Soils and Foundations 2016;56(3):460–472

HOSTED BY The Japanese Geotechnical Society

Soils and Foundations

www.sciencedirect.com
journal homepage: www.elsevier.com/locate/sandf

Technical Paper

Evaluation of CPT-based characterization methods for loose

to medium-dense sands
Abouzar Sadrekarimi
Assistant Professor, Department of Civil and Environmental Engineering, Western University, London, Ontario, Canada

Received 27 April 2015; received in revised form 28 January 2016; accepted 27 February 2016
Available online 20 May 2016

Abstract

As a result of the difficulties related to obtaining undisturbed samples of cohesionless soils, CPT-based empirical correlations, often developed
from calibration chamber experiments, are widely used for determining many soil parameters for geotechnical investigation. This paper describes
the application of 19 reduced-scale calibration chamber cone penetration tests to evaluate empirical correlations for predicting the relative density,
the unit weight, the constrained modulus, and the soil identification of loose to medium-dense sands. A subtraction cone, 6 mm in diameter with
an apex angle of 601 and a net area ratio of 0.75, is used in the laboratory tests. Due to the fine gradation of the quartz sand used in the
experiments, some of the CPT results are located within the silty sand range of the soil identification charts. An extensive evaluation is also
presented for the stress normalization process of the CPT data. It is determined that a relative density-based overburden stress normalization
method provided the best estimates for correcting the cone tip resistance for effective overburden stress.
& 2016 The Japanese Geotechnical Society. Production and hosting by Elsevier B.V. All rights reserved.

Keywords: Cone penetration test; Calibration chamber test; Sand; Relative density; Modulus; Critical state; Stress normalization; Soil identification

1. Introduction or cyclic liquefaction and their potential for liquefaction flow

failure. As CPT does not directly measure any particular soil
Geotechnical engineering analyses and designs require accurate properties, extensive research has been conducted to develop
identification and characterization of soil layers, as well as an empirical correlations of CPT measurements with soil type and
assessment of soil stratigraphy at the site. However, due to the engineering properties (including unit weight, relative density,
difficulties in obtaining undisturbed samples of cohesionless soils, and modulus) using laboratory calibration chamber experiments
geotechnical engineers often rely on field tests to obtain the in- (Schmertmann, 1978; Villet and Mitchell, 1981; Baldi et al.,
situ soil characteristics. Owing to relatively lower costs, simpli- 1986; Jamiolkowski et al., 1988, 2001; Huang and Hsu, 2005).
city, continuous measurement with depth, and excellent repeat- These experiments can provide the most reliable and precise
ability and accuracy, the electronic cone penetration test (CPT) experimental data for developing CPT-based correlations and
has emerged as one of the most popular tools for ground calibration as the entire procedure (including sample preparation,
investigation in geotechnical engineering. CPTs are particularly consolidation, and cone penetration) is conducted in the labora-
instrumental for characterizing saturated loose to medium-dense tory and can be readily monitored and controlled.
cohesionless soils due to the susceptibility of these soils to static Carrying out controlled CPT calibration chamber tests with a
standard cone (with a diameter of 35.7 mm) requires a large-
E-mail address: [email protected] diameter (typically more than 1.2 m) chamber. Such an experiment
Peer review under responsibility of The Japanese Geotechnical Society. can be expensive and time-consuming, as sample preparation

http://dx.doi.org/10.1016/j.sandf.2016.04.012
0038-0806/& 2016 The Japanese Geotechnical Society. Production and hosting by Elsevier B.V. All rights reserved.
 A. Sadrekarimi / Soils and Foundations 56 (2016) 460–472 461

involves placing a large volume of sand in the testing chamber at a are employed to evaluate a number of existing empirical correla-
controlled density. The control of sample uniformity and external tions for determining some of the geotechnical properties of loose
stress can also become difficult (Parkin and Lunne, 1982). Due to to medium-dense sands.
these challenges, several studies have employed miniature cones
and reduced-scale calibration chamber devices (Abedin, 1995; 2. Experimental procedure
Huang and Hsu, 2005; Kumar and Raju, 2009; Kokusho et al.,
2012; Franzen, 2006; Pournaghiazar et al., 2011) frequently on 2.1. Miniature CPT system
dense sands, with little experimental data on medium-dense to
loose sands. This is often because of the collapsible fabric of loose The largest cell that could be accommodated in an existing
sands which results in initially loose calibration chamber samples uniaxial loading frame is manufactured as a CPT calibration
collapsing into a denser state during sample saturation and flushing chamber in this study. The large custom-made cell contains
(Been et al., 1987b). This study presents the results of miniature cylindrical specimens of 150 mm in diameter and 195 mm
CPT calibration chamber experiments (Damavandi and high. The major components of the CPT calibration chamber
Sadrekarimi, 2015) carried out at Western University. The results are schematically illustrated in Fig. 1. A stainless steel cone,

Fig. 1. Illustration of calibration chamber testing components.

462 A. Sadrekarimi / Soils and Foundations 56 (2016) 460–472

with a diameter of 6 mm, an apex angle of 601, and a net area

ratio of 0.75 (i.e., ratio of the cone base area unaffected by
pore water pressure to total projected cone area), is employed
in the calibration chamber tests. The miniature cone, developed
at Western University, is a subtraction-type penetrometer in
which both the cone tip resistance (qc) and the sleeve friction
(fs) are measured. Similar to a type 2 cone (Lunne et al., 1997),
sample pore water pressure, including any excess pore pressure
developing during cone penetration (u2), is measured behind
the cone's shoulder above its tip by a pressure transducer with
a capacity of 1380 kPa. A thin plastic collar with small notches
is used above the cone tip to allow the passage of water while
blocking the intrusion of sand particles. A servo-controlled
triaxial loading frame, with a uniaxial loading capacity of
10 kN, is used to drive the cone. The loading frame is Fig. 2. Average particle size distribution of Ottawa sand used in this study.
equipped with a digital encoder which records its travel
distance, and therefore, the amount of cone penetration. The
encoder was calibrated with a high resolution LVDT. Two produces minimum particle segregation (Yang, 2005; Chen,
electromechanical pressure pumps with capacities of 75 and 2000). Moist tamping often replicates the soil fabric developed
170 mL are used to measure and control the fluid volume by hydraulically transported sand fills (Sladen et al., 1985) or in
changes of the soil sample and the cell. Each pressure pump is nature as loose sand layers (Schlosser, 1985). In order to improve
equipped with electrical pressure transducers, with a capacity specimen uniformity, the under-compaction moist tamping
of 2070 kPa, to measure and control the specimen and the cell method, suggested by Ladd (1978), was employed. This method
fluid pressures. A data logger with a maximum sampling involves the compaction of moist sand in layers slightly different
frequency of 1 Hz, included with the triaxial loading frame, is than the target global density, with the bottom layer compacted
used for data acquisition and for the real-time monitoring of the least and the top layer compacted the most, so that the final
the experiments. Damavandi and Sadrekarimi (2015) provide density of each layer would be equal to the target global density
further details of the miniature CPT equipment, data analysis, even with the effects of compacting the successive overlaying
verification, and the repeatability of the miniature CPT results. layers. Sand with a moisture content of 5%, required for each
layer, was weighed, poured into the mold, and tamped into 13
2.2. Tested Material layers of 15 mm thick in a circular pattern using a scaled tamper.
This moisture developed a capillary force among the sand
Reconstituted specimens of fine Ottawa sand were prepared particles that produced a stable sand fabric and kept the specimen
and tested in this experimental program. The Ottawa sand used shape once the split mold was removed. This was necessary
in this study was a uniformly graded clean sand (with no fines) during specimen preparation before filling the cell and applying a
– classified as SP according to the Unified Soil Classification seating confining pressure of 10 kPa. A latex membrane enclosed
System (Astm, 2011) – composed of white-colored quartz and sealed the specimen, thereby creating a flexible boundary via
particles with round to sub-rounded particle shapes. Sand which a uniform axisymmetric confining pressure could be
particles had a specific gravity (Gs) of 2.65, a mean diameter applied. The diameter and the height of the specimen were
(D50) of 0.193 mm, and coefficients of uniformity (CU) and carefully measured after preparation to ensure the accuracy of the
curvature (CC) of 1.71 and 1.07, respectively. Fig. 2 presents initial relative density (Dri) determination of the sample. The
the particle size distribution of this sand. Maximum (emax) and miniature cone was then mounted on the external rod and the cell
minimum (emin) void ratios, 0.821 and 0.490, respectively, was subsequently assembled on the specimen.
were determined following the ASTM standard procedures Specimen uniformity was evaluated for a number of speci-
(Astm, 2006a; Astm, 2006b). Loose specimens of this sand mens by taking plug samples from the top, middle, and bottom
exhibited non-linear isotropic compressive behavior in triaxial of each specimen prepared by following the same procedure.
compression tests with compression indices (Cc) of 0.032– Void ratio variations of about 70.006 (corresponding to a
0.050 at effective confining pressures of 7–700 kPa, while a change in relative density of 71.5%) and 70.004 (corre-
more-or-less constant recompression index (Cr) of 0.005 was sponding to a change in relative density of 71.0%) were
determined. A critical state friction angle ϕ0 cs of 30.71 was also obtained from the top to the bottom of the specimen for sample
measured for this sand in triaxial compression tests (Omar and relative densities of 10% and 26.8%, respectively. Despite
Sadrekarimi, 2015). improved specimen uniformity, small fluctuations were still
observed in some of the experiments (e.g., test No. 16 in Fig. 3)
2.3. Specimen preparation reflecting the remaining specimen non-uniformity.
A particular limitation of reduced-scale cones can result
For the calibration chamber CPT experiments, cylindrical soil from the scale of the cone diameter (dc) compared to the
samples were prepared using the moist tamping method, which diameter of the soil particles (e.g., D50). Several studies have
 A. Sadrekarimi / Soils and Foundations 56 (2016) 460–472 463

