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

net/publication/282133166

Laboratory investigation on the effect of grain size distribution of granular

material on cone penetration test results

Conference Paper · September 2015

CITATIONS READS

0 118

4 authors:

Michael Ghali Mourad Karray

Université de Sherbrooke Université de Sherbrooke
7 PUBLICATIONS   2 CITATIONS    167 PUBLICATIONS   329 CITATIONS   

SEE PROFILE SEE PROFILE

Mohamed Chekired Varvara Roubtsova

Hydro-Québec Hydro-Québec
75 PUBLICATIONS   136 CITATIONS    22 PUBLICATIONS   76 CITATIONS   

SEE PROFILE SEE PROFILE

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

THE PIEZO-ELECRIC RING ACTUATOR TECHNIQUE (P-RAT) View project

Assessing available effective stress frameworks in partially saturated granular media, experimental and numerical approaches View project

All content following this page was uploaded by Mourad Karray on 24 September 2015.

The user has requested enhancement of the downloaded file.

Laboratory investigation on the effect of grain
size distribution of granular material on cone
penetration test results
Michael Ghali & Mourad Karray
Sherbrooke university, Sherbrooke, Quebec, Canada
Mohamed Chekirad & Varvara Roubtsova
Hydro-Québec, Montréal, Québec, Canada

ABSTRACT
This paper initially compares and discusses a majority of correlations of (q c ) and (𝑓 s ) with typical geotechnical
parameters such as relative density (D r ), the effective overburden pressure (σ' v ), void ratio (e), mean grain size (D 50 ),
the degree of uniformity (C u ), lateral stress (σ' h ), mean confinement pressure (σ' m ), overconsolidation (OCR), and
angularity. The paper then presents the results of a series of experimental simulations for cone penetration test (CPT) on
both polydisperse spherical glass beads and natural sands samples with different mechanical properties as well as
different loading conditions. Stress-strain records along the soil samples are monitored during the tests. Results obtained
for (q c ) and (𝑓 s ) are correlated with such geotechnical parameters and the corresponding graphs are plotted in order to
predict the real behavior of natural soils in the field.

RÉSUMÉ
Ce papier compare et discute d'abord de la majorité des corrélations de (q c ) et (𝑓 s ) avec les paramètres géotechniques
typiques tels que la densité relative (D r ), la contrainte effective (σ' v ), l’indice des vides (e), la taille moyenne des
particules (D 50 ), le degré d'uniformité (C u ), la contrainte latéral (σ' h ), la contrainte moyenne de confinement (σ' m ), le
degré de surconsolidation (OCR), l’âge et l’angularité. L’article présente ensuite les résultats d'une série de simulations
expérimentales pour l’essai de pénétration du cône (CPT) sur des billes de verre sphérique polydisperses et sur des
échantillons de sables naturels avec des propriétés mécaniques différentes, ainsi que des conditions de chargement
différentes. Les contrainte-déformation le long de l’échantillon de sol sont surveillées et enregistrées pendant les essais.
Les résultats obtenus pour (q c ) et (𝑓 s ) sont corrélés avec les paramètres géotechniques et les graphiques
correspondants sont tracés afin de prédire le comportement réel des sols naturels in-situ.

1 INTRODUCTION
2 BACKGROUND INFORMATION
Investigating the effect of the soil mechanical parameters
on the cone penetration test results (q c ) and (𝑓 s ) is a Since the penetration resistances (q c ) and (𝑓 s ) basically
unique challenge for geotechnical engineers. Although the depends on the relative density and the stresses (e.g.
engineering experience shows that wide variations in (q c ) Salgado et al. (1997), and Tran, C. (2005)), general
and (𝑓 s ) values can occur as between different sands due relations have been presented as follow:
to the variation of the nature and the mechanical
characteristics of the sands such as; relative density (D r ), qc =ƒ ( Dr , σ 'v , σ 'h ) [1]
Void ratio (e), mean particle size (D 50 ), coefficient of
uniformity (C u ), overburden pressures (σ' v ), and mean
 f 
confinement pressure (σ' m ), stress history and aging. R f =  s  [%] [2]
This paper presents data of a series of experimental  qc 
simulations for cone penetration test (CPT) on both
polydisperse spherical glass beads and natural sands
where:
samples with different mechanical properties as well as
(q c ) is the cone-tip resistance, (𝑓) is a function of D r , σ' v ,
different loading conditions. The purpose of using glass
beads is the ability to start a parametric study on D 50 (e.g. and σ' h .
Ghali et al., 2014). In fact, parametric study for the above (R 𝑓 ) is the friction ratio, (𝑓 s ) is the sleeve friction.
mentioned factors is a very difficult task as most of these
factors not only affect (q c ) and (𝑓 s ) values but also affect Also some previous studies have been briefly
each other by somehow. Glass beads allow the presented as follows; Schmertmann's (1978) stated some
elimination of the effects of angularity and stress history correlations which is more applicable to sands of high
compressibility, while the correlation by Baldi et al. (1982,
under certain D r , σ' v , and coefficient of earth pressure at
1985, & 1986) was developed for sands of medium
rest (K 0 ). Also using glass beads allow the verification of
compressibility. Similar correlations have also been
the results obtained from a DEM software developed by
proposed by Robertson and Campanella (1983). Kulhawy
Hydro-Quebec to simulate the CPT. Procedures and
and Mayne, (1991) collected some field data of CPT and
driving method were fixed in all carried out tests.
relative density D r for 24 different sands and established where q c1 is the tip resistance normalized for the
the following general relation to correlate the CPT tip overburden stress (100 kPa), and C Q can be estimated as
resistance q c with the relative density: follow:

