3rd International Symposium on Cone Penetration Testing, Las Vegas, Nevada, USA - 2014

Relationship between CPT parameters and properties of saturated ultra-

soft clays in the Pearl River delta
Z.M. Li
Geotechnical Engineering Technology Center，Guangdong University of Technology, Guangzhou, China

ABSTRACT: Theoretical relationships between CPT parameters, elastic parameters and undrained
strength of soil are presented based on an elastic-plastic solution for cylindrical cavity expansion. The re-
lation between CPT parameters and shear strength from vane test is also presented from the theoretical
result. Thus, the CPT parameters can be determined directly by elastic parameters and shear strength or
vane shear parameters; and vice versa. That makes possible to save the high test cost and provides theo-
retical formulas to avoid some tests limited due to the site and/or other condition. Results are compared
between the theoretical results and in situ data from CPT and VST, at a large-scale ultra-soft ground
treatment project in the Pearl River Delta The results showed consistency of the two kinds of results, as
well as the feasibility and practicality of the theoretical formulas.

1 INTRODUCTION

The Cone Penetration Test (CPT) is usually used to determine the geotechnical engineering properties of
soils and to delineate soil stratigraphy, and is one of the most widely used and accepted method for soil
investigation worldwide, including in China. As large-scale basic construction is conducted in China,
especially in the Pearl River delta, they inevitably encounter a lot of soft soil ground, even ultra-soft soil
whose natural void ratio is greater than 1.5, and water content greater than the liquid limit. The CPT is a
method to determine the basic parameters of soil characteristics, including tip resistance qc, side friction
fs, friction ratio Rf（= fs/qc x 100%）and specific penetration resistance Ps （the quotient of total pene-
tration force and projected area of the probe. These parameters can be utilized to determine the bearing
capacity for foundations, but (according to the Building Code in China (Code DBJ 15-60-2008) the CPT
can not be used to estimated the undrained shear strength Su and soil sensitivity St, which are usually ob-
tained from In-Situ Vane Shear Test (VST). However, both CPT and VST reflect, to some extent, the
soil characteristic of the resistance to shear force, and there is reason to believe that certain correspond-
ing relationship exists between CPT and VST parameters. In fact, many researchers have made the link
(Chui & Ding 2004, Lin 1994, Li 2011, Robertson & Campanella 1988, Robertson & Cabal 2012, Yan
et al. 2009, Yang & Xiong 2010). However, for saturated ultra-soft clays, related work is very rare.
Based on elastic-plastic analysis of the cavity expansion process and the penetration characteristics of
saturated soft clays, this paper is aimed at establishing theoretical relations among VST, CPT and other
often used mechanical parameters, and comparing in site measured data to evaluate the relationships to
provide a basis for further understanding of the nature of ultra soft soil, and also to reduce some tests so
as to save test cost and improve test efficiency in practice.

393
2 BASIC EQUATION ESTABLISHMENT FOR PARAMETERS OF SATURATED SOFT CLAY
2.1 Empiric method
A large number of test data were utilized for statistics analysis, to establish certain relations between tip
resistance qc, side friction fs and undrained shear strength of cohesive soils Su. The typical statistical fit-
ting equations are shown in Tab. 2-1 (Li 2011, Robertson & Campanella 1988, Lin 1994).

Table 2-1. Empiric relation between qc or Ps with Su

No
Practical relation Applied conditions Sources*
.
1 Su  0.071qc  1.28 qc < 700kPa Tongji university

3
S u  0.039qc  2.7 qc < 800kPa Ministry of Railways

Su  0.0696Ps  2.7 Ps =300～1200

3 Wuhan CPT Union
Saturated soft clay
qc =100~800kPa
4 Su  0.0543qc  4.8 Soft clay in Shanghai and Guang- Sichuan institute of building
zhou
Ps =100～1500kPa
5 Su  0.0308PS  4.0 Navigation design and research institute
Newport soft clay
6 Su  0.05PS Newport soft clay The 3rd Inst. of Ministry of Railways
Su  0.0579PS  1.9 Ps =200～1100kPa
7 First industry design institute of Jiangsu
Saturated soft clay in Xuzhou
8 Su  0.0564PS  1.8 Ps <700kPa The 4th Inst. of Ministry of Railways
Su  0.057 PS lake peat and peat soil of Dian- Hydropower survey and design institute of
9
chi Hunan
10 Su  qc   c  / Nc ** ------ Lin Zong-yuan

