 authors, including:

Philippe Reiffsteck Claude Bacconnet

Université Gustave Eiffel Université Clermont Auvergne - Clermont Ferrand- France
130 PUBLICATIONS   383 CITATIONS    58 PUBLICATIONS   275 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Evaluation of preconsolidation stress from laboratory and in-situ strength parameters View project

Pieux en bois - Wooden piles View project

All content following this page was uploaded by Philippe Reiffsteck on 12 August 2015.

The user has requested enhancement of the downloaded file.

Determination of elastic modulus from stress controlled cone penetration
test
Ph. Reiffsteck,
Laboratoire Central des Ponts et Chaussées, Paris, France
C. Bacconnet, R. Gourvès, E. Godde,
LGC CUST, Clermont Ferrand, France
H. van de Graaf
Lankelma, Oirschot, Netherland

ABSTRACT: Measurement of the settlement of a cone submitted to incremental loading allows the determi-
nation of a compressibility modulus correlated to those obtained using laboratory tests and moduli derived
from other in situ tests. This test can be carried out with standard static penetrometer equipment. After stop-
ping the penetration of a static penetrometer cone at a desired depth, an incremental loading test is carried out
until the failure of the ground under the cone is observed. In this manner, a modulus can be obtained on the
stress strain relationship. Thus the classical CPT log is enriched by moduli offering new possibilities. The de-
velopment of the first prototype and its field testing, have permitted the development of the test procedure.
The interpretation method, validated using laboratory and field tests and geotechnical centrifuge models can
be use by practicing engineers.

1 INTRODUCTION
Qt
cone pressure
The incremental cone loading test presented in
this paper can be carried out during a normal static qc
∆h
penetration test as a complementary test. The test
principle is simple: it consists in increasing incre-
mentally the load on a standardized electrical cone
penetrometer (ISO/CEN 22476-1) until the failure of
z z
the ground under the cone is observed (figure 1). ∆h
Once the desired depth is reached (z*, qc*), a pre-
liminary phase permits to dissipate pore pressures, Figure 1. Principle of the cone loading test.
before the loading phase. One carries out loading by
stages of same intensity approximately equal to This paper presents the equipment developped,
qc*/n, qc* being the cone resistance recorded just be- results of a field test and the way derived parameters
fore the test. At each stage the load is hold constant such as elastic modulus can be obtained. In a first
for 60 seconds. Measurements of displacements are phase, the test procedure and measurement devices
made at 30 and 60 seconds. were developed by the Civil Engineering Laboratory
The loading curve gives the cone settlement as a of Blaise Pascal University in collaboration with La-
function of the pressure applied to the cone. There boratoire Central des Ponts et Chaussées (LCPC).
are no major difficulties to carry out cyclic tests. We present a new campaign on a LCPC experimen-
During the loading test, complementary parameters tal site with a new testing apparatus made up of a
like sleeve friction and the pore pressure are eventu- semi-heavy static penetrometer in order to validate
ally measured continuously. Thus friction ratio and the improvements made to the system and the pro-
increase of the pore pressure give precious informa- cedure. The moduli obtained by the cone loading test
tion on the failure. technique were confronted with the geotechnical in-
vestigation of this site.
2 DEVELOPMENT OF THE CONE LOADING 2.2 Protocol
TEST WITH A STATIC PENETROMETER After the installation of the thrust machine and
measuring instruments, preparation of the cone
The field campaigns presented in this paper hap-
penetrometer (saturation of the porous stone, zero
pened in summer 2006, on Merville experimental
reading), the cone penetrometer is pushed at a con-
site.
stant rate of penetration of 2 cm/s what makes it
2.1 Device used possible to record a traditional cone penetration log.
At the desired depth (z*, qc*), the penetration is
The cone loading tests were carried out with stopped and a dissipation test is started. One awaits
Lankelma small penetrometer based on a mobile the end of dissipation or 15 min to the maximum if
caterpillar type tracked vehicle installed with four pore pressure dissipation cannot be reached. One
drilled anchors. This installation allows to obtain a carries out then an unloading up to a value ranging
thrust capacity up to 180 kN. The point used is a between the zero of cone pressure and the estimated
piezocone (ISO/CEN 22476-1, 2006). A friction re- vertical stress. The loading sequence followed is
ducer located above the friction sleeve reduces fric- copied on the Ménard pressuremeter test (ISO EN
tion along the push rods. The cone and the dis- 22476-4, 2007; ASTM D4719-00). One then carries
placement transducer used for the measurement of out a loading by n stages of intensity qc*/n. n will be
the settlement are connected to an automatic data at least 10 for grounds soft to fairly dense up to 25
acquisition system (Figure 2). for dense to very dense grounds. Each stage is held
2 3 constant for 60 seconds. Measurements of displace-
ment, of cone pressure, pore water pressure and
sleeve friction are recorded every 5 seconds or if re-
6
corded manually at 30 and 60 seconds.