 1   qc / Pa 
0.5
 P 
Dr 2 =   ' 0.5 
[3] =CQ  a  ≤2 [5]
 QF   (σ v / Pa )   σ 'v 0 

where: Hence;
P a represents atmospheric pressure ≈ 100 kPa; Q F is an
empirical constant determined by least-square regression  qc   qc1 
analyses for normally consolidated sands of low, medium =
qc1N =  CQ   [6]
and high compressibility to be 332,305, and 278  Pa 2   Pa 2 
respectively.
while Lee, M.J. (2011) developed another relation where P a2 is also equal to 100 kPa but it must be
illustrating the trend between q c and σ' v based on some substituted in q c units.
high accurate field and laboratory simulations for the CPT
as shown in Figure 1. Also Lee, M.J. (2011) included the Schmertmann's (1978) figured a relation between the
effect of D r as illustrated in Figure 2 to q c and σ' v . tip resistance q c and σ' v at various D r values as shown in
Figure 3 and then some correlations with Figure 4 were
carried out to take into account the effect off over
consolidation as shown in equations 7, and 8

K' 
qcOC / qcNC =
1 + 3 / 4  oOC − 1 [7]
 K 'oNC 

K 'oOC
= ( OCR )
0.42
[8]
K 'oNC

where:
q cOC and q cNC represent the tip resistances for over
consolidated and normally consolidated sands
Figure 1. q c vs σ' v relationship. (after Lee (2011)) respectively. Also K' oOC and K' oNC represent the
coefficients of earth pressures at rest for over
consolidated and normally consolidated sands
respectively.

On the other hand Salgado et al., 1997. presented

some correlations for the effects of σ' v and σ' h on the
normalized CPT tip resistance in comparison with some
earlier studies as follow:

qc1k 0, NC = CNV CNH qc [9]

where:
q c1k0,NC and q c are the normalized and recorded tip
resistances respectively.
C NV was suggested by Liao and Whitman (1986), as
Figure 2. q c , D r , and σ' v relationship. (after Lee (2011)) follows:
Also it is worthy to mention that Robertson and Wride
Pa
( 1998 and 2000) and Wride et al . (2000) stated some CNV = [10]
important correlations for the truly normalized cone σ 'v
penetration resistance for overburden stress (q c1N ) which
is dimensionless; as follows: Salgado et al., 1997 verified the previous mentioned
relation at D r = 0.50 and generally presented C NV - σ'v
qc1 = qc CQ [4] relations at several D r values as in Figures 5 and 6. Also
C NH was stated in accordance with allot of researches
e.g. Baldi et al. (1982, 1985) as in Equation 11, and
generalized by Salgado et al., 1997 in Figure 6.
 K 0, NC
CNH = [11]
K0

where:
K 0,NC is the coefficient of earth pressure equivalent to
q c1k0,NC and K 0 is the coefficient of earth pressure for
normally consolidated soils.

Figure 5. C NV vs σ' v relationship obtained using

CONPOINT with expression of Liao and Whitman, 1985
(after Salgado et al. (1997)).

Figure 6. C NV vs σ' v relationship obtained using

Figure 3. q c - σ' v relationship for various D r values. (after
CONPOINT with values recommended by Seed and
Schmertmann's (1978))
Idriss, 1971 (after Salgado et al. (1997)).