11 qc  Cu N c   v 0 *** ------ Robertson & Campanella, (1988)

Su  0.063qc  1.91 100kPa< qc<800kPa
12 Zhang-ming Li’s Research group, (2011)
Su  0.042PS  3.74 50kPa< qc<600kPa
* Relations No. 1-10 are from Lin (1994), No.11 and No.12 are from Robertson & Campanella (1988) and Li (2011).
**σc= Gravity stress at the probe; Nc = the empirical coefficient of bearing capacity，Nc＝9~19
*** Nc= cone-tip resistance coefficient; σvo= the overlying pressure

The empiric relation in Tab.2-1 can be summed up in a generally linear formula:

qc  A1Su  B1 , or Ps  A2 Su  B2

Where the statistical fitting coefficient A1 and A2 are positive constant, while B1 and B2 could be positive
or negative. These coefficients are regarded as constant in most studies except that by Robertson and
Campanella (1983) the coefficients were thought to be related to the depth of the measuring point. The
two coefficients of the empiric expressions, in short, generally only the fitting coefficient, have no clear
physical-mechanics connotation and cannot be determined by the existing mechanical parameters.

2.2 Formula establishment based on the plastic theory of spatial axil-symmetrical problem
The following assumptions were made according to the properties of saturated soft clay and process
characteristics of CPT: (1) The cone penetration process can be considered as an elastic-plastic problem
in the process of cavity expansion; (2) Use Coulomb strength criterion and assume the value of internal
friction angle of saturated soft clay in undrained CPT process,   0 ; (3) In the cavity expansion process,
elastic volume change of the plastic-zone soil is relatively small and can be ignored.

394
 Then the radial expansion pressure Pu of the cavity wall (i.e. lateral surface of probe), after a cylindri-
cal cavity expansion, can be obtained upon the above three assumptions (Gong 1999).
 
Pu   xR n  c ln I X  1  c ln
E
 1
 21   c 
E
Ix 
21   c
Where  rRu = the stress at the inside radius of the cylinder after expansion; C = cohesive force; E = the
elastic modulus; and v = Poisson ratio.
Take the equilibrium condition of penetration probe:
D D 4L
( )2qc  DLcfs  ( )2  z ， namely qc  c fs   z
2 2 D
Where qc= Tip resistance; fs = Side friction; Lc = Effective friction length of penetration probe, D =
Probe diameter.
Then, fs can be got from the radial expansion pressure (multiplied by the friction coefficient):
E
fs   rRu   c[ln  1] （2.1）
2(1   )c
Where  = the friction coefficient between the penetration probe and the surrounding soil.
The penetration problem is a spatial axisymmetric problem in mechanics. The side and bottom of
penetration probe (corresponding to the plane, r = Ru and z = h, in the cylindrical coordinates) could
considered as principal plane, and according to the balance equation of the axisymmetric problem, that
is
d z
 z
dz
Solve the equation and obtain the constant according to the boundary conditions, we got
 z  z 2 / 2  h 2 / 2
Then, the formula on qc can be obtained from the equilibrium conditions of penetration probe:
4L h 2 4L
qc   z  c fs    c c(ln I r  1) （2.2）
D 2 D
where  = natural gravity density of soils; h = the depth of penetration probe.
In Eq. (2.2), the right side is negative, which indicates that the acted directions of qc and σz are oppo-
site.
Both single-bridge (the cone together with the outer sleeve, which can only measure a parameter Ps,
used widely in China) and double-bridge (the cone separate with the friction sleeve, which can measure
two parameters, i.e. the tip resistance and side friction) penetration probe have the same penetration way,
then get the specific penetration resistance Ps
ps  qc  mfs （2.3）
Where m = the ratio of the effective friction sleeve surface area of single-bridge probe to the double-
bridge one. As to the single-bridge and double-bridge probes of the same bottom area, m is the ratio of
side effective contact length. For a standard probe with a bottom area of 10 cm2, m=57/179.
For saturated soft clay, the internal friction angle could be assumed,   0 . According to the Cou-
lomb strength criterion, get VST shear strength Su =  =c, put in the Eq. (2.1) and Eq. (2.2), then the
value of qc and fs could be calculated. These two equations describe the relations between parameters in
CPT and VST of saturated soft clay.
Sensitivity coefficient, St = Su / Su', in which Su is undrained shear strength of undisturbed soil, while
Su' is that of disturbed soil.