consolidation pressure (p0 c) using two electromechanical pres-

sure pumps. The change in sample volume during consolida-
tion was measured by pressure pumps to determine the precise
void ratio of the sample after consolidation (ec) as well as the
consolidation relative density (Drc).

2.5. Cone penetration

Following consolidation, a uniaxial loading frame (origin-

ally designed for triaxial testing) was used to push the
miniature cone into the sample up to a penetration depth of
60 mm, while qc and fs were simultaneously measured. During
cone penetration, the sample was kept globally drained through
the top and bottom drainage ports. Table 1 summarizes the
Fig. 3. Cone penetration resistance mobilized in Tests No. 9, 11, 13, and 16: characteristics of the miniature cone penetration tests con-
(a) qc, (b) fs.. ducted in this study.
The chamber boundary effect can be a major limitation in
investigated the scale effect on cone penetration tests the comparison of CPT calibration chamber results with in-situ
(Schmertmann, 1978; Canou et al., 1988; Jacobs and Coutts, field tests. However, several studies (Parkin and Lunne, 1982,
1992; Salgado, 2013; Gui et al., 1998; Baldi and O'neill, 1995; Baldi et al., 1982, Jamiolkowski et al., 1985, Been et al.,
Huntsman, 1985; Huang and Hsu, 2005; Lee, 1990; 1987a, Mayne and Kulhawy, 1991, Salgado, 1993) have found
Balachowski, 2007; Sharp et al., 2010). For example, Canou this effect to be insignificant for loose to medium-dense sands.
et al. (1988) carried out reduced-scale cone penetration tests in For example, Parkin and Lunne (1982) investigated the effect
several sands with D50 ¼ 0.3–0.7 mm (corresponding to dc/ of flexible boundary conditions (similar to the samples of this
D50 ¼ 32–18) and observed no difference between their results study) on CPT resistance using different cone (dc) and
and those from a standard-size CPT, suggesting no scale effect. chamber (Dc) diameters. They found that the effect of chamber
Lee (1990) carried out a series of centrifuge CPT model boundary was negligible in loose sands (Drc r 30%) for Dc/
experiments on several gradations of Leighton Buzzard sand dc Z 20. While for very dense sands (with Drc E90%), Dc/
using three different miniature cones (with dc ¼ 19.05 mm, dc Z 50 was required to minimize chamber size effect.
10 mm, and 6.35 mm). They observed that the scale effect Similarly, Jamiolkowski et al. (2003) studied calibration
became appreciable only for dc/D50 o 16. Based on centrifuge chamber CPT data on Ticino and Hokksund sands, and
model experiments, Balachowski (2007) also observed no suggested correcting the CPT resistance for chamber boundary
scale effect for dc/D50 4 20. More recently, Sharp et al. effects only for Drc Z 40%. More recently, using 3D discrete
(2010) performed miniature cone penetration tests on fine element analyses, Butlanska et al. (2010) found no specimen
Nevada sand (with D50 ¼ 0.13 mm) using centrifuge testing size or boundary effects in sands at Drc o 45%. Based on the
and found no grain size effect for the dc/D50 ¼ 30.7. These
results suggest that particle size and scale effect are negligible Table 1
for the combination of cone diameter (dc ¼ 6 mm) and D50 Summary of miniature CPT tests performed in this study.
(¼ 0.193 mm) – corresponding to dc/D50 ¼ 31 – used in
Test no. Drc (%) p0 c (kPa) qc (MPa) fs (kPa)
this study.
1 22.0 100 3.96 38.54
2 24.5 100 3.99 32.32
2.4. Specimen saturation and consolidation 3 25.9 100 4.08 38.54
4 29.8 100 4.73 64.04
Although previous studies have shown similar cone pene- 5 36.6 100 5.10 50.43
6 38.6 100 5.26 52.70
tration resistance in dry and saturated clean sands (Bellotti
7 42.5 100 5.78 63.19
et al., 1988; Schmertmann, 1976; Bonita, 2000; Jamiolkowski 8 33.7 100 4.80 49.82
et al., 2001; Villet and Mitchell, 1981), moist tamped samples 9 33.7 45 2.52 14.40
of this study were saturated in order to remove the effects of 10 33.7 75 4.18 30.36
soil suction on cone resistance and to allow for the precise 11 33.3 150 6.02 47.85
12 33.7 200 7.09 49.50
measurement of the change in sample volume during con- 13 33.1 300 10.40 51.80
solidation. This was accomplished by flushing the soil speci- 14 33.7 450 13.75 63.10
men with CO2 and then de-aired water to remove air and to 15 30.8 500 10.29 42.00
dissolve the CO2. A back-pressure saturation procedure was 16 33.1 600 13.73 69.94
subsequently followed until a pore pressure coefficient of at 17 30.3 700 13.93 76.55
18 25.0 200 5.20 56.41
least 0.96 was achieved (Skempton, 1954). After specimen
19 27.4 300 6.55 51.30
saturation, the samples were subjected to an isotropic
464 A. Sadrekarimi / Soils and Foundations 56 (2016) 460–472

findings of these studies, the effects of specimen size and

boundary conditions on cone resistance are expected to be
negligible for the experiments of this study which were carried
out on sand samples with Dc/dc ¼ 25 at Drc o 43%. Therefore,
corrections for boundary conditions are not applied in the
comparisons and analyses presented here.

