CBR Soil Paper PDF

PSZ 19:16 (Pind.
1/97)
Universiti Teknologi Malaysia
BORANG PENGESAHAN STATUS TESIS

CALIFORNIA BEARING RATIO CORRELATION WITH
SOIL INDEX PROPERTIES .
.
JUDUL :
SESI PENGAJIAN : 2005 / 2006
MAK WAI KIN
Saya
(HURUF BESAR)
mengaku membenarkan tesis (PSM/Sarjana/Doktor Falsafah)* ini disimpan di Perpustakaan
Universiti Teknologi Malaysia dengan syarat-syarat kegunaan seperti berikut:
1.
2.
Tesis adalah hakmilik Universiti Teknologi Malaysia.

Perpustakaan Universiti Teknologi Malaysia dibenarkan membuat salinan untuk tujuan
pengajian sahaja.
Perpustakaan dibenarkan membuat salinan tesis ini sebagai bahan pertukaran antara
institusi pengajian tinggi.
** Sila tanda ( )
3.
4.
SULIT
(Mengandungi maklumat yang berdarjah keselamatan atau

kepentingan Malaysia seperti yang termaktud di dalam AKTA
RAHSIA RASMI 1972)
TERHAD
(Mengandungi maklumat yang TERHAD yang telah ditentukan

oleh organisasi/badan di mana penyelidikan dijalankan)
TIDAK TERHAD
Disahkan oleh
(TANDATANGAN PENULIS)
(TANDATANGAN PENYELIA)
Alamat Tetap:
404 , Blok A1 , Seksyen 2 ,

Wangsa Maju ,
53300 Kuala Lumpur.
DR NURLY GOFAR
Nama Penyelia
Tarikh:
15 MAY 2002
CATATAN: *
**
Tarikh:
15 MAY 2002
Potong yang tidak berkenaan.

Jika tesis ini SULIT atau TERHAD, sila lampirkan surat daripada pihak berkuasa/organisasi
berkenaan dengan menyatakan sekali sebab dan tempoh tesis ini perlu dikelaskan sebagai
SULIT atau TERHAD.
Tesis dimaksudkan sebagai tesis bagi ijazah Doktor Falsafah dan Sarjana secara penyelidikan,
atau disertai bagi pengajian secara kerja kursus dan penyelidikan, atau Laporan Projek
Sarjana Muda (PSM).
PSZ 19:16 (Pind. 1/97)
I hereby declare that I have read this project report and in my opinion this report is
sufficient in terms of scope and quality for the award of the degree of Master of
Engineering (Civil-Geotechnics).
Signature
Name of Supervisor :
Dr. Nurly Gofar
Date
15 May 2006
CALIFORNIA BEARING RATIO CORRELATION WITH

SOIL INDEX PROPERTIES
MAK WAI KIN
A project report submitted in partial fulfillment of

the requirements for the award of the degree of
Master of Engineering (Civil Geotechnics)
Faculty of Civil Engineering

Universiti Teknologi Malaysia
MAY 2006
ii
I declare that this project report entitled California Bearing Ratio Correlation With
Soil Index Properties is the result of my own research except as cited in the
references. The report has not been accepted for any degree and is not concurrently
submitted in candidature of any other degree.
Signature
Name
MAK WAI KIN
Date
15 MAY 2006
iii
To my beloved parents and sisters
iv
ACKNOWLEDGEMENT
I would like to take this opportunity to express my sincere appreciation to all

people and organization that had contributed towards the preparation of this final
project.
Firstly, I wish to thank my supervisor, Dr. Nurly Gofar, for spending her
precious time to supervise my works. I would not forget her invaluable guidance and
advices throughout this project.
Secondly, I am thankful to my companys director and colleagues for their

support and understandings. Their very useful assistance while I am working allows
me to concentrate and complete the project within the specified time.
Last but not least, not to forget the full supports that has been given by my
parents during my study.
ABSTRACT
California Bearing Ratio (CBR) is a commonly used indirect method to

assess the stiffness modulus and shear strength of subgrade in pavement design
works, however; civil engineers always encounter difficulties in obtaining
representative CBR value for design of pavement. Over the years, many correlations
had been proposed by various researchers in which the soil index properties were
used to develop these correlations. A study was carried out to find the correlation
between CBR values with soil index properties that best suit the type of soils in
Malaysia. Analyses were carried out based on the published correlations and soil
data obtained from two highway project sites. Based on the results, it is observed
that the current published correlations are not suitable to be used in Malaysia. In
addition, no typical range could be found based on the soil index properties. A
correlation had been proposed in the study to predict the CBR values at top face of
the soil sample for fine-grained soil based on the soil data collated.
These
correlations were developed based on the maximum dry density and optimum
moisture content.
vi
ABSTRAK
Nisbah Galas California (CBR) merupakan satu kaedah tidak langsung untuk
mengukur modulus kekerasan and kekuatan rich tanah bagi kerja-kerja rekabentuk
jalan raya berturap, tetapi; jurutera awam sentiasa menghadapi masalah untuk
mendapatkan nilai CBR yang boleh digunakan untuk rekabentuk. Tahun-tahun yang
lepas, banyak pertalian telah dicadangkan oleh banyak penyelidik dimana ciri-ciri
indeks tanah telah digunakan untuk mendapatkan pertalian ini. Satu penyelidikan
telah dijalankan untuk mendapatkan pertalian antara nilai CBR dengan ciri-ciri
indeks tanah yang boleh digunakan untuk jenis tanah di Malaysia.
Analisis
berpandukan pertalian yang telah diterbitkan dan data tanah yang didapatkan dari dua
projek lebuhraya. Keputusan analisis menunjukkan pertalian yang telah diterbitkan
ini tidak sesuai digunakan di Malaysia. Tambahan lagi, tipikal had nilai CBR tidak
diperolehi berpandukan ciri-ciri indeks tanah. Satu pertalian baru telah dicadangkan
dalam penyelidikan ini untuk menganggar nilai CBR di muka atas sampel tanah
jelekit berpandukan data tanah yang dikumpul. Pertalian ini diterbitkan berpandukan
kepada ketumpatan kering maksimum dan kandungan lembapan optimum tanah.
vii
TABLE OF CONTENTS
CHAPTER
TITLE
PAGE
DECLARATION
ii
DEDICATION
iii
ACKNOWLEDGEMENTS
iv
ABSTRACT
ABSTRAK
vi
TABLE OF CONTENTS
vii
LIST OF TABLES
LIST OF FIGURES
xi
LIST OF SYMBOLS
xiv
LIST OF APPENDICES
xv
INTRODUCTION
1.1
Background
1.2
Problem Statement
1.3
Aim and Objectives of Study
1.4
Scope of Study
LITERATURE REVIEW
2.1
California Bearing Ratio
2.1.1
Applications of California Bearing Ratio
2.1.2
Test Methods
2.1.2.1
In Situ Field Testing
2.1.2.2
Laboratory Testing
2.2
Soil Classification
11
2.2.1
Grain Size Distribution
12
2.2.2
Plasticity
15
viii
2.3
2.4
3
17
2.3.1
Design Manual for Roads and Bridges (1994)
17
2.3.2
Black (1962)
19
2.3.3
de Graft - Johnson and Bhatia (1969)
20
2.3.4
Agarwal and Ghanekar (1970)
21
2.3.5
National Cooperative
Program (2001)
Highway
Research
Current Practice in Malaysia
22
23
METHODOLOGY
26
3.1 Introduction
26
3.2
Data Collection
28
3.2.1
Source of Data
28
3.2.2
Data Selection
29
3.3
4
Correlations between CBR and Soil Classification
Data Analysis
30
RESULTS AND DISCUSSIONS
32
4.1 Introduction
32
4.2 Particle Size Distribution
32
4.3
Relationship of CBR at Top Face and Bottom Face
35
4.4
Evaluation of Published Correlations
36
4.4.1
Coarse-grained Soil
37
4.4.2
Fine-grained Soil
39
4.4.2.1
NCHRPs Correlation
39
4.4.2.2
Agarwal and Ghanekars Correlation
41
4.5
Typical Range of CBR Values
43
4.5.1
Coarse-grained Soil
44
4.5.2
Fine-grained Soil
45
4.6 Relationship of Maximum Dry Density with Optimum

Moisture Content
48
4.7
Proposed Correlation for CBR Values
50
4.8
Discussion
52
4.8.1
52
4.8.2
54
4.8.3
CBR Correlation with Soil Index Properties for

Malaysia Soils
55
ix
CONCLUSIONS AND RECOMMENDATIONS
57
5.1
Conclusion
57
5.5
Recommendations for Future Research
59
REFERENCES
APPENDIX A L
61
63 - 105
LIST OF TABLES
TABLE NO.
TITLE
PAGE
2.1
Definitions of soils classified by grading according to British