395
 Sensitivity coefficient of undisturbed soil can also be estimated by Friction ratio Rf of double-bridge.
Schmertmann (1978) came up with the relation, St=Ns/Rf , in which Rf= fs/qc×100%, then
St  N s / R f  N sqc /f s （2.4）
Ns, as a dimensionless coefficient, was obtained by comparing CPT results with laboratory results by
Robertson & Campanella (1988) and the average value is 6. Studies by Rad & Lunne (1986) showed
that the Ns value changed in the range of 5 ~ 10, average 7.5. Lunne et al (1997) argued that the value
depended on the mineral, OCR and other functions, and no unique value was suitable for all the clay.
The relationship between sensitivity St from the VST and the parameters of CPT was described in Eq.
（2.4）.
As to specific penetration resistance Ps, it can get directly from Eq. (2.3) upon qc and fs.
This far, four important parameters, including side friction fs, tip resistance qc, specific penetration re-
sistance Ps and sensitivity coefficient St can be calculated by Eqs.（2.1）~（2.4）respectively. The
three parameters fs, qc and Ps are the theoretical solutions; while St is a semi-empirical solution due to it
is related the semi-empirical relationship of Schmertmann. Note that these formulas are nonlinear in
terms of strength parameters, rather than linear which are expressed by general empirical formulas.
Also, making use of Eq. (2.1) and Eq. (2.2), the elastic modulus E can be calculated by side friction
fs and tip resistance qc, respectively:
fs D h
( 1) (qc  1)
c 4 cLc
E  2(1   )c  e and E  2(1   )c  e 2
（2.5）

Moreover, the difficulty in determining bearing capacity of deep soft soil foundation may be effec-
tively solved when using Eq. (2.5).

3 SITE CONDITIONS

A ground treatment site in the Pearl River delta region of China was used to collect in-situ data. The site
area is 186000 m2, the surface layer is saturated ultra-soft clay of 11 m thick, with an average moisture
content of 75.0% and average void ratio of 2.087.
As shown in Fig. 3-1, the engineering geological condition within the scope of soft ground treatment
was very poor. The site, once artificial-surrounded fishing ponds of different sizes, was disposed with
hydraulic reclamation. The strata distribution and physical indicators are shown in Table 3-1. The static
-dynamic drainage consolidation method (Li 2011) was utilized in improving this ultra-soft soil ground;
according to the site conditions, 3 to 4 times of point-tamping and one general-tamping were required.

Figure 3-1. In situ situation before the ground improvement

396
Table 3-1. Strata distributing in the test area

Soil name Thickness /m Soil description

Artificial hydraulic fill 0.0~5.5 Distribution is very uneven with, high water content
Average12.0m, plastic flow state (water content is greater than liquid lim-
muck 3.5~20.5 it). The water content is 45.8% ~ 114%, average 75.0%; void ratio of
1.517 ~ 2.992, average 2.087.

Alluvial source, plastic state (water content is greater than plastic

Silt clay 0.7~9.5
limit).