3. Experimental results

Fig. 3 presents cone penetration resistances (qc, fs) for the

series of miniature CPT tests of this study. According to these
plots, for the relatively uniform specimens of this study, the
cone tip resistance increased with penetration depth as soil
resistance to the insertion of the cone was fully mobilized.
After about 20 mm, qc approached a more-or-less uniform
plateau. Note that fs is plotted for penetration depths of greater Fig. 4. Comparison of CPT data of this study with soil identification zones of
than 10 mm where the friction sleeve was effectively inserted Mayne (2006).
into the soil away from the specimen's top cap. Average
magnitudes of qc and fs, determined after a cone penetration of
20 mm, are also summarized in Table 1. Due to the high
hydraulic conductivity and free-draining nature of the tested
Ottawa sand, cone penetration occurred under a primarily
drained condition with no excess pore water pressure being
generated adjacent to the cone tip (u2). Accordingly, correction
for the effect of unequal end areas was not needed (Robertson
and Cabal, 2015).

4. Comparison with existing empirical correlations

In the following paragraphs, the experimental results of this

study (summarized in Table 1) are compared with some of the
existing methods for CPT-based soil classification, stress
normalization, predicting sand relative density, unit weight,
and constrained modulus.
Fig. 5. Comparison of CPT data of this study with soil type zones of Eslami
4.1. Evaluation of soil classification methods and Fellenius (1997).

One of the useful applications of CPT is to identify the soil

type. Several studies have proposed empirical soil behavioral
charts or correlations for soil classification based on CPT
measurements. Figs. 4–6 compare the CPT data of this study
with the soil type boundaries or zones proposed by some of
these methods (Mayne, 2006; Eslami and Fellenius, 1997;
Robertson, 2009). In these plots, s0 vc and svc are the initial
effective and total vertical stresses, respectively, which corre-
spond to p0 c for the CPT experiments of this study, Pa ¼ 100 kPa
is a reference pressure, and qc,net ¼ qc  svc. Regarding Fig. 6,
Qtn ¼ qc,net/Pa  (Pa/s0 vc)n and FR¼ fs/qc,net  100 are the nor-
malized cone parameter and the friction ratio, respectively, and
n is a stress normalization exponent which depends on the soil
behavior (Robertson, 2009). According to Figs. 4 and 5, the soil
type identification charts of Mayne (2006) and Eslami and
Fellenius (1997) provide reasonable predictions of the clean
Ottawa sand used in this study. A few data points plot between Fig. 6. Comparison of CPT data of this study with soil identification zones of
Robertson (2009). Numbered zones correspond to 1: sensitive fine-grained soil,
the sands and silts or within the silt zones of these charts. This is 2: organic soil, 3: silty clay to clay, 4: clayey silt to silty clay, 5: silty sand to
likely to be because of the fine gradation (D50 ¼ 0.19 mm) of the sandy silt, 6: sand, 7: gravelly sand to dense sand, 8: very stiff sand to clayey
Ottawa sand used here compared to that used for developing the sand, and 9: very stiff fine-grained soil.
 A. Sadrekarimi / Soils and Foundations 56 (2016) 460–472 465

soil identification charts of Mayne (2006) and Eslami and s0 vc ¼ 100 kPa. These methods are briefly described and
Fellenius (1997). However, the experimental data plot entirely compared with the CPT experiments of this study.
as silty sands to sandy silts (zone 5) in Fig. 6, suggesting that An ideal stress normalization method should produce an
the empirical chart of Robertson (2009) might be less accurate equal normalized cone resistance from different depths or s0 vc
for identifying quartz fine sands. A few field samples from each in the same soil at the same Drc. As tests Nos. 8–17 were
soil stratum are recommended to confirm the predictions of soil conducted at close Drc (¼ 30.3–33.7%), they provide an
classification techniques. exceptional opportunity to evaluate the stress normalization
techniques of Table 2. Fig. 7a compares the stress normal-
ization methods for qc. In this figure, the qc measured in test
4.2. Evaluation of overburden stress normalization methods No. 8 (carried out at p0 c ¼ 100 kPa) is normalized by qc at other
p0 c in order to obtain stress normalization factors (qc1/qc). Note
In order to compare soil behavior from different depths, the that for the isotropically consolidated experiments of this
cone resistance is often normalized (or corrected) to a common study, s0 vc ¼ p0 c. According to Fig. 7, the widely used stress
effective overburden stress of 100 kPa (Wroth, 1984). As normalization factor of Liao and Whitman (1986) somewhat
summarized in Table 2, a number of techniques (Wroth, 1984; overestimates the experimental data at s0 vc 4 100 kPa, which
Robertson, 2009; Olsen and Mitchell, 1995; Kayen could lead to unconservative values for normalized qc (qc1) at
et al., 1992; Moss et al., 2006; Liao and Whitman, 1986; s0 vc 4 100 kPa. Based on a combination of theory and
Cetin and Isik, 2007; Idriss and Boulanger, 2006) are empirical observations, Moss et al. (2006) suggested a stress
suggested for converting the total (qc) or the net (qc, normalization exponent for normalizing both qc and fs, as
net ¼ qc  sv) cone tip resistances and sleeve friction to those follows:
that would have been measured if CPT had been carried out at !  ð  0:32qc ðMPaÞ  0:35 þ 0:49Þ
f s ðMPaÞ
 0:33 qc ðMPaÞ  100
c ¼ 0:78qc ðMPaÞ 
1:21
Table 2 abs log 10 þqc ðMPaÞ
CPT overburden stress normalization methods.
ð1Þ
Normalized parametera Reference
The above equation results in c E 0.5, which is the same as
Cone tip resistance Sleeve friction that suggested by Liao and Whitman (1986); and therefore, the
 0:5 resulting qc1 becomes relatively larger than those of this study
– Wroth (1984)
qc1;net ¼ qc;net sP0a at s0 vc 4 100 kPa. As shown in Fig. 7a, while the stress
 0:5
vc

– Liao and Whitman (1986) normalization method suggested by Kayen et al. (1992)
qc1 ¼ qc s0Pa
vc

qc1 ¼ 1:8
qc – Kayen et al. (1992) predicts slightly smaller qc1 than those of this study, the Drc-
0:8 þ s0vc =Pa
 c  c based normalization profile proposed by Idriss and Boulanger
qc1;net ¼ qc;net sP0a f s1 ¼ f s sP0a Olsen and Mitchell (1995)
 c
vc vc (2006) provides the best estimates for qc1/qc from the CPT
qc1;net ¼ qc;net Pa – Cetin and Isik (2007); Robertson experiments. Nevertheless, the choice of the overburden stress
s0vc
(2009)
 0:784  0:521Drc correction method seems to have a negligible impact on qc1 for
– Idriss and Boulanger (2006)
qc1 ¼ qc s0 vc/Pa ¼ 0.5  2.0 in Fig. 7a.
Pa
s0
 vc c  c
qc1 ¼ qc Pa
f s1 ¼ f s Pa Moss et al. (2006) Similar to Fig. 7a, Fig. 7b compares the stress normalization
s0vc s0vc
methods based on qc,net with the laboratory CPT data of this
a
svc and s0 vc are total and effective initial vertical stresses, respectively; Pa: study. According to this figure, these methods provide
atmospheric pressure ( E100 kPa); c: stress normalization exponent. different normalized net cone resistance (qc1,net), indicating a