Soil Classification System
13
2.2
Relationship of plasticity with liquid limit
16
2.3
Subgrade CBR estimation of British soils compacted at

natural moisture content (The Highway Agency, 1994)
18
4.1
Particle size distribution test results for fine-grained soils
33
4.2
Particle size distribution test results for coarse-grained soils
34
xi
LIST OF FIGURES
FIGURE NO.
TITLE
PAGE
2.1
Dynamic cone penetrometer equipment
2.2
Test equipment for determination of CBR value in

laboratory
11
2.3
Example of grading curves
14
2.4
Plasticity chart
16
2.5
Relationship between CBR and plasticity index at

various liquidity index values
19
2.6
Correction of CBR values for partial saturation
19
2.7
Relationship between suitability index and soaked CBR

values
20
2.8
Relationship between the ratio of maximum dry density

to plasticity index and CBR for laterite-quartz gravels
21
3.1
Flowchart of the study
27
4.1
Relationship between CBRTOP and CBRBOTTOM values
35
4.2
Comparison of CBRTOP with NCHRPs line for coarsegrained soil
37
xii
4.3
Comparison of CBRBOTTOM with NCHRPs line for

coarse-grained soil
38
4.4
Comparison of CBRTOP with NCHRPs line for finegrained soil
40
4.5
Comparison of CBRBOTTOM with NCHRPs line for finegrained soil
40
4.6
Relationship between CBRTOP and CBRA&G values
42
4.7
Relationship between CBRBOTTOM and CBRA&G values
42
4.8
Numbers of measurement of CBRTOP for coarse-grained

soil
44
4.9
Numbers of measurement of CBRBOTTOM for coarsegrained soil
45
4.10
Numbers of measurement of CBRTOP for fine-grained

soil
46
4.11
Numbers of measurement of CBRTOP(3%) for finegrained soil
47
4.12
Numbers of measurement of CBRBOTTOM for fine-grained

soil
47
4.13
Numbers of measurement of CBRBOTTOM(3%) for finegrained soil
48
4.14
Relationship of maximum dry density with optimum

moisture content
49
4.15
Proposed correlations for CBRTOP for fine-grained soil
51
xiii
LIST OF SYMBOLS
Percentage passing 2.4 mm BS sieve
CBR
CBRA&G
CBR value predicted by the Agarwal and Ghanekars

correlation
CBRTOP
CBR value at top face of soil sample
CBRTOP(3%)
Minimum CBRTOP within the range of 3% of OMC
CBRBOTTOM
CBR value at bottom face of soil sample
CBRBOTTOM(3%)
Minimum CBRBOTTOM within the range of 3% of OMC
DCP
Dynamic Cone Penetrometer
D60
Diameter at 60% passing from grain size distribution (mm)
LL
Liquid Limit
MDD
Maximum Dry Density
OMC
Optimum Moisture Content
PI
Plasticity Index
Percentage passing No.200 U.S. sieve (in decimal)
xiv
LIST OF APPENDICES
APPENDIX
TITLE
PAGE
Atterberg limits test results for coarse-grained soils
63
Atterberg limits test results for fine-grained soils
64
Compaction test results for coarse-grained soils
66
Compaction test results for fine-grained soils
67
Measured laboratory CBR values for coarse-grained

soils
68
Measured laboratory CBR values for fine-grained

soils
70
Measured soil index properties required for

NCHRPs correlations
72
Estimated CBR values from NCHRPs correlation for

coarse-grained soils
74
Estimated CBR values from NCHRPs correlation for

fine-grained soils
75
Estimated CBR values based on Agarwal &

Ghanekars Correlation
77
Determination of the CBR value extracted from

BS1377 Part 4:1990
78
xv
Determination of the CBR value extracted from

ASTM D 1883 - 92
98
CHAPTER 1
INTRODUCTION
1.1
Background
California Bearing Ratio (CBR) is frequently used index test value for civil
engineer particularly those in pavement construction to assess the stiffness modulus
and shear strength of subgrade. It is actually an indirect measure which represents
comparison of the strength of subgrade material to the strength of standard crushed
rock quoted in percentage values.
The method was originally developed at
California Division of Highways in 1930s to provide an assessment of the relative

stability of fine crushed rock base material.
California Bearing Ratio is not something new to civil engineers in Malaysia

especially for those involved in road and airport pavement works. Usually, the CBR
values are used by pavement engineers to design the thickness of pavement that will
be laid on top of the subgrade. Subgrade that has lower CBR value will have thicker
pavement compared with the subgrade that has higher CBR value. In other words,
the design of pavement is very much dependent on the CBR value of subgrade.
Different soil types give different values of CBR although it is compacted at the
same amount of energy and rate of penetration.
Conventionally, CBR value can be measured directly in the laboratory test in

accordance with BS1377 on soil sample acquired from site. The soil sample will be
compacted as required in a standard mould and then a plunger is made to penetrate
the soil at a specified penetration rate. Load deflection curve plotted from the
2
result of the penetration will be compared with that obtained from the standard crush
rock.
Apart from CBR test carried out in laboratory, engineer frequently conducts
indirect measurement of CBR value at project site. Dynamic Cone Penetrometer
(DCP) is a popular in-situ test method commonly used to estimate the in-situ CBR
value. However, the CBR value obtained from DCP test shall not be relied upon for
pavement design as it may represent unsoaked CBR value rather than soaked CBR
value which is required for design. Therefore, engineer is advised not to use the
CBR value obtained from DCP test for pavement design but only as a comparison
and estimation of CBR values that can be achieved by the subgrade.
DCP test although does not give exact soaked CBR value for design, it is
always proposed by engineers for subgrade assessment because it is an easy, cheap
and fast method compared with laboratory test. While laboratory test takes at least
four (4) days to measure the CBR value for each soil sample, DCP tests can give
immediate results of CBR values at various locations just in one day. Nevertheless,
it is still a good engineering practice that DCP test is being carried in a project as a
supplement to laboratory testing when assessing the shear strength and stiffness
modulus of subgrade.
A more reliable method of predicting CBR value of subgrade shall be

explored so that the engineers will have more options and confidence in obtaining a
representative soaked CBR value for pavement design.
One of the methods is by developing a correlation between CBR values with

soil index properties. There are few correlations that have been published by many
researchers since 1960s.
In Malaysia, practising engineers seldom use these
correlations as it may be due to its unproven results on the Malaysia soils. Although
there are some researches had been carried out by our local universities, no extensive
data have been collated from a number of projects in Malaysia for verification
purposes.
3
1.2
Problem Statement
Civil engineers always encounter difficulties in obtaining representative

CBR value for design of pavement. Inadequate soil investigation data due to budget
constraint and poor planning of soil investigation works are regularly happened here
in Malaysia. In addition, laboratory CBR test required a relatively large soil sample
and is time consuming. Furthermore, the results sometimes are not accurate due to
the poor quality of handling and laboratory testing on the soil samples.
Thus,
identification of factors that governs the CBR value such as index properties and
classification of the soil can be used as a base of the judgement on the validity of the
CBR values obtained in the field.
1.3
Aim and Objectives of Study
The aim of the study is to find correlation between CBR values with soil
index properties that best suit the type of soils in Malaysia. In order to achieve this
aim, three objectives have been identified for the study:
1.
To evaluate published correlation for CBR value and the index properties of
soil based on collated data acquired from a number of projects in Malaysia.
2.
To tabulate the CBR values obtained from collated soil samples and propose
a typical range of CBR values samples based on the soil index properties.
3.
To obtain a correlation between CBR values with soil index properties that is
best suited for the type of soils in Malaysia.
4
1.4
Scope of Study
The study covers only the Malaysian practices in predicting CBR values for
pavement design. Site and laboratory tests will not be carried out thus all the soil
information and test results will be obtained from soil investigation contractors and
commercial laboratories.
The correlations to be reviewed and analysed in this study will be limited to

published correlations of CBR values with soil index properties that are generally
acceptable by engineers worldwide.
CHAPTER 2
LITERATURE REVIEW
2.1
The California Bearing Ratio (CBR), defined as the ratio of the resistance to
penetration of a material to the penetration resistance of a standard crushed stone
base material. The ratio has been used as an empirical measure of the road subgrade
strength. The method was developed by the California Division of Highways in
1930s as part of their study in pavement failure.
The CBR ratio was originally established to provide an assessment of the

relative stability of fine crushed rock base material. Then the use of CBR test was
extended to assessment of subgrade material. It is primarily intended for, but not
limited to, evaluating the strength of cohesive materials having maximum particle
sizes less than 19 mm (0.75 in.). Since it establishment in 1930s, it has been widely
used by many countries and considerable experience has been developed ever since
then. Ironically, the method was for pavement design in California for only a few
years, and was superseded by the Hveen Stabilometer test (Carter and Bentley,
1991).
The method of evaluating CBR standardized in ASSHTO T193 and

BS1377:1990. The test is an index test, thus it is not a direct measure of stiffness
modulus or shear strength. The ratio provides a means of comparison between the
strength of subgrade and the standard crushed stone base kept in California Division
of Highways laboratory. CBR test is frequently used as an index test to evaluate the
6
strength of road subgrade as direct determination of stiffness modulus and shear
strength is difficult. This strength value is often used as a guide to the design of road
pavement thickness or to assess compliance of subgrade against minimum
specification values set by the design engineer.
2.1.1
Applications of California Bearing Ratio
The main application of California Bearing Ratio (CBR) is to evaluate the

stiffness modulus and shear strength of subgrade.
Generally, the subgrade soil
cannot bear the construction and commercial traffic without any distress, therefore; a
layer of rigid or flexible pavement is required to be laid on top of the subgrade to
carry the traffic load.
The determination of the thickness of the pavement layer is governed by the