Eluvial source, brown, maroon-based, hard plastic, local plastic -

Sandy clay 1.0~12.7
like, decomposed granite
Completely decomposed Gray, maroon-based, with a hard core of soil column, easily disi n-
2.1~10.5
granite tegrating by water
Strongly decomposed
0.7~13.2 Purple, gray, soil cores were folder or chunky rock, soft rock
granite

4 TEST METHODS AND EQUIPMENTS

Before the test, according to geologic condition and working condition of the site, choose the penetra-
tion equipment in combination with engineering requirements for the test depth. Select thrust tonnage of
the penetration equipment and prepare a reaction system to ensure the thrust force.
Choose probes on the basis of the suggestion given by Yilmaz (1991), as shown in Table 4-1, and
choose different capability of probes according to soil conditions, seen in Table 4-2.
Table 4-1. Probe selection

Soil conditions qc (MPa) Probe capability (total capacity) (kN)

extra soft 5~12.5 10~25
soft ~ medium soft 12.5~25.0 25~50
Clay
medium soft ~ hard 25.0~50.0 50~100
Dense sand ＞75 ≥150

Table 4-2. Probe dimensions

Single-bridge
Projected Double-bridge probe
Probe probe
area of the Apex angle
diameter Effective friction Friction sleeve Friction sleeve
Cone α/°
D/ mm side wall length surface area length
A/ cm2
L1 / mm /cm2 Lc /mm
10 35.7 60 57 200 179

Monitoring instruments and precision:

(1) Instrument for monitoring point layout: Leica TC307 Total Station; precision﹤1/4 s;
(2) Single-bridge probe: DQ-10Y, measuring range 0~30KN, precision 10N;
(3) Double-bridge probe: SQ-10Y, measuring range 0~30KN, precision 10N;
(4) DY - 2000 multi-purpose digital tester: testing accuracy≦±0.5%±1 word.
Calibrations of the probes including strain gauge load cells and pressure transducers were carried out

397
at regular intervals according to relative order, all according to the relevant regulations, to verify and en-
sure the quality of the probes. Initial induce measure and sensibility of sensors was described in Table 4-
3.
Table 4-3. Initial induce measure and sensibility of sensors

Level I Level II Level III

Initial induce level
Maximum penetration Maximum penetration Maximum penetration
measure /kPa
resistance resistance resistance
2.5～5.0MPa 7.0～12.0MPa 12.0～20.0MPa
parameter
ps , qc 10~20 30~50 50~100
fs 0.1~0.2 0.3~0.5 0.5~1.0
u 2 5 10

Penetration locations were set up roughly-evenly throughout the site, and the penetration test of each
location was conducted three (3) times, that is, before tamping, after the 1st point-tamping and after
tamping, with a test depth of 8～11m. The monitoring points were set up with the use of Total Station,
to make sure the uniformity of these points before and after tamping.
VST were carried out at the same plane position of penetration points. The parameters of vane head:
H—100mm，D—50mm，thickness—3mm; vane shear probe: SB-1Y，measuring range 0~30kN, pre-
cision 1Nm. VST were conducted below the surface (the original ground face) with a depth of 2 m or 4
m, 6 m and 8 m.
Quite a few measuring points were difficult to test because of the extra-poor site condition before
tamping, therefore, test data after the 1st pass of point-tamping were utilized in this article.

5 ANALYSIS AND COMPARISON BETWEEN TEST AND THEORETICAL RESULTS

5.1 In-site test results

qc－h curve of CPT qc－h curve of CPT Su-h curve of VST

（Orifice No.5，elevation：+7.96m） （Orifice No.5，elevation：+7.96m） (Orifice No.12，elevation: +7.50m）
qc/MPa qc/MPa Su/kPa
0 2 4 6 8 10 12 14 16 0 0.2 0.4 0.6 0.8 1 1.2 0 10 20 30 40
0 5 3

2 6 4

4 5
7
h/m
h/m

6
h/m

6
8

before tamping 7
8
before tamping
the 2nd tamping
9
the 2nd tamping
after tamping
after tamping 8
10

10 before tamping the 1st tamping

12 after tamping

Figure 5-1(a). Relation curve Figure 5-1(b). Amplificatory curve Figure 5-2(a).Typical relation curve
between qc with h（Deepness） between qc with h in the soft soil of VST