Fig. 7. Comparison of existing stress normalization techniques with (a) qc1/qc and (b) qc1,net/qc,net from CPT experiments of this study.
466 A. Sadrekarimi / Soils and Foundations 56 (2016) 460–472

lack of consensus in stress normalization methods which could

lead to very different CPT interpretations. The overburden
stress correction of Wroth (1984) largely underestimates the
qc1,net measured in the CPT experiments at s0 vc 4 100 kPa.
Olsen and Mitchell (1995) collected over 2 decades of field
data and an extensive database of CPT calibration chamber
tests conducted by other researchers in order to deduce a CPT
tip normalization factor. They suggested the following rela-
tionship for the stress normalization exponent:
c ¼ 1 ðDrc  10%Þ  0:007 ð2Þ
The above equation yields c ¼ 0.86  0.83 for Drc ¼ 30.3–
33.7% in this study and the resulting stress normalization
factors tend to underestimate the qc1,net measured in the CPT
experiments at s0 vc 4 100 kPa. Cetin and Isik (2007) used a
Bayesian probabilistic analysis for the compilation of a
database of field tests, CPT calibration chamber tests, and
finite element numerical analysis data. They developed an
iterative procedure based on the following equations for
estimating a range of stress normalization exponents, as
follows:
R  272:38
c¼ 7 0:085; 272:38o R o 275:19 ð3Þ Fig. 8. Comparison of existing stress normalization techniques with fs1/fs from
275:19  272:38 the CPT experiments of this study.
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 2
f qc1;net
R¼ log s  100 þ 243:91 þ log  126:24 According to Fig. 8, the existing methods largely overestimate
qc Pa the effect of the overburden stress on fs. The experiments of
ð4Þ this study suggest a bilinear trend (see Fig. 8) with much larger
stress normalization factors for fs. As displayed in Fig. 8, fs1/fs
According to Fig. 7b, the qc1,net from the CPT experiments
decreases sharply with an increase in s0 vc up to about 100 kPa.
in this study tend to be smaller than the ranges of those
However, with further increases in s0 vc, the effect of s0 vc on fs1
produced by Eq. (3) at s'vc 4 100 kPa. More recently,
is significantly reduced. The physical mechanism underlying
Robertson (2009) suggested another iterative procedure for
this trend is investigated below.
determining a variable stress normalization exponent based on
The sleeve friction (fs) measured during a cone penetration
a soil behavior type (SBT) index, Ic, as seen below:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi test is essentially a function of the radial stress (s0 rf) and the
I c ¼ ð3:47  logQÞ2 þ ðlogF R þ 1:22Þ2 ð5Þ interface friction angle (δf) between the soil and the cone
sleeve (Huntsman, 1985). Based on observations made in the
The iteration starts with c¼ 1.0 in order to calculate the first driving of small-scale instrumented steel piles (Lehane et al.,
trial of Ic. If Ic r 1.64, then c¼ 0.5; otherwise, it is calculated 1993), the magnitude of s0 rf depends on the dilatancy of the
from the equation given below: sand adjacent to the cone sleeve (Lehane and White, 2005).
0 Since neither δf nor s0 rf were directly measured in the CPT
s
c ¼ 0:381I c þ 0:05 v  0:15 r 1:0 ð6Þ experiments of this study, s0 rf is approximately computed in
Pa Fig. 9 as s0 rf ¼ fs/tan(δf) for Drc ¼ 30.3–33.7%. A cone pene-
As demonstrated in Fig. 7b, this method is found to be tration of about 2–3 mm is often sufficient to mobilize a
conservatively biased by underestimating qc1,net as a result of constant-volume interface sleeve friction corresponding to the
the relatively large normalization exponents (0.67–1.0) pro- critical state of the sand adjacent to the cone sleeve (Lehane
duced by Eq. (6). As shown in the plots of Fig. 7, the CPT and White, 2005). Accordingly, δf ¼ 1/2ϕ0 cs ¼ 15.41
experiments suggest c¼ 0.612 for the quartz Ottawa sand used (ϕ0 cs ¼ 30.71 for the Ottawa sand of this study) is assumed in
in this study. This produces the same normalization factors as this analysis (Tejchman and Wu, 1995). According to Fig. 9,
those provided by Idriss and Boulanger (2006). the computed s0 rf is greater than the imposed horizontal
Although many studies have developed stress normalization boundary stress, s0 hc (which is the same as s0 vc and p0 c in
methods for cone tip resistance, according to Table 2, very few this study) at s0 vc r 150 kPa, where sand dilatancy is strong.
(Olsen and Mitchell, 1995; Moss et al., 2006) consider However, s0 rf drops sharply with an increase in s'vc beyond
correcting fs by merely extending the same factors (used for 150 kPa as the dilantancy of the sand is suppressed and its
qc) to fs1. Fig. 8 compares the normalization factors of these contractive tendency improves with increasing s0 vc. The
studies with the fs1/fs from the CPT experiments in this study. observed pattern is analogous to those measured in skin
Similar to qc1, fs1 is measured in test No. 8 at p0 c ¼ 100 kPa. friction on reduced-scale instrumented piles (Boulon and
 A. Sadrekarimi / Soils and Foundations 56 (2016) 460–472 467

Fig. 9. Effect of s0 vc on radial stress amplification (s0 rf/s0 hc) on cone sleeve Fig. 10. Comparison of dry unit weights (γd) predicted by Mayne (2007) with
based on computations of this study and measurements made in other those from experiments of this study for tests Nos. 1–8.
calibration chamber tests (Boulon and Foray 1986; Balachowski 2006).

Foray, 1986) as well as in rigid-wall calibration chamber

experiments on clean quartz sands (Balachowski, 2006). The
wide difference between the radial stress amplification factors
(s0 rf/s0 hc) of loose and dense sands in Fig. 9 emerges from the
greater dilatancy of the denser samples. Similar observations
have also been reported in several other experimental inves-
tigations (Baldi et al., 1981) as well as in in-situ tests on piles
with different lengths (Coyle and Castello, 1981). The trend of
fs1/fs with s0 vc/Pa observed in Fig. 9 reflects the changes in
(s0 rf/s0 hc), and thus, reduced sand dilatancy and increased
contractiveness with an increasing s'vc.

4.3. Evaluation of soil unit weight Fig. 11. Comparison of total unit weights (γt) predicted by Mayne et al. (2010)
with those from experiments of this study for tests Nos. 1–8.
Soil unit weight is a critical information for calculating the
initial geostatic and overburden stresses for CPT data proces-
0 0:05
sing and for interpreting many other geotechnical parameters. svc qc  svc 0:017
fs 0:073

0:16
The soil unit weight can be measured directly from undis- γ t ¼ 1:81γ w : : : Bq þ 1
Pa Pa Pa
turbed samples collected by thin-wall tube samples or ground
freezing techniques. However, such techniques can often be ð8Þ
expensive, difficult, and onerous in saturated clean sands or
in which Bq ¼ Δu2/qc,net is the normalized pore water pressure
gravels and they require costly equipment. Therefore, indirect
parameter. Figs. 10 and 11 compare these correlations with γd
empirical correlations with CPT measurements are developed
and γt of test Nos. 1–8, conducted at p0 c ¼ 100 kPa. According
and regularly used for quicker processing of cone penetration
to these figures, Eq. (7) provides close estimates of γd as
data and preliminary geotechnical analyses.
Eq. (7) is derived based on calibration chamber tests on silica
From the regression analysis of a large database of calibra-
sands (Mayne, 2007). However, the predictions of Eq. (8) are
tion chamber cone penetration tests in clays, silts, and sands,
on average about 1 kN/m3 lower than those from the CPT
Mayne et al. (2010) and Mayne (2007) suggested the following
experiments of this study.
correlations for predicting dry (γd) and total (γt) unit weights
This probably stems from the original development of Eq. (8)
from the cone tip resistance:
as an average correlation based on regression analysis of CPT in
! different soil types, including clays, silts, sands, tills, and largely

 q =P a
γ d kN=m3 ¼ 1:89log
0 c
0:5 þ 11:8 ð7Þ dominated by clayey soils. Based on the CPT tests in this study
svc =Pa (Figs. 10 and 11), the following relationships are suggested for
estimating γd and γt for clean silica sands (similar to the Ottawa
468 A. Sadrekarimi / Soils and Foundations 56 (2016) 460–472

sand used here): Table 3

! Summary of correlation parameters suggested by earlier studies for fitting

 q =Pa Eq. (11) based on s0 vc.
γ d kN=m3 ¼ 3:00log
0 c 0:5 þ 9:44 ð9Þ
svc =Pa No. Sand C0 C1 C2 Reference
!