strength of subgrade, thus the information on the stiffness modulus and shear
strength of subgrade are required before any pavement design is carried out. These
parameters are necessary to determine the thickness of the overlying pavement in
order to achieve optimum and economic design. The stiffness modulus and shear
strength of subgrade are controlled by soil type, particularly plasticity, degree of
remoulding, density and effective stress (The Highway Agency, 1994). The effective
stress is dependent on the stress from the overlying soil layers, the stress history and
the suction. In turn, suction is dependent on the moisture content history, the soil
type and the depth of the water table.
Due to the number of factors that make the measurement of stiffness modulus
and shear strength of subgrade complicated, it is necessary to adopt a more
simplified test method that can be used as an index test. This is where CBR test
come into frame in measurement of subgrade strength. The CBR test is a simple
strength test that compares the bearing capacity of a material with that of a well
graded standard crushed stone base material. This means that the standard crushed
stone material should have a CBR value of 100%. The resistance of the crushed
7
stone under standardised conditions is well established. Therefore, the purpose of a
CBR test is to determine the relative resistance of the subgrade material under the
same conditions.
If the CBR value of subgrade is high, it means that the subgrade is strong.
Accordingly, the design of pavement thickness can be reduced in conjunction with
the stronger subgrade. Thus, it will give a considerable cost saving in term of
construction besides an optimum design. However, if the CBR value of subgrade
indicates that the subgrade is weak i.e. low reading of CBR reading, the thickness of
pavement shall be increased in order to spread the traffic load over a greater area of
the weak subgrade. This is important to prevent the weak subgrade material to
deform excessively and causing the road pavement fail.
Alternatively, the easiest method to overcome this weak subgrade before the
construction of pavement is by replacing the soil with adequately compacted soil in
layers. Otherwise, the subgrade can be stabilised by lime, cement, or the use of a
geotextile to produce a stable platform for construction equipment and traffic load in
long term.
The CBR test is used exclusively in conjunction with pavement design
methods and the method of sample preparation and testing must relate to the
assumptions made in the design method as well as to assumed site conditions. For
instance, the design may assume that soaked CBR values are always used, regardless
of actual site conditions (Carter and Bentley, 1991).
2.1.2
Test Methods
California Bearing Ratio (CBR) values can be measured by in situ field

testing and laboratory testing. The tests are based on the principle of the testing is to
make a plunger of standard area penetrating into a soil sample. The force or load
required to cause the penetration will be recorded. This information is plotted on a
8
standard graph, and the plot of the test data will establish the CBR result of the soil
tested.
2.1.2.1 In Situ Field Testing
California Bearing Ratio (CBR) test can be carried out in situ if requested by
engineer. It will cover the determination of the CBR value of a soil tested in situ,
with a selected overburden pressure, by causing a cylindrical plunger to penetrate the
soil at a given rate. The force required to cause the plunger to penetrate the in situ
soil together with the penetration depth would be recorded at a specified interval.
Later, the results shall be compared with the relationship between force and
penetration into the in situ soil to that for a standard crushed stone base material.
The field CBR testing is seldom used in Malaysia. Instead, a more popular
test method known as dynamic cone penetrometer test or commonly referred as
Dynamic Cone Penetration (DCP) test is widely used due to its simplicity. The
method can be carried out with minimum of three people with a capacity of six tests
per day.
The DCP was originally designed and used for determination of strength
profile of flexible pavement structures. It consists of a long steel rod fitted with 60
degrees hardened cone with 20mm larger diameter at the penetrating end. The length
is about two meters and can be extended by extension rods if required to penetrate
deeper into subgrade.
The DCP is driven into soil by dropping an 8 kg sliding hammer located at

the upper rod through a height of 575mm. This weight is able to fall freely through a
given drop height to achieve a standard amount of penetrative effort at every drop.
The depth of penetration is then measured using the ruler attached to the rod.
9
Handle
8 kg Hammer
575 mm
Hand
Guard
Hardened 60o Cone
Figure 2.1 : Dynamic cone penetrometer (DCP) equipment
By measuring the penetration of the cone against number of drops of the

weight it is possible to plot resistance to penetration which means the strength of the
subgrade being tested can be measure indirectly. It is possible to correlate the DCP
readings with CBR values by applying a simple formula.
The recommended
equation by Steve, Richard and Thomas (1992) is given as:
CBR =
292
DCP 1.12
(2.1)
2.1.2.2 Laboratory Testing
The laboratory testing for determination of California Bearing Ratio (CBR)

values is almost similar to the in situ CBR field testing as described above, except
the latter soil sample is undisturbed.
10
The test follows a well published standard procedure and there is little
difference between the BS1377 (1990) Part 4 under Clause 7 and AASTHO T193.
The main difference between the two standards lied on the practice during
preparation of compacted samples for testing.
The AASTHO specification stipulates the use of dynamic compaction by

using a rammer to compact the soil sample for testing. On the other hand, British
Standard specification allows the use of static compaction by using a load frame to
compact the soil sample. In addition, dynamic compaction using either a rammer or
a vibrating hammer is also acceptable by British Standard.
In Malaysia, BS 1377 (1990) Part 4 standard procedure is commonly adopted

for determination of CBR values in laboratory. Generally, it is being carried out on
remoulded soil samples obtained from site in bulk. The soil sample is placed into a
mould and compacted to the efforts as stipulated in the standard.
The sample is the compacted to the expected dry density at the appropriate
water content in a mould. It can be compacted at natural moisture content for
determination of CBR value at in situ condition or else compacted to optimum
moisture content for determination of CBR value at maximum dry density. The soil
sample is frequently soaked for 4 days before testing.
A cylindrical plunger is made to penetrate the soil sample at a given rate.

The amount of seating force to be applied on the plunger depends on the expected
CBR value from the soil. It is recommended to apply 10N for expected CBR value
up to 5%, 50N for expected CBR value from 5% to 30% and 250N for CBR value
above 30%.
Surcharge weight in the form of annular discs with a mass of 2kg can be
placed on the soil sample to simulate the weight of pavement materials overlying
subgrade. The plunger is then penetrated into the soil at a constant rate of 1mm/min
and the forces recorded at penetration intervals of 0.25mm and the total penetration
11
should not exceed 7.5mm. Figure 2.2 shows typical test equipment for determination
of CBR value in laboratory.
The recorded results of force applied and penetration are plotted in the form
of a load-penetration diagram by drawing a curve through the experimental points.
These results are then compared to a standard curve from standard crushed stone
base material where the forces on the standard curve are 13.24kN at 2.5mm
penetration and 19.96kN at 5.0mm penetration. The CBR value of the soil is then a
simple ratio of the corresponding values. The higher CBR values shall be taken.
Figure 2.2 : Test equipment for determination of CBR value in laboratory
2.2 Soil Classification
The soil classification systems are used extensively by engineers to

distinguish between the different types of soils within broad categories.
Some
classification systems are available at present i.e. Unified Soil Classification System
and British Soil Classification System. Most classification system divided the soil
into two groups: cohesive or fine-grained soils and cohesionless or coarse-grained
soils.
12
Some tests are required to evaluate the probable behavior of soil.
For
cohesionless soils the density and grain size distribution or grading are most
indicative of its behaviour. On the other hand, plasticity gives a better appraisal of
the behaviour of cohesive soils. The concept of grading and plasticity and the use of
these properties to identify, classify and assess soils are the oldest and most
fundamental in soil mechanics (Carter and Bentley, 1991). These properties will be
the main focus throughout this study.
2.2.1
Grain Size Distribution
Grain size distribution is known as a major factor in determining the

properties of soils, mainly coarse-grained or cohesionless soils. This is particularly
important where mineral composition is relatively insignificant.
The grain size
distribution provides the engineer with valuable information on the soils

characteristics. Grain size tests are conducted not only to determine the size of the
individual grains in soil, but also to determine the relative distribution of the sizes.
For cohesionless or coarse-grained soils, the grain size distribution is

determined by passing soil samples either by wet or dry shaken through a nest of
sieves of descending size. Generally, it is a screening process in which coarse
fractions of soil are separated by means of series of sieves.
On the other hand, the grain size of fine-grained cohesive soils must be
determined by more complicated methods. One such test is the hydrometer test
which involves the measurement of the specific gravity of a soil-water suspension at
fixed time intervals. Sizes are determined from the settling velocity and times
recorded using Stokes law.
In Malaysia, the British Soil Classification System (BSCS) as stipulated in

BS 5930 (1999) has been adopted by engineers in classifying the soils. The soils are
classified into groups according to size, and the groups are then further divided into
13
coarse, medium and fine sub-groups. Table 2.1 below shows the definition of soils
classified by grading according to BSCS.
Table 2.1 : Definitions of soils classified by grading according to British Soil

Classification System
TYPES OF SOIL
SIZE
(mm)
Very Coarse
Boulders
> 200
Soils
Cobbles
60 200
Coarse Soils
Gravel
Sand
Fine Soils
Silt
Clay
coarse
20 60
medium
6 20
fine
26
coarse
0.6 2.0
medium
0.2 0.6
fine
0.06 0.2
coarse
0.02 0.06
medium
0.006 0.002
fine
0.002 0.006
< 0.002
The grain size distribution is summarised as percentages of clay, silt, sand

and gravel which is determined by weighing on cumulative quantities in certain sieve
sizes. Points are then plotted of percentage passing on a semi-logarithmetric chart.
A smooth S-shaped curve drawn through these points is called a grading curve.
The position and shape of the grading curve can determine whether the soil is
well graded, poorly graded or uniformly soil. Figure 2.3 shows the example of five
types of typical grading curves. Grading curve type A illustrates the soil is well
graded material, type B is poorly graded material, while type C shows the soil is a
uniform cohesionless material. Meanwhile, type D and type E are well graded with
some clay and well graded with excess of fine-grained materials.
14
A uniform soil is one that consists of particles with a very narrow range in
sizes. This soil is likely to be in loose state. A well graded soil, on the other hand,
tends to be in dense state and can be compacted even more with mechanical
equipment.
Soil that has a very small range of grain sizes is considered poorly graded
soil. This type of soil contains a higher proportion of voids than well-graded soils in
which the finer grains will fill the voids between the coarser grains. Thus, grading
influences the density of soils.
100
E
80
% Finer
60
C
40
D
20
0
0.0001
0.001
0.01
0.1
10
100
Particle size (mm)