2.3 Tests made on Merville and Compiegne sites

5
The cone loading tests realized in Merville were car-
1 4 ried out in a very plastic and overconsolidated Flan-
dres clay; the site is homogeneous (figure 3).
Figure 2. 1- device allowing to visualize and record the pa- cone resistance qc in MPa
rameters during the cone loading tests and the penetration. 2 - friction ratio fs/qc×100 Ménard Pressuremeter Modulus EPMT (MPa)
pump, 3 - Connection of the hand pump on the hydraulic sys- 0 4 8 12 16 20 24 28 0 10 20 30 40 50
tem of the CPT and pressure gauge, 4 - rigid structure anchored 0 0

in the ground, independent of the thrust machine, 5 - displace-

PM1
1 PM1 bis

ment transducer, 6 - Plate attached on the push rods for the

PM2
2 CLT
2
measurement of displacements, 3
To carry out the cone loading test, a manual 4 4

pump is connected on the hydraulic jack. The 5

depth (m)
depth in m

loading of the cone is ensured by the jack which 6 6

transmits the effort to the push rods. A pressure 7

transducer connected to the pump permits to obtain 8 8

indirectly the total penetration force. A calibration 9

curve connecting the reading on the pressure 10 10

transducer to the effort to the top of the rods was 11

12
elaborate for this purpose. A displacement 12

transducer is fixed on a rigid structure anchored in Figure 3. Cone penetration tests C1 and pressuremeter tests log
the ground and independent of the thrust machine. In the clay layer where tests were carried out
This frame is used as a reference during pressuremeter module EM varies linearly according
measurement of the settlement of push rods by the to the depth between 10 and 30 MPa and the limit
displacement transducer as shown on figure 2. The pressure varies also linearly between 1 and 1,3 MPa
measuring range of the transducer used is 50 mm for (Canépa et al., 2002) (figure 3). Several cone load-
an absolute precision of 0,2%, i.e. 0,1 mm. An ing tests were carried out on this site. The cone pres-
interface digitalizes and records these displacements sure is represented versus the settlement of the top of
and pressure. the push rods (figure 4).
 2,5 Merville seem to represent the different phase of contractancy
and dilatancy as observed in undrained triaxial tests
2 C1 4 m
(figure 6).
cone resistance (Pp in MPa)

C1 7,1 m
C1 8 m We need to make a correction of settlement due
1,5 C1 9,8 m
C2 4 m
to the compression of the rods. The correction can
1
C2 7,1 m be obtained from top and base resistances which
C2 8 m
C3 6,5 m have been measured separately. Another way to es-
0,5
C3 7,1 m
C3 9,8 m
timate the rod compressibility is to calculate the av-
erage shaft friction as the sum of shaft resistance
0 measured locally during cone loading tests. In
0 5 10 15 20 25 30
cone settlement (∆h in mm)
35 40 45 evaluating the shaft resistance the stable post peak
values are used and not the peak values (figure 7).
Figure 4. Raw curves of cone loading tests performed in Mer- However this last method gives an overestimate as
ville
we have used a friction reducer during our tests.
The curve is composed of a linear part in small 80 Merville
strain, then of a curve and a linear part in higher
70
strain level. The quality of these curves is very good. C1 4 m

shaft resistance (τsf in kPa)