398
 Su '-h curve of VST qc/MPa fs/kPa
(Orifice No.12，elevation: +7.50m） 0 0.4 0.8 1.2 1.6 0 2 4 6 8
Su '/kPa
3.5 3.5
0 3 6 9 12 15
3 4.0

4.5 4.5
4
5.0

5 5.5 5.5

h/m
h/m

h/m
6.0
6
6.5 6.5

7 7.0

7.5 7.5
8
8.0

8.5 8.5
before tamping the 1st tamping
Test value Theoretical value Test value Theoretical value
after tamping

Figure 5-2(b). Typical relation curve of VST Figure 5-3. Relation curve of qc Figure 5-4. Relation curve of fs

Typical CPT and VST curves are shown in Fig. 5-1 and Fig.5-2 separately.
Fig. 5-1 and Fig. 5-2 displayed the change in mechanical strength characteristics of this ultra-soft clay
before, during and after the improvement. It is clear that the improving effect was obvious in the upper
4m.

5.2 Comparison between test and theoretical value of penetration parameters

According to soil tests, the relevant parameters were: Su = 8.4kPa, E = 1.31MPa, ν = 0.35，γ =
18.5kN/m3, μ= 0.085; Geometric parameters of probes: Lc = 179mm, D = 35.7mm. Took Ns = 6, which
is suggested by Robertson, when calculating sensitivity coefficient St. Calculation depth was 0.5 ~ 8m.
Because of the poor permeability of clay, gravity stress was obtained by calculating water and earth
pressures together, and qc or fs was calculated once per 0.1m respectively.
Results of comparison between theoretical and test data were shown in Fig. 5-3. As shown in Fig. 5-3,
the relative laws of the test parameters of CPT were as follows:
(1) In process of penetrating into saturated soft soil, qc grew slowly with h, the slope of the calculated
curve was not a constant, but changed a little; fs kept essentially unchanged with the depth h.
(2) The critical depth was not obvious when the probe was penetrated in saturated soft soil. The cause
might be ultra-soft clay is different with other soils; on account of the small value of friction angle  in
saturated soft clay, shear strength and friction kept minor changes with the increase of depth.
The figure also indicated that the theoretical values were in good consistency with the test ones.

5.3 Comparison between measured and theoretical values of Su with other parameters
(1) On the relationship between qc with Su

399
 Test value Theoretical value Test value Theoretical value
600 14

500 12

10
400
qc/kPa

fs/kPa
8
300
6
200
4
100
2

0 0
5 10 15 20 25 30 35 40 45 5 10 15 20 25 30 35 40 45
Su/kPa Su/kPa

Figure 5-5. Relation curve between qc with Su Figure 5-6. Relation curve between fs with Su

Seen from Fig.5-5, the test data shows a positive correlation between qc and Su. However, the curve
deviates from linearity. The results also indicate that the value of elastic parameters have an effect on re-
sults from theoretical calculations. In general, it is showed the overall consistency of theoretical and test
value.
（2）On the relationship between fs with Su
From Fig. 5-6, the test data shows a general linear trend of fs ~ Su. The theoretical value has a better
consistency with the test value in a certain range (Su <40kPa) , indicating that the theoretical formula is
suitable for soft clay.
（3）On the relationship between Ps with Su

Test value Theoretical value Test value Theoretical value

600 4.0

500 3.5

3.0
400
Ps/kPa

2.5
300
St

2.0
200
1.5
100
1.0
0 0.5
5 10 15 20 25 30 35 40 45 0 5 10 15 20 25 30 35 40 45
Su/kPa Su/kPa
Figure 5-7. Relation curve between Ps with Su Figure 5-8. Relation curve between St with Su