Figure 2.3 : Example of grading curves
It is important to classify the type of soil based on its particle size distribution
prior in carrying out any analysis on the soil data. Based on the particle size
distribution test results, the soil data can be grouped into two (2) broad types, i.e.
coarse-grained soil or fine-grained soil.
According to BS 5950 (1999), fine-grained soil is soil that contains silt and
clay content more than 35% from the total weight of soil. Soil that contains gravel
and sand content more than 65% will be classified as coarse-grained soil. Both types
15
of soil have different characteristics and reacts in a different way when subjected to
stress and strain
2.2.2
Plasticity
Plasticity is the ability of a material to be moulded without fracturing. In soil,

it is due to the electrochemical behaviour of clay minerals and is unique to soils
containing clay-mineral particles. Clay soils have flaky particles to which water
adheres, thus imparting the property of plasticity. Natural water content is also of
vital interest. It is important to determine the plasticity characteristic of a cohesive
soil.
Cohesive soil can exist is three states depending on the amount of water
present in the soil. The three states are liquid slurry, a plastic substance or a solid. In
order to distinguishing them, a test called Atterberg limits was developed. The test
procedure details can be found in BS 1377 (1990).
When water is added to a dry cohesive soil, it changes from solid to semisolid to plastic to liquid state. The moisture content in the soil at the threshold
between semi-solid and plastic is called the plastic limit whereas the moisture
content in the soil at the threshold between plastic and liquid is called the liquid
limit. The numerical difference between the liquid limit and plastic limit is defined
as the plasticity index of the soil. The greater the plasticity index means that the soil
is more plastic and compressible and the greater the volume change characteristic of
the soil. It has proven to be one of the most helpful of all soil indices and is essential
to the description of a cohesive soil.
It has been observed that many properties of clays and silts, such as their dry
strength, their compressibility, their reaction to the shaking test, and their consistency
near the plastic limit, can be correlated with the Atterberg limits by means of the
plasticity chart (Terzaghi, Peck, and Mesri, G., 1996). The plasticity chart as shown
16
in Figure 2.4 has an empirical boundary known as the A line which separates the
inorganic clays form silty and organic soils.
Figure 2.4 : Plasticity chart
The relation of the natural moisture content to the liquid and plastic limits is
indicative of soil behaviour. When the natural moisture content is above or close to
the liquid limit, the soil may be sensitive. A sensitive soil wills loss its shear strength
easily when it is disturbed. This sensitivity complicates sampling and testing and
special measures frequently have to be used.
Table 2.2 : Relationship of plasticity with liquid limit

PLASTICITY
LIQUID LIMIT
(%)
Low
< 35
Intermediate
35 50
High
50 70
Very High
70 90
Extremely High
> 90
17
Based on the British Soil Classification System, fine-grained soil can be
divided into five classes of plasticity. This plasticity classes are dependent on liquid
limit which is as shown in Table 2.2.
2.3
Correlations between CBR and Soil Classification
It is reasonable to assume that California Bearing Ratio (CBR) values are

related to soil index properties in some ways. Many researches had been carried out
to show the effect of soil types and characteristics on CBR values such as Black
(1962), de Graft-Johnson & Bhatia (1969), Agarwal & Ghanekar (1970) and NCHRP
(2001).
2.3.1 Design Manual for Roads and Bridges (1994)
The Highway Agency (1994) had proposed an estimation of CBR values

based on plasticity index for British soils compacted at natural moisture content
which is shown in Table 2.3. The precise density and moisture content conditions
corresponding to the predicted CBR values are not specified. As such, this table
shall be limited for the use of British soils only. Furthermore, the predicted CBR
values as stated in this table do not take into the account of water table depth below
subgrade formation level.
As shown in the table, it can be observed that the soil types play the most
important role in determination of CBR values. Predicted CBR values of 20% to
60% can be obtained from coarse-grained soil whereas for fine-grained soils CBR
values in the range of 2% to 5% are expected. Therefore, cohesionless soils can
provide higher CBR values compared with cohesive soils.
18
Table 2.3 : Subgrade CBR estimation of British soils compacted at natural moisture
content (The Highway Agency, 1994)
TYPE OF SOIL
PLASTICITY
PREDICTED
INDEX (%)
CBR (%)
70
60
50
40
2 to 3
30
3 to 4
20
4 to 5
10
4 to 5
Heavy Clay
Silty Clay
Sandy Clay
Sand (Poorly Graded)
20
Sand (Well Graded)
40
Sandy Gravel (Well Graded)
60
Soil grading characteristics is also one of the factors that affecting the CBR
values. Poorly graded sand shows predicted CBR of 20% while well graded sand
give CBR value as high as 40%. This indirectly confirms that the relative density of
soils is essential in determination of CBR values.
It is worth to note that plasticity index has impact on the predicted CBR
values as shown in the Table 2.3. The predicted CBR values of 2% to 5% were
obtained from soil with plasticity index ranging from 10% to 70%. Hence, it shows
that CBR values can be correlated with soil plasticity.
CBR values depend not only on soil index properties but also on the density,
moisture content, and to some extent, method of sample preparation during
laboratory testing.
These factors must therefore be taken into account when
considering correlations between CBR and soil classification tests.
19
2.3.2
Black (1962)
A number of attempts have been made to correlate California Bearing Ratio

(CBR) with soil grain distribution and plasticity.
Among them, Black (1962) had
developed a method of estimating the CBR value for cohesive soils. He had obtained
the correlations between CBR and plasticity index for various values of liquidity
index which is shown in Figure 2.5.
Note that the values given in Figure 2.5 are
only referred to saturated soil only. For unsaturated soils, the CBR values obtained
can be corrected by applying the correction factor as shown in Figure 2.6.
Figure 2.5 : Relationship between CBR and plasticity index at various liquidity
index values
Figure 2.6 : Correction of CBR values for partial saturation
20
2.3.3
de Graft Johnson and Bhatia (1969)
A correlation of CBR with plasticity and grading using the concept of

suitably index was developed by de Graft-Johnson and Bhatia (1969).
The
correlation is shown as below and Figure 2.7. In this case, the suitability index is
defined as:
Suitability Index =
where;
A
LL log(PI )
Percentage passing 2.4mm BS sieve
LL
Liquid Limit
PI
Plasticity Index
(2.2)
It is worth to note that the soil samples were compacted to maximum dry
density at optimum moisture content and soaked for 4 days according to the Ghana
standard of compaction. This specifies the use of a standard CBR mould and a 4.5kg
rammer with 450mm drop height to compact the soil in 5 layers using 25 blows per
layer.
Figure 2.7 : Relationship between suitability index and soaked CBR values
21
2.3.4
Agarwal and Ghanekar (1970)
Based on their research on 48 samples of fine-grained soils found in India,

Agarwal and Ghanekar (1970) had tried to develop a correlation between CBR
values and either liquid limit, plastic limit or plasticity index. However, there failed
to found any significant correlation between them.
Instead, they did found an
improved correlation when they include the optimum moisture content and liquid
limit. The correlation is defined as below.
CBR = 2 16 log(OMC ) + 0.07 LL
where;
OMC =
LL
Liquid Limit
(2.3)
The 48 soil samples tested by them had CBR values of not more than 9% and
the standard deviation obtained was 1.8. Hence, they suggested that the correlation
is only of sufficient accuracy for preliminary identification of material. They also
recommended that this correlation may be of more use of derived for specific
geological regions.
Figure 2.8 : Relationship between the ratio of maximum dry density to plasticity
index and CBR for laterite-quartz gravels
22
2.3.5
National Cooperative Highway Research Program (2001)
National Cooperative Highway Research Program (2001) of United States of

America through the Guide for Mechanical-Empirical Design of New and
Rehabilitated Pavement Structures had developed some correlations that describe
the relationship between soil index properties and CBR values. Two equations were
established where one was for non-plastic coarse-grained material and the other was
for soils which contains 12% fines and exhibit some plasticity.
The best-fitted equation proposed by NCHRP for clean, coarse-grained soil is

shown as below.
CBR = 28.09(D60 )
0.358
where;
D60
(2.4)
Diameter at 60% passing from grain size distribution

(Unit in millimeter)
Equation above is limited to D60 values greater than 0.01mm and less than
30mm. For D60 less than 0.01mm, the recommended value of CBR is 5% whereas
CBR value of 95% is recommended for D60 greater than 30mm.
For plastic, fine-grained soils, the soil index properties chosen to correlate
CBR are the percentage passing No.200 U.S. sieve or 0.075mm size sieve and
plasticity index. The suggested equation by NCHRP is shown below.
CBR =
where;
75
1 + 0.728(wPI )
(2.5)
Percentage passing No.200 U.S. sieve (in decimal)
PI
Plasticity Index
23
2.4
Current Practice in Malaysia
The knowledge on the strength as indicated by California Bearing Ratio