The figure 4 shows the repetitivity observed by three 60
C1 7,1 m
C1 8 m
loading made at 4 7.1, 8 and 9.8 m: the results are 50
C1 9,8 m
homogeneous. The residual strength also called limit 40
C2 4 m
C2 7,1 m
pressure reached is close to the mean value of con- 30 C2 8 m

ventional cone penetration resistance qc mean interpo- 20

C3 6,5 m
C3 7,1 m
lated starting from measurements immediately above C3 9,8 m
10
and under the test level (table I).
0
0,25 4m 0 5 10 15 20 25 30 35 40 45
7m cone settlement (∆h in mm)
8m
pore pressure (MPa)

0,2 Figure 7. Measured shaft resistance during cone loading test

9.8 m
0,15 2.4 Interpretation of the curves
0,1 Many moduli can be defined from these stress-strain
curves. One calls below penetrometer modulus Epn
0,05 the quantity defined by equation (1) where ∆Pp/∆h is
the slope of the linear part at small strain and he is
0 the equivalent elastic height introduced to keep Epn
1 10 100 1000
time (s) homogeneous to a pressure.
Figure 5. Dissipation of pore pressure
Before loading all excess pore pressure in the ∆Pp ∆Pp πR
E pn = × he = × × (1 − ν 2 ) (1)
clay set up during installation are almost completely ∆h ∆h 4
dissipated (figure 5).
500 Merville One makes in a first approach the assumption that
400
C1 4 m
C1 7,1 m
the probe ended by a cone can be assimilate to a
C1 8 m small rigid circular plate of radius R deep embedded
pore pressure (u en MPa)

300 C1 9,8 m
C2 4 m
at a depth H inside a semi-infinite elastic solid with-
200 C2 7,1 m out weight of characteristics E and n. The Boussi-
C2 8 m
C3 6,5 m nesq formula is used to compute the displacement of
100
C3 7,1 m
C3 9,8 m
the probe with a factor of 2 taking into account the
0 embedment in the ground, (Butterfield and Banerjee,
-100
0 5 10 15 20 25 30 35 40 45
1971).
In tables I, the cone resistance qc at the depth of
-200
cone displacement ( mm)
the cone loading test is the average value of cone
values measured 5 cm before and after the initial
Figure 6. Pore pressure variation measured during cone loading depth of the test. Bachelier and Parez (1965) have
tests performed in Merville
proposed to link the œdometric moduli to cone resis-
Figure 6 shows how excess pore pressure change
tance using the expression Eoed = α × qc with α = 7
during cone loading test. Even if the results show a
for an overconsolidated clay as Merville Flandres
small scatter, they all follow the same trend and
clay. This relation seems to be valid in the view of baoui et al. (2006). A similar degradation of elastic
oedometric moduli obtained in laboratory. modulus has been observed. It has been shown that
Table I. Modules deduced from the Merville tests (Flandres this non-linear stress behaviour of soil can be ap-
clay)
proached by the hyperbolic family of soil models.
Survey Prof. Epn ql qc Eoed EPMT
(m) (MPa) (MPa) mean (MPa) (MPa)
(MPa) 3 CONCLUSION ET PERSPECTIVES
C1 4 7.5 1.04 1.38
12.9 13.8
C2 4 12.4 1.13 1.52
C3 6.5 8.4 1.78 1.93 - 15.8 The results showed that the cone loading test al-
C1 7.1 12.9 1.72 2.25 lows to obtain exploitable curves for the determina-
C2 7.1 14.7 2.13 2.51 - 21 tion of elasticity moduli of the ground, this accord-
C3 7.1 15.8 1.74 2.04 ing to the deformation, even for very reduced
C1 8 9.1 1.55 1.94 deformations. An advantage is the integration of the
17.1 18.7
C2 8 9.3 1.99 2.68
C1 9.8 8.9 1.64 2.1 test of loading in the normal cone penetration test. In
26.1 19.7 addition, the boundary conditions are well controlled
C3 9.8 9.45 1.77 2.36
and the insertion is more repetitive than for borehole
The data show also a reasonable agreement with expansion tests realised in prebored hole. The con-
the correlation Eoed = β.EPMT proposed by Cassan frontation of cone loading tests results obtained on
(1978) between pressuremeter moduli and œdomet- Merville site with pressuremeter results, and oe-
ric moduli: Eoed = 1.55xEM. dometer and triaxial compression tests have shown
It is important to notice that each test has an ap- the pertinence of the technique. The simple and im-
plication range depending of strains generate in the mediate relations, which can be drawn from results
ground, deformations which must be related with obtained on sites, are promising.
those appearing during construction of geotechnical Then, it is planned to more develop the method of
works or during their lifetime (Burland, 1989). On interpretation of the test in order to lead to the esti-
figures 9, we represented the evolution of secant mate of settlements.
shear modulus versus strain obtained with cone load-
ing tests at Merville site and the other in situ tests as
4 REFERENCES
proposed by Atkinson et al., (1991), Tani (1995) and
Tatsuoka et al. (1997). ASTM D4719 (2000). Standard Test Method for Prebored
60000 Pressuremeter
TRIAXIAL
BE
Arbaoui H., Gourvès R., Bressolette Ph., Bodé L. (2006), Me-
DH sure de la déformabilité des sols in situ à l'aide d'un essai de
50000
CH chargement statique d'une pointe pénétrométrique, Can.
Geotech. J. 43(4) 355-369
secant shear modulus (kPa)