Figure 4. Normalized tip resistance to normalized K for Figure 6. C NH vs K 0 / K 0,NC relationship (after Salgado et
some field and champer tests (after Schmertmann's (1978)). al. (1997)).
3 EXPERIMENTAL WORK Table 2b. Minimum void ratio for glass beads (After Ghali
et al., 2014).
An objective was set, from a series of experimental D 50 (mm)
simulations for the CPT on natural disturbed sand Sample No.
0.50 1.00 2.00 3.00 4.00
samples as well as artificial (glass beads) representing
sand sizes, to justify and develop new equations 2.5 0.45 0.42 0.40 0.39 0.39
governing the relations between (q c ) and (𝑓 s ) values and Cu 4 0.38 0.35 0.33 0.31 0.30
the previously discussed geotechnical parameters. 8 0.32 0.30 0.29 0.27 0.27
12.5 0.32 0.30 0.28 0.26 0.26
3.1 Testing Assembly and Samples:

An assembly developed earlier in the geotechnical

laboratory at Sherbrooke University ( Ghali, et al., 2014)
has been used to simulate Laboratory CPTs.This
assembly as shown in Figure 7 consists of loading arm,
guide frame, fully confined cell with several inner semi-
flexible walls, lateral and vertical pressure. Simply as
illustrated in the above mentioned figures, the loading arm
works as an oedometer with arm ratio of 5.45 and own
weight reaction of 1.875 kN on the top surface of the
samples. Vertical loads applied at the end of the loading
arm by a calibrated hydraulic compressor and the applied
loads were monitored instantly during all tests. the total
vertical stresses were calculated and verified using
pressure sensor was put at the bottom of the cell. Several
artificial and natural soil samples were prepared in the cell Figure 7. The assembly used to simulate the CPT.
(340 mm inner diameter and 500 mm height) shown in
Figure 8.

3.2 Testing Program and Results:

Several tests were carried out on glass beads samples

and some corresponding natural clean sand soil samples
(previously prepared and tested by Ghali et al, 2014) as
illustrated in Table 1 and Figure 9, under certain given
initial relative densities, specific gravity, dry densities,
initial volume, and known volumetric changes during
testing. The maximun void ratios (e max ) and the minimum
void ratios (e min ) were also determined for all tested
samples as presented in Tables 2a and 2b, respectively.
Figure 10 shows a sample of the angularity of a used Figure 8. Testing cell equipped by several horizontal and
sand particles. vertical pressure sensors.
Table 1. Samples coding numbers (After Ghali et al.,
2014).
D 50 (mm)
Sample No.
0.50 1.00 2.00 3.00 4.00
2.5 A-1 A-2 A-3 A-4 A-5
Cu 4 B-1 B-2 B-3 B-4 B-5
8 C-1 C-2 C-3 C-4 C-5
12.5 D-1 D-2 D-3 D-4 D-5

Table 2a. Maximum void ratio for glass beads(After Ghali

et al., 2014).
D 50 (mm)
Sample No.
0.50 1.00 2.00 3.00 4.00
2.5 0.61 0.56 0.53 0.51 0.50
Cu 4 0.51 0.47 0.44 0.42 0.40
8 0.43 0.39 0.37 0.36 0.35 Figure 9. Sample for the tested glass beads gradation: C u
12.5 0.42 0.39 0.36 0.33 0.33 = 4 (After Ghali et al., 2014).
 Figure 11b. Sample schematic curve for the relation
between (𝑓 s ) and the effective overburden stress.

Figure 10. Sample for clean tested sand of average

angularity (A 2D ) = 650.

The cone tip resistances (q c ) and the sleeve friction

resistances (𝑓 s ) were recorded among all tests under
several applied effective overburden pressures as well as
different lateral pressures, sample of the results are
illustrated in Figures 11a-11d.
Also several tests were carried out under effective
overburden pressure of 98 kPa and different coefficient of
lateral pressures, and tri-axial tests were performed for all
the tested samples in order to obtain the internal friction
(ϕ'). Equation 12 is used to calculate the coefficient of
lateral earth pressure at rest (K 0NC ), hence the normalized
(q c1 ) and (𝑓 s1 ) at (σ' v = 98 kPa) can be obtained for each
individual test as shown in Figure 12.

K 0 NC = 1 − sin φ ' [12] Figure 11c. Sample schematic curve for the variation of
(q c /K 0 ) with depth.