(CBR) value of subgrade is crucial to pavement design works because the CBR value
has been used widely by practising engineers for the design of pavement thickness
overlying the subgrade. The CBR of subgrade can be obtained by in situ field
testing. The in situ method frequently used by practising engineers in Malaysia is
dynamic cone penetrometer test (DCP). Although the DCP test is not a direct
method of measuring the CBR values at site, however this method is more preferable
to in situ direct CBR testing.
DCP test can be carried out easily at site to check on subgrade resistance
against penetration. The recorded DCP blows or readings will be converted into
CBR value based on empirical formulation represented by Equation (2.1). Since the
works are carried out at site, the CBR value obtained is best representative of CBR
values under unsoaked condition subjected to ground water level.
As discussed earlier, the CBR value used for pavement design is under
soaked condition.
Therefore, the value obtained from the DCP test which is
unsoaked CBR value, cannot be used directly for the design work.
Practising
engineer usually used the unsoaked value to indicate the bearing capacity of the
subgrade is adequate or meet the design requirement if not subjected to soaked
condition.
As such, good drainage is essential for road construction to allow the

optimum CBR that value can be obtained from the subgrade. Drainage must be kept
operating efficiently during the life of the pavement to prevent the CBR value from
decreasing due to rising of water table.
Generally, minimum soaked CBR value of 5% of subgrade has always been

adopted for the new road construction in Malaysia. Any values lowered than this is
rarely being used except for pavement rehabilitation works. For the latter case, a
24
capping layer with fill material capable of achieving soaked CBR of 15% will be laid
on top of the poor subgrade.
Laboratory testing for determination of CBR value is usually carried out on

the fill material. The fill material will be tested to ensure that the soil will be capable
of achieving required soaked CBR value for road construction works. Prior to the
CBR testing, compaction test is usually conducted to obtain maximum dry density
and optimum moisture content.
Soil samples taken will be compacted in five moulds with each having the
different moisture content in the range of -3% and +5% of the determined optimum
moisture content. The samples shall be soaked for four days in a water tank. After
four days, measurement of CBR values as per requirement of BS 1377 will be carried
out on the samples to obtain the soaked CBR.
The minimum CBR value is derived form the CBR versus moisture content
plot over a range of 3% of the optimum moisture content. If the soaked CBR value
is less than the minimum requirement value, the fill material has to be rejected and
replaced with higher strength material. Although the fill material does not meet the
minimum soaked CBR requirement, it will still be used as fill material below some
depth from the base of pavement which is subjected to approval from engineer.
Occasionally, the in situ CBR values of subgrade can be obtained from

laboratory testing. The in situ density and moisture content of the subgrade has to be
identified first. Then, the sample will be compacted in a mould at its in situ moisture
content. Sample is soaked for four days and subsequently CBR measurement will be
taken on the sample. The CBR value obtained is best representative of the in situ
CBR of subgrade under soaked condition.
The above method is usually being carried out for pavement rehabilitation
works. The subgrade under the pavement needs to be tested to evaluate whether the
underlying subgrade is capable of achieving the minimum soaked CBR value for
25
design. If the existing subgrade is poor, either subgrade reconstruction or a capping
layer overlying the subgrade can be adopted.
At present, practicing engineers in Malaysia seldom use the correlations of

CBR value with soil index properties as described in Section 2.3 above. One of the
main reason is the correlation may not representing the types of soils in Malaysia.
Moreover, the practice of samples preparation and testing procedures may as well
affect the accuracy of the correlations as most of the established correlations had
used some different methods.
The need for reliable correlation between CBR value and soil index
properties is required for the pavement design in Malaysia.
This is important
particularly for pavement rehabilitation works where existing subgrade underlying

the pavement needs to be assessed prior to the design and construction of pavement
works. DCP test could be carried out but it is not representing soaked CBR value
which is required for design works.
27
CHAPTER 3
METHODOLOGY
3.1
Introduction
The first objective of the study utilised the available correlation methods to
meet the objectives as stated in Chapter 1. Therefore, literature review was carried
out in the early stage of the study to enhance the understanding of the California
Bearing Ratio (CBR) and soil classification. Technical papers from the international
proceedings, journals and published reports were reviewed to keep up to date on the
published correlations that relate the CBR values with the soil index properties.
Summary of the literature review have been presented in the Chapter 2.
Data were collected from field and laboratory test results of the CBR values
and soil classification from a number of projects in Malaysia. These soil data shall
be obtained from the soil investigation contractors and commercial laboratories. The
suitability of published correlations in predicting the CBR values on collected data
were analysed in this study. New correlations are developed and analysed to fit the
collected data.
The overall methodology of the study has been summarised in a flowchart as

shown in Figure 3.1.
27
Literature
Study
Problem
Identification
Published
Correlation
Field Test
Results
Laboratory
Test Results
Collate
Data
CBR Values
Soil Index Properties
Data Analysis
Review of Existing
Correlations with
New Data
Tabulation of CBR
Values Based on Soil
Index Properties
Develop a
Correlation for
Malaysia Soil by
Curve Fitting
Figure 3.1 : Flowchart of the study
28
Soil properties such as particle size distribution, Atterberg limits, density and
moisture content were collected for the analysis and substituted into the existing
correlations to find the estimated CBR values.
The estimated CBR values which
derived from the existing correlations were compared with the CBR values obtained
from the laboratory. Comparison results were reported and discussed in the study to
evaluate the appropriateness of the correlations for the type of soils in Malaysia.
At the same time, attempt to find the typical range of CBR values of soil in
Malaysia were carried out. The proposed typical range values shall be based on the
soil classification properties of soils in Malaysia. Efforts were also made to develop
a more suitable correlation between CBR value and soil index properties for the
types of soil in Malaysia by curve fitting method.
3.2
Data Collection
3.2.1
Source of Data
Adequate data is important for carrying out the required analyses in order to
achieve the objectives of the study. The data were obtained from one or more
reliable sources to ensure that the results from the analysis are correct.
Data was obtained from a well-established and experienced soil investigation

contractor who has wide experience in carrying out soil investigation works
particularly in roads and highways construction. They also owned a well-equipped
laboratory and able to carry out most of the laboratory tests as accordance to relevant
British Standards.
The data for the study was obtained from one of the current highway project
in Malaysia.
It is located along the existing highway from Tanjung Malim to
Rawang and Senawang to Ayer Keroh. Soil investigation works were carried out to
29
obtain the subsoil profile of the project sites as well as assessing the soils for the
construction of new pavement and rehabilitation of existing pavement.
An extensive data on the soil properties and California Bearing Ratio (CBR)
values were obtained from the project sites. Laboratory testing such as particle size
distributions tests, Atterberg limits tests, compaction tests and CBR tests were
conducted by the soil investigation contractor as instructed by the engineers for the
projects.
3.2.2
Data Selection
As stated above, data was acquired from a soil investigation contractor that
had carried out extensive field works and laboratory testing for the above-mentioned
project sites. A total numbers of 65 soil data from the project sites have been used in
the study. The selected data shall consist of CBR value, optimum moisture content,
maximum dry density, particle size distribution, plastic limit, liquid limit and
plasticity index.
Optimum moisture content and maximum dry density was carried out by the
contractor on soil samples that collected by either trial pits or in bulk samples. This
is important as the CBR values obtained for the analyses are corresponding to the
optimum moisture content and maximum dry density of the soil samples.
The CBR values obtained were corresponded to the CBR test conducted
accordance to Method 5 in BS 1377 (1990) Part 4 in which it refer to the rammer
compaction to specified effort. For these project sites, 4.5kg rammer method had
been adopted to compact the soil samples in the moulds.
The soil samples were subjected to four days of soaking in the water prior to
the CBR tests. CBR values at optimum moisture content were obtained from the top
and bottom ends of soil samples and will be used in the analyses. In addition,
30
minimum CBR values in between the range of +3% and -3% of the optimum
moisture content are also recorded for both the top and bottom ends of the soil
sample.
Total contents of gravels, sands, silts and clays for each of the soil sample are
recorded. This is important to identify the type of soil, whether it is coarse-grained
soil or fine-grained soil, before carrying out any analysis. If the sample is finegrained soil, the plastic limit, liquid limit and plasticity index of the soil samples will
be used for analysis.
3.3
Data Analysis
In order to meet the objectives of the study, detailed analyses need to be

carried out on the collected data. Graphical and correlation method as well as
statistical functions integrated in Microsoft Excel were used for the analysis of data.
Graphical method is a widely used method for carrying out most analyses
particularly that involved quantitative data. It is an analytical tool that permits the
expression and comparison of control and experimental data in a graphical model.
Two types of graphical models will be used in the analysis. There are polygram and
histogram.
Polygram is also referring to line graph relating two or more variables. This
model will be adopted when carrying out evaluation on the existing established CBR
correlations with soil index properties. CBR values and soil index properties will be
plotted and compared with the lines generated from the existing CBR correlations.
Histogram is also referring to bar graph capable of representing large amount

of data. The CBR values obtained from the project sites were grouped and plotted as
histograms in relative to the soil index properties in order to obtain a typical range of
CBR values based on the soil index properties.
31
A relationship between two or more variables can be obtained by correlation

method. Subsequently, an equation can be established based on the correlation
developed. This method is not an experimental but it is a mathematical technique for
summarising the data that corresponding to one or more variables. Correlation
developed will be used as a basis for prediction. Therefore, this method will be
adopted to find the new CBR correlation with soil index properties that best suited
the type of soil in Malaysia.
Graphical analyses were used together with the correlation method in order to
find the best correlation that can be developed. Data were plotted against the CBR
values and efforts were made to establish a best fit curve or best fit line that can fit
the points as closely as possible. Equations were developed based on the best fit line
to predict CBR values with the soil classification and index properties.
35
CHAPTER 4
RESULTS AND DISCUSSION
4.1
Introduction
Numerous analyses were carried out to meet the objective of the study based
on the 65 numbers of the soil data obtained from highway projects connecting
Rawang and Tanjung Malim, and connecting Senawang and Ayer Keroh.
4.2
Particle Size Distribution
Out of the 65 soil data obtained from the project, 57 of the soil data have been
classified as fine-grained soil. The remaining of the soil data are coarse-grained soil
group. Table 4.1 summarizes the particle size distribution for fine grained soil based
on fifty seven soil data. It can be seen that the fine grained soil consist mostly of silt
contents with average of 45% and sand with average of 26%. Gravel and clay
content both recorded average percentage of about 15% only. The soils can be
described as slightly gravelly slightly sandy SILT based on the BS 5950 (1999) soil
description methods. This type of soil is soil is obviously not belong to alluvium soil
types and therefore will give some high CBR values.
33
Table 4.1 Particle size distribution test results for fine-grained soils
Soil
No.
F1
F2
F3
F4
F5
F6
F7
F8
F9
F10
F11
F12
F13
F14
F15
F16
F17
F18
F19
F20
F21
F22
F23
F24
F25
F26
F27
F28
F29
F30
F31
F32
F33
F34
F35
F36
F37
F38
F39
F40
F41
F42
F43
F44
F45
Gravel
15
22
13
17
37
31
21
7
14
17
2
0
6
4
1
14
24
23
26
4
6
3
3
4
16
15
14
31
9
1
25
1
13
11
26
24
29
22
21
18
39
24
31
19
14
Sand
15
32
24
21
22
27
30
23
14
37
11
7
13
50
47
26
26
30
28
50
37
20
25
19
35
49
28
29
40
24
33
29
39
22
24
24
17
30
26
32
19
36
26
34
20
Total Contents (%)