 q =Pa a Several NC sands 0.050 0.700 2.91 Schmertmann (1978)
γ t kN=m3 ¼ 2:00log
0 c 0:5 þ 15:69 ð10Þ b Ticino 0.157 0.550 2.41 Baldi et al. (1986)
svc =Pa c Hokksund 0.086 0.530 3.29
d Ticino 0.140 0.550 2.90 Jamiolkowski et al.
A particular difference between Eqs. (8) and (10) is the e Ticino, Toyoura, 0.175 0.500 3.10 (2001)
consideration of fs in Eq. (8). The nearly perfect relationship Hokksund
between γt and qc, provided by Eq. (10), implies a weaker
connection between fs and γt (compared to qc), and that fs could
be excluded for estimating γt. Table 4
Summary of correlation parameters suggested by earlier studies for fitting
Eq. (11) based on p0 c.
4.4. Evaluation of sand relative density
No. Sand C0 C1 C2 Reference
In addition to the soil unit weight, the relative density is f Ticino 0.181 0.550 2.61 Baldi et al. (1986)
extensively used in geotechnical engineering as an index g Hokksund 0.153 0.550 2.88
parameter for characterizing the shearing response of granular h Ticino 0.205 0.510 2.93 Jamiolkowski et al.
soils and it is often used in ground improvement projects for (1988)
i Ticino 0.174 0.560 2.97 Jamiolkowski et al.
compaction quality control. Considering the difficulties in j Ticino, Toyoura, 0.297 0.460 2.96 (2001)
obtaining undisturbed samples, CPT is a viable tool for Hokksund
estimating the Drc of in-situ sands. The experimental data k Da Nang (Vietnam) 0.369 0.500 2.34 Huang and Hsu (2005)
from this study are used here to evaluate some of the existing
predictive correlations for Drc. For normally consolidated and
unaged fine to medium clean sands, Schmertmann (1976)
suggested the following correlation between qc and Drc:
!
100 qc ðMPaÞ
Drc ð%Þ ¼ ln
C ð11Þ
C2 C 0 s0 1
vc

in which C0, C1, and C2 are empirical fitting parameters and s0 vc

is the initial effective vertical stress. Since the pioneering work
of Schmertmann (1976), several investigators (Baldi et al.,
1986; Villet and Mitchell, 1981; Jamiolkowski et al., 2001;
Schmertmann, 1978) have fitted Eq. (11) to CPT calibration
chamber tests on many sands. However, some studies indicate
that the qc measured in calibration chamber tests is largely
influenced by the horizontal effective stress, s0 hc (Baldi et al.,
1986; Houlsby and Hitchman, 1988; Jamiolkowski et al., 2001;
Holden, 1971). Accordingly, Eq. (11) has also been expressed
(Baldi et al., 1986; Jamiolkowski et al., 2001; Huang and Hsu, Fig. 12. Comparison of Drc from CPT experiments of this study with those
2005; Jamiolkowski et al., 1988) as a function of the mean predicted by other studies based on s0 vc. Line labels are a: Schmertmann
consolidation pressure, p0 c, in order to include the effects of both (1978), b: Baldi et al. (1986) based on Ticino sand, c: Baldi et al. (1986) based
s0 vc and s0 hc. For field applications, p0 c can be determined by on Hokksund sand, d: Jamiolkowski et al. (2001) based on Ticino sand, e:
assuming Ko ¼ 1 sin (ϕ0 ) for normally consolidated young Jamiolkowski et al. (2001) based on Ticino, Toyoura and Hokksund sands, and
l: Kulhawy and Mayne (1990).
sand deposits in which ϕ' is the sand's effective stress friction
angle. Tables 3 and 4 summarize the fitting parameters of Eq.
(11) suggested by these studies. suggesting a greater effect of s0 vc than that incorporated
Figs. 12 and 13 compare the predictions made in these through p0 c. The s0 vc-based correlation of Baldi et al. (1986),
studies with the CPT experiments based on s0 vc and p0 c, from calibration chamber CPT tests on Hokksund sand,
respectively. For the isotropically-consolidated samples in this provides the closest estimates to the Drc in this study, while
study, s0 vc ¼ p0 c. Except for the s0 vc-based correlations pro- those of Schmertmann (1978) largely overestimate Drc.
posed by Baldi et al. (1986) (for Hokksund sand) and by In summary, the correlations proposed by different studies
Schmertmann (1978), the remaining methods underestimate predict a very wide range of Drc because of the different types
the Drc values in this study. The overall amount of under- of sands used in their databases. Therefore, a unique form of
estimation seems to be greater for the p0 c-based methods, Eq. (11) cannot be determined for all sands, and particularly
 A. Sadrekarimi / Soils and Foundations 56 (2016) 460–472 469

Fig. 13. Comparison of Drc from CPT experiments of this study with those Fig. 14. Comparison of Drc from CPT experiments of this study with ranges of
predicted by other studies based on p0 c. Line labels are f: Baldi et al. (1986) those predicted by Eq. (13).
based on Ticino sand, g: Baldi et al. (1986) based on Hokksund sand, h:
Jamiolkowski et al. (1988), i: Jamiolkowski et al. (2001) based on Ticino sand,
j: Jamiolkowski et al. (2001) based on Ticino, Toyoura and Hokksund sands, is consistent with the stress normalization exponent determined
and k: Huang and Hsu (2005). in Fig. 7.

for variable soil deposits. In other words, different sands will 4.5. Evaluation of soil stiffness
not necessarily behave the same at the same Drc and p0 c. For
the CPT experiments in this study, fitting parameters of Soil stiffness describes the soil deformation behavior under
C0 ¼ 0.119, C1 ¼ 0.612, and C2 ¼ 2.55 are found following an an increment in confining stress or shear stress. Several
optimization process by minimizing the standard deviation investigators have attempted to relate soil stiffness to cone
between the calculated and the measured qc. However, the penetration resistance (Schmertmann, 1978; Tanaka and
values for Drc predicted with these parameters in Eq. (11), are Tanaka, 1998; Mayne, 2006; Tonni et al., 2010). Here, the
still within 7 6.5% of the actual values. This indicates the bulk moduli (K) of the samples were determined from the
limited accuracy of Eq. (11) which merely uses p0 c and qc isotropic consolidation stages of the CPT experiments as the
(Sladen, 1989; Huang and Hsu, 2005; Hamidi et al., 2013). incremental change in effective confining pressure (∂p0 )
Accordingly, Kulhawy and Mayne (1990) and Jamiolkowski divided by the volumetric strain (εv). These were then
et al. (2001) have proposed empirical relationships for converted to constrained moduli (M) using the following
estimating Drc with additional consideration regarding the relationship for each experiment:
effect of sand compressibility. Kulhawy and Mayne (1990) 3K ð1  νÞ
compiled the results of 24 sets of calibration chamber tests on M¼ ð14Þ
ð1 þ νÞ
fine to medium sands and suggested the following relationship:
A Poisson's ratio ν (¼  εr/εa) of 0.36 was also determined
1 q =Pa
D2rc ¼ :
c 0:5 ð12Þ based on the measurements of axial strain (εa) and volumetric
305Qc s0 =Pa strain (εv ¼ εa þ 2εr) of the same sand in consolidated drained
vc
triaxial compression shear tests. In these experiments, the sand
in which Qc is a compressibility factor equal to 0.91, 1.00, and
specimen was loaded in axial compression while maintaining a
1.09 for high, medium, and low compressibility sands,
constant radial confining pressure. In practice, M is often
respectively. According to Fig. 12, Eq. (12) (labeled “l”)
related by a constant coefficient to qc,net or qc (Schmertmann,
provides close estimates of Drc for qc/(s0 vc)0.5 ¼ 200–327,
1978). As illustrated in Fig. 15a, constrained moduli from the
while underestimating at greater qc/(s0 vc)0.5 values. From a
experiments of this study are very close to the empirical
review of the calibration chamber tests data, Jamiolkowski et
relationship (M ¼ 5qc,net) suggested by Mayne (2006) for
al. (2001) suggested a modified form of Eq. (11) as below:
normally consolidated clean sands. The ranges of M are also