Fig.5-7 shows that the Ps ~ Su curve is a combination of the above two curves, fs ~ Su and qc ~ Su, and
in the same way, the theoretical value has a better consistency with the test value.
(1) On the relationship between St with Su
Fig.5-8 shows that in theory, St is reduced with the increase of Su. However, the test data indicates the
presence of one kind of soil, whose St remains unchanged as soil shear strength increases. It is not hard
to see, semi-empirical formula (2.4) suggested by Schmertmann etc., only shows that St would decrease
with soil shear strength. The relation of St with Su would be further developed.
Seen from the above comparison between the test results and theoretical values, the two kinds of val-
ues have good consistency for saturated soft clay. If there is some deviation, the reason is the complex
mechanical properties of soft clay, including: (1) the theory assumptions are simplified, such as the soil
is assumed to be a simple elastic-plastic body; (2) the complex mechanical process in the penetration
process is not taken into account, like the generation of pore water pressure (tip resistance qc can be cor-
rected as pore pressure is included), etc.; (3) the variability of soils properties, like the effect of residual
shells in the soft soil formation, etc.
The theoretical formula built the relationship among these important mechanical quantities of in situ
tests, which not only makes for saving huge amounts of test time and cost for the project design, but also
provides a way for the further development of relative theory.
400
6 CONCLUSIONS

(1) The existed empirical relationship between the parameters of CPT and shear strength can be
summed up as a general linear formula, in which the constant coefficient could be positive or negative.
Physical-mechanical connotation of these coefficients is not clear and even conflicting.
(2) As a spatial-axisymmetric problem, with plasticity theory of cylindrical cavity expansion, theoret-
ical relationship between the parameters of CPT (including tip resistance qc, side friction fs and specific
penetration resistance Ps), the elastic parameters and undrained stength of soil were established. Mean-
while, the relationship between CPT parameters and shear strength value of vane test was also obtained.
(3) The above established theoretical relationship is nonlinear in terms of strength parameters, rather
than linear expressed by the general empirical formulae.
(4) Comparing results from theoretical calculation with in situ data of CPT and VST in an actual ul-
tra-soft soil treatment project in the Pearl River Delta, it indicated that the good consistency between
theoretical relations (fs ~ Su, qc ~ Su and Ps ~ Su) and the test data. And semi-empirical formula on St of
VST, only showed that sensitivity coefficient would decrease with soil shear strength, so this relation-
ship on sensitivity coefficient should be further developed.
(5) Seen from these derived theoretical relationships, the penetration parameters can be directly de-
termined by the conventional mechanical parameters (such as elastic parameters and undrained strength)
or vane shear parameters, and vice versa; Elasticity modulus E can also be calculated by tip resistance qc
or side friction fs. All these make for saving the high cost of testing and provide theoretical methods and
formulas to avoid some tests limited due to the site and/or other condition.

ACKNOWLEDGEMENTS

Thanks to Miss Zeng Wenxiu, the graduate of the author, who collates data and drawings for this paper.
This work was supported by National Natural Science Foundation of China (Project No. 51178122).

REFERENCES

Code for testing of building foundation.（DBJ 15-60-2008）. 2008. Beijing: China Architectural Press.
Chui, X.Z. & Ding, H. 2004. Approximate theoretical and experimental research progress on the head resistance of static cone
penetration. Advances in Mechanics 34(2):251-262.
Gong, X.N. 1999. Soil Plasticity. Hangzhou: Zhejiang University Press.
Lin, Z.Y. 1994. Test and monitoring handbook of geotechnical engineering. Shenyang: Liaoning Sci. & Tech. Press.
Li, Z.M. 2011. Soft soil foundation reinforcement with quality control. Beijing: China Building Industry Press.
Robertson, P. K. & Campanella, R.G. 1988. Guidelines for geotechnical design using CPT and CPTU. Vancouver: Universi-
ty of British Columbia.
Robertson, P.k. & Cabal, K.L. 2012. Guide to cone penetration testing. California: Gregg Drilling & Testing, Inc.
Yan, S.W. & Feng, X.W. & Hou, J.F. & Li, W. 2009. Deduction and application of strength parameters of soft clay by use of
vane strength. Chinese Journal of Geotechnical Engineering 31(12):1805-1810.
Yang, J.H. & Xiong, L. 2010. Analysis of correlation of situ-test results in the Yangtze Delta. Soil Engineering and Founda-
tion 24(1):79-80.

401
402