Silt
Clay
64
6
42
4
54
9
58
4
40
1
38
4
46
3
66
4
66
6
40
6
84
3
69
24
73
8
42
4
48
4
42
18
39
11
15
32
43
3
38
8
48
9
57
20
37
35
75
2
46
3
31
5
45
13
38
2
46
5
37
38
26
16
64
6
21
27
63
4
41
9
34
18
45
9
42
6
49
4
47
3
37
5
33
7
41
2
41
6
60
6
Clay + Silt
70
46
63
62
41
42
49
70
72
46
87
93
81
46
52
60
50
47
46
46
57
77
72
77
49
36
58
40
51
75
42
70
48
67
50
52
54
48
53
50
42
40
43
47
66
34
Table 4.1 (Contd) Particle Size Distribution of Soil Data for Fine-grained Soils
Soil
No.
F46
F47
F48
F49
F50
F51
F52
F53
F54
F55
F56
F57
Gravel
10
24
8
11
8
39
30
2
11
4
9
22
Sand
22
36
55
5
6
12
15
18
46
7
11
21
Total Contents (%)

Silt
Clay
36
32
35
5
28
9
61
23
55
31
20
29
29
26
30
50
38
5
57
32
47
33
31
26
Clay + Silt
68
40
37
84
86
49
55
80
43
89
80
57
Table 4.2 tabulates the results from the particle size distribution tests for
eight numbers of soil data that are classified as coarse-grained soils. The coarsegrained soils consist of mostly gravel contents with average of 47%. Sand, silt and
clay sum up the remaining contents with average of about 28%, 19% and 6%
respectively. Based on the BS 5950 (1999), the soil can be classified as silty slightly
sandy Gravel. It is expected that this type of soil will give higher CBR values due to
the presence of large gravel contents.
Table 4.2 Particle size distribution test results for coarse-grained soils
Soil
No.
C1
C2
C3
C4
C5
C6
C7
C8
Gravel
44
44
41
58
53
30
51
52
Sand
21
32
40
33
26
37
19
16
Total Contents (%)

Silt
Clay
33
2
21
3
17
2
9
0
20
1
26
7
9
21
17
15
Gravel + Sand
65
76
81
91
79
67
70
68
35
4.3
Comparison of CBR Values Obtained from Top and Bottom Face
According to BS 1377 (1990) Part 4, California Bearing Ratio (CBR) values

can be obtained from the top and bottom end of the soil sample and the values
obtained shall be indicated separately in the test report.
As stated in BS, the CBR
values shall be reported as CBR value at top face (CBRTOP) and CBR value at bottom
face (CBRBOTTOM) in two significant values. If the results from the both end of the
sample are within 10% of the mean value, the average results may be reported.
For the purpose of this study CBRTOP and CBRBOTTOM values were analysed
and reported separately. Figure 4.1 shows the relationship between the CBRTOP and
CBRBOTTOM values that were obtained from the soil data. A 45o line was drawn on
the graph indicating that the CBRTOP value is equal to CBRBOTTOM value. Hence,
when the plotted soil data are below the line, the CBRBOTTOM values are higher than
CBRTOP values or vice versa.
Figure 4.1 Relationship between CBRTOP and CBRBOTTOM values
36
Figure 4.1 shows that all the data are plotted below the linear line, which
indicates that the CBR values obtained from the bottom face of the soil samples in
laboratory CBR test are always higher than the CBR values recorded on the top face
of the soil samples.
This is true as soils located at the bottom of the mould are
always stiffer than the soils located at the top when subjected to compaction from the
rammer method.
A smooth line had been drawn on the figure to get a possible line that can
represent the actual relationship of the CBRTOP value with CBRBOTTOM value. As
shown in the figure, it can be presented by the following equation:
CBRTOP = 0.43 CBRBOTTOM

Where;
CBRTOP
CBR value at top face of soil sample
CBRBOTTOM
CBR value at bottom face of soil sample
(4.1)
Based on the equation above, it can be predicted that the CBRTOP value is
about 43% of the CBRBOTTOM value. However, this equation should only be used
merely for the estimation of CBRTOP and CBRBOTTOM values as it main objective is to
find the relationship between CBRTOP and CBRBOTTOM values.
4.4
Three published correlation have been selected for evaluation in the study.
The evaluation was differentiated based on the broad groups of soil, i.e. coarsegrained and fine-grained soil.
37
4.4.1
Coarse-Grained Soil
The existing published correlation for coarse-grained soil that will be

evaluated is proposed by the National Cooperative Highway Research Program
(NCHRP) of United States of America. The CBR correlation for coarse-grained soil
can be expressed as follows:
CBR = 28.09(D60 )
0.358
where;
D60
(4.2)
Diameter at 60% passing from grain size distribution

(Unit in millimeter)
Based on the above equation, analyses were carried out based on the CBRTOP
and CBRBOTTOM values obtained from the soil data. Both values are plotted in Figure
4.2 and 4.3 respectively against the soil grain diameter at 60% passing the grain size
distribution. A NCHRPs line is plotted in the figures to represent the CBR values
estimated by the CBR correlation for coarse-grained soil.
Figure 4.2 Comparison of CBRTOP with NCHRPs line for coarse-grained soil
38
Figure 4.2 shows that all the measured CBRTOP values are plotted below the
NCHRPs line. This indicates that the measured CBRTOP values are less than the
CBR values estimated from the NCHRPs correlation for coarse-grained soils by 4%
to 39%.
Six out of the eight numbers of the measured CBRBOTTOM values are plotted
below the NCHRPs line which is as shown in Figure 4.3. The predicted CBR values
based on the NCHRPs correlation for coarse-grained soils have overestimated the
measured CBRBOTTOM values within 5% to 37% for these six soil samples. For the
two measured CBRBOTTOM plotted above the NCHRPs line, the predicted CBR
values have under-estimated it as much as 28%.
Figure 4.3 Comparison of CBRBOTTOM with NCHRPs line for coarse-grained soil
Figures 4.2 and 4.3 have shown that NCHRPs CBR correlations for coarsegrained soil are unable to predict the measured CBR values correctly. This is due to
the NCHRPs correlation is for clean coarse-grained soil which is not suitable for the
coarse-grained soil that is mix with some fine-grained soils.
39
4.4.2
Fine-Grained Soil
There are two well-established correlations that are used to predict the CBR
values for the fine-grained soil: that proposed by NCHRP (2001), and by Agarwal
and Ghanekar (1970).
4.4.2.1 NCHRPs Correlation
National Cooperative Highway Research Program (NCHRP) of United States

of America has proposed the following equation below for estimating the CBR
values for fine-grained soil:
CBR =
where;
75
1 + 0.728(wPI )
Percentage passing 0.075mm sieve (in decimal)
PI
Plasticity Index
(4.3)
For comparison with the NCHRPs correlation above, the measured CBR
values from the soil data are plotted against the wPI values. Figure 4.4 and 4.5 show
the plot of the CBRTOP and CBRBOTTOM values against the wPI values respectively.
A NCHRPs line that represents the predicted CBR values based on the correlation
above is plotted in both of the figures.
As shown in the Figure 4.4, 44 out of the 57 numbers of measured CBRTOP

values are plotted below the NCHRPs line. These represent about 77% of the
measured CBRTOP values that are under-estimated when using this correlation. The
difference of the CBR values predicted by the correlation with the measured CBRTOP
values is within the range of -11% to +19%.
40
Figure 4.4 Comparison of CBRTOP with NCHRPs line for fine-grained soil
Figure 4.5 Comparison of CBRBOTTOM with NCHRPs line for fine-grained soil
41
About 22 out of 57 numbers or about 39% of measured CBRBOTTOM values
are below the NCHRPs line, as demonstrated in Figure 5.5. The difference of the
predicted CBR values with measured CBRBOTTOM values is within the range of -38%
to +16%. Hence, the range is wider and larger compared to the range when using
CBRTOP values.
Based on these two figures, it can be seen that there is no obvious trend of
plotted measured CBR values against wPI value for either CBRTOP value or
CBRBOTTOM value is observed. The plotted data are scatter particularly as shown in
Figure 4.5 for measured CBRBOTTOM values. Moreover, most of the measured CBR
values are far from the CBR values predicted by NCHRPs line.
4.4.2.2 Agarwal and Ghanekars Correlation
Agarwal and Ghanekar have proposed a CBR correlation for fine-grained soil
in year 1970. The correlation can be expressed by the following equation:
CBR = 2 16 log(OMC ) + 0.07 LL
where;
OMC =
LL
Liquid Limit
(4.4)
The accuracy of the correlation above in predicting CBR values for finegrained soil had been evaluated by using the measured laboratory CBR values.
Figure 4.6 and 4.7 demonstrate the relationship of the predicted CBR values with the
CBRTOP and CBRBOTTOM values respectively. A linear line has been drawn in the
figures indicating the measured CBR value is equal to the CBR value predicted by
Agarwal and Ghanekars correlation.
42
Figure 4.6 Relationship between CBRTOP and CBRA&G values
Figure 4.7 Relationship between CBRBOTTOM and CBRA&G values
43
Based on the Figure 4.6, most of the measured CBRTOP values are not plotted
near to the linear line except for some measured values of less than 10%. It is
observed that all the predicted CBR values are less than 10% based on the optimum
moisture content and liquid limit of the soil samples. This is not true as the measured
CBRTOP values recorded can be as high as 23%.
Figure 4.7 has shown some similarity as in the Figure 4.6. The measured
CBRBOTTOM values are plotted almost vertically against the predicted CBR values
and away from the linear line. In addition, all the predicted CBR values are less than
10% although the measured CBRBOTTOM values are within the range of 2% to 52%.
Out of the 57 data obtained, only two data or about 4% of the data are seen below the
linear line. This indicates that the correlation has under-estimated about 96% of the
measured CBRBOTTOM values.
It can be summarised that the CBR values predicted by the Agarwal and
Ghanekars correlation are less than 10% based on the soil index properties for all
the soil data. This is not correct as many of the measured CBR values are more than
10% especially for measured CBRBOTTOM values. Therefore, the correlation is not
suitable for predicting CBR value of soil that is higher than 10%.
4.5
The CBR values from the data had been plotted in histograms to permit
analysing the typical CBR values that can be obtained from the soil samples. For
purpose of analysis, the data had been divided into two main groups, i.e. coarsegrained soil and fine-grained soil.
44
4.5.1
Coarse-Grained Soil
CBRTOP values of coarse-grained soils were plotted in the Figure 4.8 based
on the number of measurements from the eight soil data obtained. As can be seen in
the figure, the values scatter quite widely within 2% to 40% and not even one of the
CBRTOP value is being measured twice.
Figure 4.8 Numbers of measurement of CBRTOP for coarse-grained soil