 !
qc =Pa close and within the lower bound of those determined by
Drc ð%Þ ¼ 26:8ln
 C  bx ð13Þ Veismanis (1974) and Lunne and Christoffersen (1983), as
s0vc =Pa 1
seen in Fig. 15b.
in which C1 is a stress normalization exponent (¼ 0.50) similar
to Eq. (11), and bx ¼ 52.5, 67.5, 82.5 are for high, medium, and 4.6. Effect of lateral stress ratio (Kc)
low compressibility sands, respectively. As illustrated in Fig.
14, the experiments of this study plot within the medium to Vast experimental evidences (Huntsman, 1985; Jamiolkowski
low compressibility range (bx ¼ 70.2) using C1 ¼ 0.612, which et al., 1985; Baldi et al., 1986; Houlsby and Hitchman, 1988;
470 A. Sadrekarimi / Soils and Foundations 56 (2016) 460–472

Fig. 15. Comparison of constrained moduli for CPT samples with empirical correlations based on (a) qc,net and (b) qc u2..

Mayne and Kulhawy, 1991; Salgado, 1993; Ahmadi et al., 2005), or around the boundaries for sand and silty sand of the soil type
as well as theoretical analyses of cone penetration tests using identification plots. The experimental results suggested an over-
bearing capacity, cavity expansion, strain path, and finite element or burden stress normalization exponent of 0.612 for the quartz sand
a combination of these techniques (Vesic, 1975; Teh and Houlsby, used in this study for Drc ¼ 30.3–33.7%. Different overburden
1991; Salgado et al., 1997b; Yu et al., 2000), indicate that both qc stress normalization techniques were then evaluated by compar-
and fs are strongly correlated with the horizontal effective stress ison with the cone resistance measured in this study. The Drc-
(s0 hc), rather than s0 vc, whereas the relative density, the unit weight, based stress normalization scheme of Idriss and Boulanger (2006)
and the modulus of cohesionless soil are controlled by both s'hc and provided the closest estimates to qc1/qc from the CPT experiments.
s'vc. On the other hand, the existing overburden stress correction
Despite the primary effect of s0 hc on qc, CPT interpretation methods largely underestimated the normalized sleeve friction
methods are predominantly based on s0 vc, as s0 vc is easily resistance (fs1). Hence, a bilinear relationship was proposed for
calculated with reasonable certainty from the soil unit weight normalizing sleeve friction based on the experiments of this study.
and a knowledge of the location of the groundwater table. The relationship indicates a sharply reducing normalization factor
Thus, these methods are compared based on s0 vc with the CPT with increasing effective stress up to about 100 kPa and a
experiments of this study for which s0 vc ¼ s0 hc, whereas field significantly reduced gradient with further increases in effective
stress conditions are seldom isotropic. This could have stress (4100 kPa). Specific CPT-based correlations were sug-
produced some unknown bias in the comparisons of Figs. gested for estimating the dry and saturated unit weights of quartz
10–12. sands from the CPT experiments in this study. Several empirical
Note that both the normalized penetration resistances (qc1, correlations for predicting relative density from CPT resistance
fs1, or qc1,net) and the original penetration resistances (qc, fs, or were also reviewed in this study. The existing correlations
qc,net) are similarly affected by changes in Kc; and therefore, exhibited very wide ranges in relative density predictions. These
the stress normalization factors (qc1/qc, fs1/fs, or qc1,net/qc,net) of relationships are not universally applicable to all sand types, and
Figs. 7 and 8 are unaffected by the lateral stress ratio, Kc ¼ s0 hc/ therefore, cannot be used as reliable ground improvement
s0 vc. This is further supported by cavity expansion analysis acceptance criteria. The constrained moduli of the CPT samples
(Salgado et al., 1997a; Moss et al., 2006), in which the stress also agreed very well with an empirical correlation proposed by
normalization exponent is found to vary by less than 1% with a Mayne (2006) for clean sands.
change in Kc from 0.5 to 1.0 for loose sands with Drc o 45%.
The effect of the change in Kc on the constrained moduli Acknowledgments
comparison in Fig. 15 is also small because of the much larger
magnitude of qc compared to s0 vc (in qc,net). The research described in this study was carried out with
funding provided to the author by the Natural Sciences and
Engineering Research Council of Canada (NSERC: Discovery
5. Summary and conclusions Grant). The author is also grateful to Mr. Erol Tas, a technician
at Western University's Soil Mechanics Laboratory, and to
A series of 19 miniature cone penetration tests was described in graduate student Ms. Sepideh Damavandi for their assistance
this study. The tests were used to evaluate some of the existing in performing the CPT experiments.
empirical methods for soil type identification and to determine the
relative density, the unit weight, the stress normalization, and the References
modulus of loose to medium-dense sands. Due to the fine
gradation (Dmax ¼ 0.85 mm) of the Ottawa sand used in the Abedin, M.Z., 1995. The characterization of unsaturated soil behaviour from
CPT experiments in this study, the results were plotted within penetrometer performance and the critical state concept. In: Department of
 A. Sadrekarimi / Soils and Foundations 56 (2016) 460–472 471