SASW
SC
40000
4m PMT
Bachelier M., Parez L., (1965) Contribution à l’étude de la
CPM compressibilité des sols à l’aide du pénétromètre à cône, 6th
STAF ICSMFE, Montréal, Vol. 2, pp. 3-7
30000 LCPC SBP
CLT
Butterfiels R. Banerjee P.K. (1971) A rigid disc embedded in
an elastic half space, Geotechnical Engineering, 2, 35-52
20000 Burland J.B. (1989) Ninth Laurits Bjerrum memorial Lecture :
« Small is beautiful » the stiffness of soils at small strains,
10000
Can. Geotech. J., 26, 499-516
Burland J.B., Twine D. (1988) The shaft friction of bored piles
in terms of effective strength, Deep Foundations on Bored
0 and Auger Piles, Van Impe Ed; Balkema, 411-420
0,0001 0,001 0,01 0,1 1
Canépa Y., Borel S., Deconinck J. (2002). Détermination de la
deformation (-)
courbe d’évolution du module de cisaillement d’un sol en
fonction de sa déformation à partir d’essais en place. PA-
Figure 9. Variation of secant shear modulus obtained from RAM 2002, presses de l’ENPC, pp. 25-32.
cone loading test (CLT) compare to several field techniques at Cassan M. (1978) Les essais in situ en mécanique des sols,
4 m on Merville site tome 2 applications et méthodes de calcul, Ed. Eyrolles,
Figures 9 shows clearly the ability of the cone 331 pages
loading test to represent the non linear stress strain ISO/CEN 22476-1 (2006) Geotechnical investigation and test-
behaviour from the early stage of loading and to the ing - Field testing - Part 1: Electrical cone and piezocone
penetration tests (CPT et CPTU), 35 pages
range of deformation in which the definition of the ISO/CEN 22476-4 (2007) Geotechnical investigation and test-
modulus is pertinent for design of geotechnical ing - Field testing - Part 4: Ménard pressuremeter test, 66
structures. pages
Tests results for cone loading tests made in sand Van Baars S. Van De Graaf H. (2005) L’utilisation du piezo-
cone pour la détermination de la perméabilité des sols com-
tank at small strain level have been compiled by Ar- pressibles, Géoline 2005,

View publication stats