Figure 11a. Sample schematic curve for the relation Figure 11d. Sample schematic curve for the variation of
between (q c ) and the effective overburden stress. (𝑓 s /K 0 ) with depth.
 While the effect of the void ratio range on the
normalized tip resistance and the normalized sleeve
friction for the tested clean sands as well as spherical
glass beads were presented in Figures 14a, & 14b.

Figure 12. Example for predicting (q c1 ) and (𝑓 s1 ) at (σ' v =

98 kPa) for each individual test.

4 ANALYSIS OF RESULTS AND PROPOSED

RELATIONS 2
Figure 14a. Proposed correlation between (q c1 / D r ) and
The relation between (D 50 , A 2D and C u ) and the void ratio void ratio range.
range were presented earlier by Ghali et al., 2014 in
Equation 13 as a best fitting for the relations Figure 13.

0.06* D50 (0.046log Cu −0.17) [13]

emax − emin = − A2 D
e 1000 *log( D50 + 1) * Cu 0.285

2
Figure 14b. Proposed correlation between (𝑓 s1 / D r ) and
void ratio range.

On the other hand several tests were carried out

under σ'v = 98 kPa and several lateral pressures, in order
to predict the effect of mean confinement pressure on the
normalized q c and 𝑓 s values as shown in:

x1
qc  1 + 2 K 
= 
[14]
qc 0  1 + 2 K 0 

and

x2
fs  1 + 2K 
= 
[15]
fs0  1 + 2K0 

where q c0 and 𝑓 s0 are the tip resistance and sleeve

Figure 13. Sample of the present database for the effect friction at K=K 0 respectively, and x1 & x2 are constants.
of D 50 , C u and A 2D on (e max - e min ) (after Ghali et al.,
2014).
New proposed relation was preliminarily developed to D 50 , C u , ϭ' m , OCR and angularity). The key finding of this
predict the effect of each individual parameter included in study can be summarized as follows:
the present study on the CPT tip resistance values as a) (q c and 𝑓 s ) values were verified to be
follows: proportional with the square of the relative density.
b) (q c and 𝑓 s ) values were verified to be
where ϭ'v in kPa.
0.50
qc − f ( q0 ) =
f q ( D ,U , A) . f q ( N ,OCR ) .Dr 2 [16] proportional with (ϭ'v/98)
c) (q c ) values were found to be related with the
OCR as well as the mean confinement pressure by the
where x1
relation ((1+2k)/(1+2k 0 )) .
f ( q0 ) is a function representing the CPT tip resistance d) (𝑓 s ) values were found to be related with the
OCR as well as the mean confinement pressure by the
values for the soil at loosest state (D r = 0.0), which is very x2
relation ((1+2k)/(1+2k 0 )) .
small value ≈ 0.0.
e) The effect of angularity on (q c and 𝑓 s ) values
and
decreases with the increase of D 50 .
f q ( D ,U , A) is a function of D 50 , A 2D and C u, developed
under σ'v = 98 kPa and K = K 0 as follows:
REFERENCES
f q ( N ,OCR ) is a function of ϭ'v and OCR. as follows: Baldi, G., Bellotti, R., Ghionna, V" Jamiolkowski, M., and
Pasqualini, E. (1982). "Design parameters for sands
x1
 1 + 2K   σ ' v 
0.5 from CPT." Proc., 2nd Eur. Symp. on Penetration
f q ( N ,OCR ) =   * 
[17] Testing, A. A. Balkema, Rotterdam, The Netherlands.
 1 + 2 K 0   98  Vol. 2, 425-432.
Baldi, et a!. (I985). "Laboratory validation of in-situ tests."
On the other hand the preliminary proposed relations for Italian Geo­ tech. Assoc. (AGI) Jubilee Vol., Proc.,
the sleeve friction can be summarized as follow: 11th Int. Conf. on Soil Mech. and Found. Engrg., A.
A. Balkema, Rotterdam, The Netherlands.
Baldi, G., Bellotti, R., Ghionna, V" Jamiolkowski, M., and
f s − f ( f0 ) =
f f ( D ,U , A) . f f ( N ,OCR ) .Dr 2 [18]
Pasqualini, E. (1986). "Interpretation of CPT's and
CPTU's. 2nd Part: drained penetration on sands."
where Proc., 4th Int. Geotech. Seminar: Field Instrumentation
f ( f0 ) is a function representing the CPT sleeve friction and In Situ Measurements, Nanyang Technol. Univ"
Singapore. 143-162.
values for the soil at loosest state (D r = 0.0), which is very Ghali M., Hussien M. N., Karray M., Chekirad M., &
small value ≈ 0.0. Roubtsova V. 2014. "Laboratory investigation on the
and effect of grain size distribution of granular material on
f f ( D ,U , A) is a function of D 50 , A 2D and C u, developed penetration test results". 67th Canadian Geotechnical
Society (CGS) conference, , Regina; October 2014,
under σ'v = 98 kPa and K = K 0 as follows:
Paper NO. 306.
Kulhawy, F. H., and Mayne, P. W., 1990. Manual on
f f ( N ,OCR ) is a function of ϭ'v and OCR. as follows: Estimating Soil Properties for Foundation Design,
Final Report 1493-6, EL-6800, Electric Power
x2 Research Institute, Palo Alto, CA.
 1 + 2K   σ ' v 
0.5