Civil Engineering and Geosciences (Ph.D. thesis). Newcastle University, Eslami, A., Fellenius, B.H., 1997. Pile capacity by direct CPT and CPTu
Tyne and Wear, United Kingdom. methods applied to 102 case histories. Can. Geotech. J. 34 (6), 886–904.
Ahmadi, M.M., Byrne, P.M., Campanella, R.G., 2005. Cone tip resistance in Franzen, J.H., 2006. Cone penetration resistance in silt (M.S. thesis).
sand: modeling, verification, and applications. Can. Geotech. J. 42 (4), Department of Civil & Environmental Engineering, University of Rhode
977–993. Island, Kingston, RI.
Astm, 2006a. Standard D4253: standard test methods for maximum index Gui, M.W., Bolton, M.D., Garnier, J., Corte, J.F., Bagge, G., Laue, J., Renzi,
density and unit weight of soils using a vibratory table. In: Annual Book of R., 1998. Guidelines for cone penetration tests in sands. Taylor & Francis
ASTM Standards, vol. 04.08. ASTM International, West Conshohocken, 155–160.
PA. Hamidi, B., Varakshin, S., Nikraz, H., 2013. Relative density correlations are
Astm, 2006b. Standard D4254: Standard test methods for minimum index not reliable criteria. Ground Improv. 166 (GI4), 196–208.
density and unit weight of soils and calculation of relative density. In: Holden, J., 1971. Research on Performance of Soil Penetrometers. Gainesville,
Annual Book of ASTM Standards, vol. 04.08. ASTM International, West Florida. Report Report CE-SM-71-71.
Conshohocken, PA. Houlsby, G.T., Hitchman, R., 1988. Calibration chamber tests of a cone
Astm, 2011. Standard D2487: standard Practice for Classification of Soils for penetrometer in sand. Geotechnique 38 (1), 39–44.
Engineering Purposes (Unified Soil Classification System). In: Annual Huang, A.B., Hsu, H.H., 2005. Cone penetration tests under simulated field
Book of ASTM Standards, vol. 04.08. ASTM International, West conditions. Geotechnique 55 (5), 345–354.
Conshohocken, PA. Huntsman, S., 1985. Determination of in-situ lateral pressure of cohesionless
Balachowski, L., 2006. Penetration resistance of Lubiatowo sand in calibration soils by static cone penetrometer (Ph.D.). Department of Civil and
chamber tests. Arch. Hydro-eng. Environ. Mech. 53 (4), 311–329. Environmental Engineering,University of California, Berkeley, California.
Balachowski, L., 2007. Size effect in centrifuge cone penetration tests. Arch. Idriss, I.M., Boulanger, R.W., 2006. Semi-empirical procedures for evaluating
Hydro-eng. Environ. Mech. 54 (3), 161–181. liquefaction potential during earthquakes. Soil Dyn. Earthq. Eng. 26 (2-4),
Baldi, G., Bellotti, R., Ghiona, V., Jamiolkowski, M., Pasqualini, E., 1986. 115–130.
Interpretation of CPTs and CPTUs. Part II: drained penetration of sands. In: Jacobs, P.A., Coutts, J.S., 1992. A comparison of electric piezocone tips at the
Fourth International Geotechnical Seminar on Field Instrumentation and in Bothkennar test site. Geotechnique 42 (2), 369–375.
situ Measurements. Nanyang Technological Institute, Singapore143–156. Jamiolkowski, M., Ladd, C.C., Germaine, J.T., Lancelotta, R., 1985. New
Baldi, G., Bellotti, R., Ghionna, V., Jamiolkowshi, M., Pasqualini, E., 1981. developments in field and laboratory testing of soils. vol. 1. A.A. Balkema,
Cone resistance of a dry medium sand. In: Proceedings of 10th Interna- San Fransisco, CA57–153.
tional Conference on Soil Mechanics and Foundation Engineering, vol. 2. Jamiolkowski, M., Lo Presti, D.C.F., Manassero, M., 2001. Evaluation of
pp. 427 432. rntive density and shear strength of sands from CPT and DMT. Germaine,
Baldi, G., Bellotti, R., Ghionna, V.N., Jamiolkowshi, M., Pasqualini, E., 1982.
J.T., Sheahan, T.C., Whitman, R.V. (Eds.), Soil Behavior and Soft Ground
Design parameters for sands from CPT. In: Proceedings of Second
Construction, ASCE Geotechnical Special Publication, 119. American
European Symposium Penetration Testing, ESOPT II. pp. 425 432.
Society of Civil Engineers (ASCE), Reston, Virginia, pp. 201–238.
Baldi, G., O'neill, D.A., 1995. Developments in penetration technology for
Jamiolkowski, M., Lo Presti, D.C.F., Manassero, M., 2003. Evaluation of
geotechnical and environmental applications. In: Proceedings of the
relative density and shear strength of sands from CPT and DMT.
International Symposium on Cone Penetration Testing.
Germaine, J.T., Sheahan, T.C., Whitman, R.V. (Eds.), Soil Behavior and
Been, K., Jefferies, M.G., Crooks, J.H.A., Rothenburg, L., 1987a. The cone
Soft Ground Construction, ASCE Geotechnical Special Publication, 119.
penetration test in sands. Part II, general inference of state. Geotechnique
American Society of Civil Engineers (ASCE), pp. 201–238.
37 (3), 285–299.
Jamiolkowski, M.J., Ghionna, V.N., Lancellotta, R., Pasqualini, E., 1988. New
Been, K., Lingnau, B.E., Crooks, J.H.A., Leach, B., 1987b. Cone penetration
correlations of penetration tests for design practice. In: Ruiter, J.D. (Ed.),
test calibration for Erksak (Beaufort Sea) sand. Can. Geotech. J. 24,
Penetration Testing 1988: Proceedings of the First International Sympo-
601–610.
Bellotti, R., Crippa, V., Pedroni, S., Ghionna, V.N., 1988. Saturation of sand sium on Penetration Testing, ISOPT-1, Orlando, 20–24 March 1988. A.A.
specimen for calibration chamber tests. Deruiter, J. (Ed.), Proceedings of Balkema, Rotterdam; Brookfield, VT, USA, pp. 263–296.
the 1st International Symposium on Penetration Testing (ISOPT-1), vol. 2. Kayen, R.E., Mitchell, J.K., Seed, R.B., Lodge, A., Nishio, S., Coutinho, R.,
Balkema, Rotterdam, The Netherlands, Orlando, Florida, pp. 661–671. 1992. Evaluation of SPT-, CPT-, and shear wavebased methods for
Bonita, J.A., 2000. The Effects of Vibration on the Penetration Resistance and liquefaction potential assessments using Loma Prieta data. In: Proceedings
Pore Water Pressure in Sands. In: Civil Engineering. Virginia Polytechnic of the 4th Japan–U.S. Workshop on Earthquake Resistant Design of
Institute, Blacksburg, VA vol. PhD. Lifeline Facilities and Countermeasures for Soil Liquefaction. Honolulu,
Boulon, M., Foray, P., 1986. Physical and numerical simulation of lateral shaft Hawaii, vol. 1, pp. 177–192.
friction along offshore piles in sand. In: Proceedings of 3rd International Kokusho, T., Ito, F., Nagao, Y., Green, R.A., 2012. Influence of non/low-
Conference on Numerical Methods in Offshore Piling. pp. 127 147. plastic fines and associated aging effects on liquefaction resistance. J.
Butlanska, J., Arroyo, M., Gens, A., 2010. Size effects on a virtual calibration Geotech. Geoenviron. Engineering, ASCE 138 (6), 747–756.
chamber. In: Benz, T., Nordal, S. (Eds.), Numerical Methods in Geotech- Kulhawy, F.H., Mayne, P.H., 1990. Manual on estimating soil properties for
nical Engineering. Taylor & Francis Group, London, pp. 225–230. foundation design. Palo Alto, California. Report Report EL-6800. pp. 2–
Canou, J., El Hachem, M., Kattan, A., Juran, I., 1988. Mini piezocone (M- 38.
CPTU) investigation related to sand liquefaction analysis. In: Proceedings Kumar, J., Raju, K.V.S.B., 2009. Miniature cone tip resistance of sand with fly
of the First International Symposium on Penetration Testing, ISOPT-1. ash using triaxial setup. Can. Geotech. J. 46 (2), 231–240.
Orlando, Florida. pp. 699 706. Ladd, R.S., 1978. Preparing test specimen using undercompacton. Geotech.
Cetin, K.O., Isik, N.S., 2007. Probabilistic assessment of stress normalization Test. J. ASTM 1 (1), 16–23.
for CPT data. J. Geotech. Geoenviron. Eng. 133 (7), 887–897. Lee, S.Y., 1990. Centrifuge modeling of cone penetration testing in cohesion-
Chen, C.C., 2000. Shear induced evolution of structure in water-deposited sand less soils (Ph.D. thesis). Cambridge University, Cambridge, UK.
specimens. In: School of Civil and Environmental Engineering. Georgia Lehane, B.M., Jardine, R.J., Bond, A.J., Frank, R., 1993. Mechanisms of shaft
Institute of Technology, Atlanta, GA, 497. friction in sand from instrumented pile tests. J. Geotech. Engineering,
Coyle, H.M., Castello, R.R., 1981. New design correlations for piles in sand. J. ASCE 119 (1), 19–35.
Geotech. Eng. Div., ASCE 107 (GT7), 965–985. Lehane, B.M., White, D.J., 2005. Lateral stress changes and shaft friction for
Damavandi, S., Sadrekarimi, A., 2015. Development of a miniature cone model displacement piles in sand. Can. Geotech. J. 42 (4), 1039–1052.
penetrometer for calibration chamber testing. Geotech. Test. J., ASTM, 38; Liao, S.S.C., Whitman, R.V., 1986. Overburden correction factors for SPT in
878–892. sand. J. Geotech. Eng. 112 (3), 373–377.
472 A. Sadrekarimi / Soils and Foundations 56 (2016) 460–472