Number of measurements for CBRBOTTOM values of coarse-grained soils was
presented in Figure 4.9. Based on the eight soil data obtained, it can be observed that
the range of CBRBOTTOM values is within 10% to 39% except for one sample that had
recorded 79%. Each of the CBR values recorded only once for all the soil data
acquired.
Based on all the figures, it can be deduced that the measured CBRTOP values
for coarse-grained soils are within the range of 2% to 40%. As for CBRBOTTOM, the
range is between 10% to 79%. However, the ranges for coarse-grained soil above is
based on eight soil data only and should be analysed further with more data.
45
Figure 4.9 Numbers of measurement of CBRBOTTOM for coarse-grained soil
4.5.2
Fine-Grained Soil
CBRTOP values of fine-grained soils are plotted in the Figure 4.12 based on
the number of measurements from the 57 soil data. As can be seen in the figure, the
recorded values are within the range 2% to 23%.
It shall be noted from the Figure 4.10 that the data are concentrated within the
range of 2% to 10%. Generally, the CBRTOP values within this range have been
measured at least twice. The highest number of measurements for CBRTOP values is
4% where it had been recorded for 12 times.
Figure 4.11 shows the lowest CBR values on top face of the soil samples
within the range of 3% of the optimum moisture content. It can be seen that the
range of CBRTOP(3%) now is smaller than the CBRTOP values which is from 1% to
12% with one sample recorded CBRTOP(3%) more than 10%.
46
Figure 4.10 Numbers of measurement of CBRTOP for fine-grained soil
Based on the Figure 4.11, it can be seen that the CBRTOP(3%) values mostly
concentrated within the range of 1% to 7%. It is observed that the CBRTOP(3%)
values of 3% had recorded 13 times which is the highest number of measurements
followed by CBRTOP(3%) values of 2% and 1%.
Number of measurements of CBRBOTTOM values of fine-grained soils was

plotted in Figure 4.13. Based on the 57 soil data, the recorded range of CBRBOTTOM
values are widely spread within 2% to 52%.
Most of the data in Figure 4.13 had recorded CBRBOTTOM values of 6% to

20% in which 7% and 15% had recorded five times each. CBRBOTTOM values had
recorded once only for value higher than 20% except for CBRBOTTOM values of 30%
which had recorded thrice.
Figure 4.13 presented the number of measurements for CBRBOTTOM(3%)

values for fine-grained soils from the soil data obtained. It is to highlight again that
47
the CBRBOTTOM(3%) value refers to lowest CBRBOTTOM values obtained within the
range of 3% of optimum moisture content.
Figure 4.11 Numbers of measurement of CBRTOP(3%) for fine-grained soil
Figure 4.12 Numbers of measurement of CBRBOTTOM for fine-grained soil
48
Figure 4.13 Numbers of measurement of CBRBOTTOM(3%) for fine-grained soil

Based on the Figure 4.13, the recorded values are within 1% to 17% in which
it has a smaller range compared to CBRBOTTOM values. The highest number of
measurement of CBRBOTTOM(3%) values is 5% which had been recorded for a total
number of ten times.
Based on the CBR values concentration in all of the figures, it can be

summarised that the measured CBRTOP values for fine-grained soils are within the
range of 2% to 10% and measured CBRTOP(3%) values are within 1% to 7%. As for
CBRBOTTOM and CBRBOTTOM(3%) values, the range is between 6% to 20% and 1% to
17% respectively.
4.6
Relationship of Maximum Dry Density with Optimum Moisture Content
Based on the soil data acquired, the relationship of the maximum dry density
with the optimum moisture content had been analysed. If there is a correlation
49
between these two index properties of soil, it can be used for analysing the data to
find the correlation between CBR values with soil index properties.
Figure 4.14 Relationship of maximum dry density with optimum moisture content
The plot of the maximum dry density against the optimum moisture content is
presented in Figure 4.14 based on the 57 soil data for fine-grained soils. Figure 4.14
shows that the maximum dry density can be correlated with optimum moisture
content using the following equation below.
MDD
where;
22.842 - 0.2976 OMC
MDD =
Maximum Dry Density (kN/m3)
OMC =
Optimum Moisture Content (%)
(4.5)
The fine-grained soils data used in the analysis have optimum moisture
content within the range of 9.8% to 24.8% and maximum dry density of 14.5 kN/m3
to 20.5 kN/m3. It is observed that the maximum dry density will be lower if the
optimum moisture content is getting higher. As the maximum dry density can be
50
correlated with the optimum moisture content, it is a good indication that these soil
properties can be used to find a CBR correlation with these soil properties.
4.7
Proposed Correlation for CBR Values
Based on the soil data and the relationship between maximum dry density and
optimum moisture content as stated in Section 4.3 and Section 4.5 above, a
correlation between CBR values with these soil index properties had been analysed.
As highlighted earlier, the soil data for coarse-grained soils will not be used to obtain
CBR correlation with its soil index properties. This is due to the shortage of data
available for coarse-grained soils. Therefore, the correlation obtained will be strictly
applied for the fine-grained soils only.
It is current practice in Malaysia that CBR at top face is taken as design value
of subgrade for pavement. As our scope of study is focus on our country only, a
correlation between CBR value at top face of soil sample, the CBRTOP value for finegrained soil can be correlated with maximum dry density and optimum moisture
content. The Figure 4.15 shows the plot of maximum dry density against the ratio of
CBRTOP value with optimum moisture content.
CBRBOTTOM values from soil data have been used for the correlation. The
values were converted to CBRTOP values by using the Equation (4.1) above before
being plotted in the figure. As such, the total number of soil data being analysed is 57
numbers.
51
Figure 4.15 Proposed correlations for CBRTOP for fine-grained soil

A best fit curve line was obtained from the plotted data as shown in the
figure. The line can be represented by a power equation as shown below:
MDD
= 19.2 (CBRTOP/OMC) 0.05
(4.7)
or
CBRTOP = OMC (MDD/19.3)20
where;
(4.8)
CBRTOP
CBR Values at Top Face of Sample (%)
MDD
Maximum Dry Density (kN/m3)
OMC
Optimum Moisture Content (%)
Based on the equation, CBRTOP value can be predicted based on the

maximum dry density and optimum dry density of a given soil. It is worth to be
noted that these index properties can be obtained from compaction test results. As
52
such, one can estimate the CBRTOP value of a given soil based on the compaction test
results.
4.8
Discussion
Soil data had been obtained and analysed accordingly within the scope of the
study.
All soil information were obtained from a established soil investigation
contractor which also has own a laboratory for carrying out most of the laboratory
tests accordance to British Standard.
Data acquired for analyses ranging from CBR values for top and bottom end
of soil samples, maximum dry density, optimum moisture content, particle size
distribution and Atterberg limits for each of the soil samples.
Total soil data
obtained are 65 numbers with eight numbers are for coarse-grained soils and the
remaining 57 numbers are belong to fine-grained soils. As such, the soil data were
grouped into coarse-grained soils and fine-grained soils for purpose of analyses.
4.8.1
Total of three published CBR correlations had been evaluated using the soil
data obtained for the study. Generally, these correlations can be divided into two
groups based on the type of soils i.e. coarse-grained soil and fine-grained soil. For
coarse-grained soil, only one correlation evaluated using the eight soil data and it is
proposed by the National Cooperative Highway Research Program (NCHRP) of
United States of America. The CBR correlation can be expressed as Equation (4.2).
The results of the evaluation were presented in Section 4.4.1 where CBR
values for top and bottom end of soil sample had been analysed separately. It is to be
noted that CBR value at top face of soil sample (CBRTOP) are always lower than
CBR value at bottom face of soil sample (CBRBOTTOM) as stated in Section 4.3.
53
Hence, it is prudent to evaluate the correlation with these two CBR values. Based on
the analysis results, it is observed that the estimated CBR values are very high
compared with measured CBRTOP values.
It is understood that the difference
between estimated CBR values and measured CBRTOP values are ranging from 4% to
39%. This indicates that the correlation will overestimate the CBRTOP values.
As for estimating CBRBOTTOM values, the analysis shows that the result is
almost similar for estimating the CBRTOP values. Six out of the eight soil samples
recorded lower CBRBOTTOM values than estimated CBR values. The remaining soil
samples recorded CBRBOTTOM values higher than estimated CBR values and the
difference is as high as 28%.
Both the measured CBRTOP and CBRBOTTOM values appeared to be