f f ( N ,OCR ) =  * 
[19] Kulhawy, F.H. and Mayne, P.W., 1991. " Relative density,
 1 + 2 K 0   98  SPT and CPT interrelationships", in: Proc. Ist Int.
Symp. on Calibration Chamber Testing, New York, pp.
197–211.
5 RELIABILITY OF THE PRESENT CORRELATION Lee, M.J. et al., 2011. Effect of stress history on CPT and
DMT results in sand. ELSEVIER, Engineering
Results were found to be matching with existing field Geology, vol 117, pp 259-265.
tests with very small deviation which may be due to: Liao, S.C., and Whitman, R.V. 1986. Overburden
a) The presence of fine grained soil content and correction factors for SPT in sand. Journal of
water content which are not included in this study. Geotechnical Engineering Division, ASCE, 112(3): 373
b) Neglecting the effect of surface roughness of - 377.
coarse grained particles. Robertson, P. K., and Campanella. R. G. 1983.
"Interpretation of cone penetration tests. Part I: Sand."
6 CONCLUSIONS Can. Geotech. 1., Ottawa, Canada. 20. 718-733.
Robertson, P. K., and Campanella. R. G., and Wightman
Based on the preliminary data collected and analyzed A., 1983. "SPT-CPT correlations". ASCE Journal of
from the present experimental simulations, new Geotechnical and Geonvironmental Engineering,
correlations were developed to give reliable relations 109(11): 1449-1459.
between (q c and 𝑓 s ) values and (D r , ϭ'v, void ratio range,
 Robertson, P.K., Campanella, R.G., Gillespie, D. and
Greig, J. 1986." Use of piezometer cone data".
Proceedings of the ASCE Specialty Conference In
Situ '86: Use of In Situ Tests in Geotechnical
Engineering, Blacksburg, pp.1263-80.
Robertson, P.K., and Wride, C.E. (Fear) 1998. "Evaluating
cyclic liquefaction potential using the cone penetration
test". Canadian Geotechnical Journal, 35(3): 442–459.
Robertson, P.K., Wride, C.E. (Fear), List, B.R., Atukorala,
U., Biggar, K.W., Byrne, P.M., et al. 2000." The
CANLEX project: summary and conclusions".
Canadian Geotechnical Journal, 37(3): 563–591.
Salgado, R., Boulanger, R.W., and Mitchell, J.K. 1997.
Lateral stress effects on CPT liquefaction resistance
correlations. ASCE Journal of Geotechnical and
Geonvironmental Engineering, 123(8): 726 - 735.
Salgado, R., Mitchell, J.K., and Jamiolkowski, M. 1997.
"Cavity expansion and penetration resistance in
sands". ASCE Journal of Geotechnical and
Geonvironmental Engineering, 123(4): 344 - 354.
Seed, H.B., and Idris, I.M., (1971). "Simplified procedures
for evaluating soil liquefaction potential". J. of Mech.
and Found. Dev., ASCE, 97(9), 1249-1273.
Schmcrtmann, J. H. (1978). "Guidelines for cone
penetration test, performance and design." Rep.
FHwA-TS-78-209, Fed. Hwy. Admin. (FHwA),
Washington. D.C.
Tran, C. 2005. "Truth: non-additive measures for the
determination of relative density of sands using CPT
measurements". ELSEVIER C. R. Mecanique, 333:
605-611
Wride, C.E. (Fear), Robertson, P.K., Biggar, K.W.,
Campanella, R.G., Hofmann, B.A., Hughes, J.M.O.,
Kupper, A., and Woeller, D.J. 2000. "Interpretation of
in situ test results from the CANLEX sites". Canadian
Geotechnical Journal, 37(3): 505–529.

View publication stats