Lunne, T., Christoffersen, H.P., 1983. Interpretation of cone penetrometer data Salgado, R., Mitchell, J.K., Jamiolkowski, M., 1997b. Cavity expansion and
for offshore sands. In: Proceedings of 1st Offshore Technology Con- penetration resistance in sand. J. Geotech. Geoenviron. Eng. 123 (4),
ference, vol. 1, pp. 181 192. 344–354.
Lunne, T., Robertson, P.K., Powell, J.J.M., 1997. Cone Penetration Testing in Schlosser, F., 1985. Liquefaction de veines de sable lache dans des talus aous-
Geotechnical Practice, 1st edn. Blackie Academic & Professional, London marins. In: Proceedings of 11th International Conference on Soil
New York. Mechanics and Foundation Engineering.
Mayne, P.W., 2006. In-situ test calibrations for evaluating soil parameters. In: Schmertmann, J.H., 1976. An Updated Correlation Between Relative Density,
Phoon, K.K., Hight, D.W., Leroueil, S., Tan, T.S. (Eds.), Second Dr, and Fugro-Type Electric Cone Bearing, qc. Vicksburg, MS, Report
International Workshop on Characterisation and Engineering Properties Report DACW 39-76 M6646.
of Natural Soils, 29 November1 December. Singapore, vol. 3; 2006, pp. Schmertmann, J.H., 1978. Guidelines for Cone Penetration Test: Performance
1–56. and Design. Washington, D.C., Report Report FHWA-TS-78-209.
Mayne, P.W., 2007. Invited overview paper: in-situ test calibrations for Sharp, M.K., Dobry, R., Phillips, R., 2010. CPT-Based Evaluation of
evaluating soil parameters. In: Characterization and Engineering Properties Liquefaction and Lateral Spreading in Centrifuge. J. Geotech. Geoenviron.
of Natural Soils, vol. 3. Taylor and Francis, London1602–1652. Eng. 136 (10), 1334–1346.
Mayne, P.W., Kulhawy, H., 1991 Calibration chamber database and boundary Skempton, A.W., 1954. The pore pressure coefficient A and B. Geotechnique 4
effects correction for CPT data. In: Huang, A.-B. (ed.) Proceedings of 1st (4), 143–147.
International Symposium on Calibration Chamber Testing. pp. 257264. Sladen, J.A., 1989. Problems with interpretation of sand state from cone
Mayne, P.W., Peuchen, J., Bouwmeester, D., 2010. Soil unit weight estimation penetration test. Geotechnique 39 (2), 323–332.
from CPTs. In: Proceedings of the 2nd International Symposium on Cone Sladen, J.A., D'hollander, R.D., Krahn, J., 1985. The liquefaction of sands, a
Penetration Testing. collapse surface approach. Can. Geotech. J. 22 (4), 564–578.
Moss, R.E.S., Seed, R.B., Olsen, R.S., 2006. Normalizing the CPT for Tanaka, H., Tanaka, M., 1998. Characterization of sandy soils using CPT &
overburden stress. J. of Geotech. and Geoenviron. Eng. 132 (3), 378–387. DMT. Soils Found. 38 (3), 55–65.
Olsen, R.S. & Mitchell, J.K. (1995) CPT stress normalization and prediction of Teh, C.I., Houlsby, G.T., 1991. An analytical study of the cone penetration test
soil classification. In Proceedings of the International Symposium On Cone in clay. Geotechnique 41 (1), 17–34.
Penetration Testing, CPT'95.), Linkoping, Sweeden, vol. 2, pp. 257 262. Tejchman, J., Wu, W., 1995. Experimental and numerical study of sand–steel
Omar, T., Sadrekarimi, A., 2015. Specimen size effects on behavior of loose interfaces. Int. J. Numer. Anal. Methods Geomech. 19 (8), 513–536.
sand in triaxial compression tests. Can. Geotech. J. 52 (6), 732–746. Tonni, L., Gottardi, V., Berengo, V., Simonini, P., 2010. Classification,
Parkin, A.K., Lunne, T., 1982. Boundary Effects in the Laboratory Calibration overconsolidation and stiffness of Venice lagoon soils from CPTU. In:
of a Cone Penetrometer for Sand, vol. 2. A.A. Balkema, Rotterdam Proceedings of 2nd International Symposium on Cone Penetration Testing.
761–767. Veismanis, A., 1974. Laboratory investigation of electrical friction - cone
Pournaghiazar, M., Russell, A.R., Khalili, N., 2011. Development of a new penetrometers in sand. In: Proceedings of European Symposium on
calibration chamber for conducting cone penetration tests in unsaturated Penetration Testing, vol. 2, pp. 407 420.
soils. Can. Geotech. J. 48 (2), 314–321.
Vesic, A.S., 1975. Bearing capacity of shallow foundation. In: Winterkorn,
Robertson, P.K., 2009. Interpretation of cone penetration tests – a unified
Fang (Eds.), Foundation Engineering Handbook. Van Nostrand Reinhold,
approach. Can. Geotech. J. 46 (11), 1337–1355.
New York, pp. 121–147.
Robertson, P.K., Cabal, K.L., 2015. Guide to Cone Penetration Testing for
Villet, W.C.B., Mitchell, J.K., 1981. Cone resistance, relative density, and
Geotechnical Engineering, 6th edition Gregg Drilling & Testing Inc.,
friction angle. In: ASCE National Convention on Cone Penetration Testing,
Martinez,California.
ASCE. American Society of Civil Engineers, St. Louis, Missouri178–208.
Salgado, R., 1993. Analysis of penetration resistance in sands (Ph.D.).
Wroth, C.P., 1984. The interpretation of in situ soil tests. Geotechnique 34 (4),
Department of Civil and Environmental Engineering, University of
449–489.
California, Berkeley, California.
Yang, C.T., 2005. Boundary condition and inherent stratigraphic effects on
Salgado, R., 2013. The mechanics of cone penetration: contributions from
microstructure evolution in sand specimens (Ph.D.). Department of Civil
experimental and theoretical studies. In: Coutinho, R.Q., Mayne, P.W.
and Environmental Engineering, Georgia Institute of Technology, Atlanta,
(Eds.), Geotechnical and Geophysical Site Characterization 4, ISC4. CRC
Georgia, 560.
Press, Boca Raton, FL, USA, pp. 131–153.
Yu, H.S., Herrmann, L.R., Boulanger, R.W., 2000. Analysis of steady cone
Salgado, R., Boulanger, R.W., Mitchell, J.K., 1997a. Lateral stress effects on
penetration in clay. J. Geotech.Geoenviron. Eng. 126 (7), 594–605.
CPT liquefaction resistance correlations. J. Geotech. Geoenviron. Eng. 123
(8), 726–735.