overestimated by the NCHRPs correlation for coarse-grained soils. It is to be noted
that the correlation is meant for clean coarse-grained soil whereas for the soils
obtained are coarse-grained soils mixed with small percentage amounts of finegrained soils. Therefore, it is believe that the type of soil will has lower CBR value
compared to pure coarse-grained soil.
In summary, the results of the analysis show that the NCHRPs correlation is
not suitable for estimating the CBR value for coarse-grained soils. However, the
results are based on limited soil data. More soil data shall be obtained to carry out
further evaluation on the correlation.
Two correlations had been evaluated with the fine-grained soil data and are
presented in Section 4.4.2. The first CBR correlation evaluated is proposed by the
National Cooperative Highway Research Program (NCHRP) of United States of
America. The correlation can be expressed as Equation (4.3).
The results of the analysis for the correlation above are presented in Section
4.4.2.1. Based on the results, 77% of the total estimated CBR values are lower than
the measured CBRTOP values indicating underestimate of CBRTOP values. As for
54
estimating CBRBOTTOM values, it is appear that the correlation is unable to predict the
CBR values correctly. The results shown that the difference of the estimated CBR
values with measured CBRBOTTOM values had a very wide range i.e. between -38% to
+16%.
Therefore, it can be summarised that the NCHRPs CBR correlation for finegrained soil is not appropriate to be used for fine-grained soil in Malaysia. The
chances for overestimating the CBRTOP values are likely to be high. In addition, the
predicted CBRBOTTOM values may just be too far from the actual values.
Agarwal and Ghanekars correlation for fine-grained soil was evaluated with
the fine-grained soil data and is presented in Section 4.4.2.2. The CBR correlation
can be expressed as Equation (4.4). It is appeared that the CBR correlation will
underestimate the CBRTOP and CBRBOTTOM values if the soil has CBR value over
10%. The analysis results show that all predicted CBR values are below 10%. In
addition, it is observed that about 96% of the total recorded CBRBOTTOM values are
higher than the predicted CBR values. It indicates that the correlation is very prone
to underestimate the CBRBOTTOM values.
It can be concluded that the Agarwal and Ghanekars CBR correlation is

definitely not suitable for estimating fine-grained soil that has CBR values more than
10%.
4.8.2
It is the objective of the study to find the typical range of CBR values based
on soil index properties. However, no significant range of CBR values is observed
based on the soil index properties acquired from the soil data. Hence, the typical
range of CBR values will be based on the numbers of time the CBR values was
recorded.
55
Based on the analysis results for coarse-grained soil which is presented in
Section 4.5.1, the range of CBRTOP value is within 2% to 40%. Meanwhile for CBR
value at bottom face, the range obtained for CBRBOTTOM value is within 10% to 79%.
For fine-grained soil, the CBRTOP values recorded a range within 2% to 23%.
However, the analysis results as presented in Section 4.5.2 shows that the CBRTOP
values are concentrated within the range of 2% to 10%. The recorded range of
CBRTOP(3%) value is within the range of 1% to 12% with the soil data concentrated
in between 1% to 7%. CBRBOTTOM for fine-grained soils had been recorded within
the range of 2% to 52%. From the soil data concentration, the typical range can be
narrow downed to within 6% to 20%. The typical range of CBRBOTTOM(3%) for finegrained soil as recorded in the analysis is between 1% and 17%.
4.8.3
CBR Correlation with Soil Index Properties for Malaysia Soils
Soil index properties can be used to correlate with CBR values. This has
been shown in many published CBR correlations in the world.
However, the
published correlations are only reliable for predicting CBR values for the type of soil
where the correlation is origin. Hence, it is important to find a correlation that can
estimate CBR values based on soil index properties in Malaysia.
Prior to the analysis to find the CBR correlation with soil index properties,
the relationship of maximum dry density with optimum moisture content is analyses.
Based on the results as presented in Section 4.6, it is found that these soil properties
can be interrelated by a linear line. This is essential finding that can be used to find
CBR correlation based on these index properties. The linear line can be expressed by
the Equation (4.8).
Based on the finding above, CBRTOP value had been correlated with
maximum dry density and optimum moisture content for all the soil data. The
analyses carried out are presented in Section 4.7. Correlations were obtained for
56
estimating the CBRTOP value of fine-grained soils. The correlation is expressed as
Equation (4.8).
The Equation (4.8) are valid for estimating CBRTOP value for fine-grained
soil only in which it had been developed from 57 numbers of soil data. Therefore,
the CBR correlations above shall be good enough to estimate CBRTOP value for finegrained soil in Malaysia based on soil index properties for preliminary design.
Generally, the typical range of CBRTOP values obtained from the study is between
2% to 10%. As such, if the predicted CBR value from the equation give value more
than 10%, it is prudent to carry out laboratory tests to confirm this value prior to
design.
57
CHAPTER 5
CONCLUSIONS AND RECOMMENDATIONS
5.1
Conclusion
Soil data had been obtained and analysed accordingly within the scope of the
study.
All soil information were obtained from a established soil investigation
contractor which also has own a laboratory for carrying out most of the laboratory
tests accordance to British Standard.
Data acquired for analyses are the from CBR values for top and bottom end
of soil samples, maximum dry density, optimum moisture content, particle size
distribution and Atterberg limits for each of the soil samples.
Total soil data
obtained was 65 numbers with eight numbers can be classified as coarse-grained

soils and the remaining 57 numbers were belong to fine-grained soils. As such, the
soil data were grouped into coarse-grained soils and fine-grained soils for purpose of
analyses.
Based on the analyses carried out, the conclusion of the study can be
summarised as follow:
1.
The published correlations are not suitable to estimate CBR values fo the
soil under study.
58
a)
CBR correlation for coarse-grained soils proposed by NCHRP is

likely to overestimate the actual CBR values. This may due to the
fact that this correlation is meant for clean coarse-grained soil and
not for coarse-grained soils that are mixed with small percentage
amounts of fine-grained soils.
The NCHRPs correlation to predict CBR value for fine-grained

soil is not appropriate to be used in Malaysia. The chances for
overestimating the CBRTOP values are likely to be high and the
predicted CBRBOTTOM values may just be too far from the actual
values.
b)
The Agarwal and Ghanekars CBR correlation is definitely not

suitable for estimating fine-grained soil that has CBR values more
than 10%.
2.
Typical range of CBR value can be found for residual soil in Malaysia.
For coarse-grained soils, the range of CBRTOP value is within 2% to 40%

whereas for CBRBOTTOM value it is within 10% to 79%.
For fine-grained soils, the CBRTOP values are concentrated within the
range of 2% to 10% whereas CBRBOTTOM values have typical range that
concentrated within 6% to 20%. The recorded range of CBRTOP(3%)
value is within the range of 1% to 12% with the soil data concentrated in
between 1% to 7%. The typical range of CBRBOTTOM(3%) as recorded in
the analysis is between 1% and 17%.
59
3.
New correlations have been proposed for predicting the CBR values at
top face of soil sample for fine-grained soil in Malaysia based on
maximum dry density and optimum moisture content of soil. The
following correlation is:
CBRTOP
OMC (MDD/19.2)20
Generally, the typical range of CBRTOP values obtained in the study is

between 2% to 10%. Therefore, if the predicted CBRTOP value from the
above correlation is more than 10%, laboratory testing should be carried
out to confirm the CBR value for design.
5.2
Recommendations for Future Study
Due to time constraints and limited soil data obtained for coarse-grained
soils, there are some aspects which have not been covered in the study. Following
are some recommendations that can be carried out future study or research in the
subject of CBR correlation with soil index properties in Malaysia.
1.
The type of soils obtained for this study is most probably the residual
soils since it consist of gravels, sands, silts and clays.
It will be
interesting to obtain different types of soil such as alluvium soils, marine

clays, pure sands, etc. for further study.
2.
More soil data for coarse-grained soils should be obtained to evaluate the
published correlations and find the correlation with soil index properties
that can be used in Malaysia.
3.
Published correlation proposed by other researchers should be evaluated

to find the suitability to the types of soil in Malaysia.
4.
Different soil index properties other than maximum dry density and
optimum moisture content should be used to find a CBR correlation.
60
5.
Typical range of soils based on the soil index properties are not found in
this study. Further study or research shall be carried out on this topic.
6.
Experimental research by carrying out laboratory tests should be

conducted to evaluate the accuracy of the proposed CBR correlations that
were developed from this study.
7.
Soil data from other project sites in different states within Peninsular
Malaysia as well as East Malaysia should be collated and analysed to
confirm the finding of the study.
8.
The effects of using maximum dry density and optimum moisture content
obtained from the compaction test by 2.5kg rammer method if used in the
new proposed correlations.
9.
Study on the unsoaked CBR values correlated with soil index properties
can be carried out.
61
REFERENCES
Agarwal, K.B. and Ghanekar, K.D. (1970). Prediction of CBR from Plasticity
Characteristics of Soil. Proceeding of 2nd South-east Asian Conference on Soil
Engineering, Singapore. June 11-15, 1970. Bangkok: Asian Institute of
Technology, 571-576.r
American Standard Test Method (1992). Standard Test Method for CBR (California
Bearing Ratio) of Laboratory-Compacted Soils. United States of America,
ASTM Designation D1883-92.
Black, W.P.M. (1962). A Method of Estimating the CBR of Cohesive Soils from
Plasticity Data. Geotechnique. Vol.12: 271 - 272.
British Standards Institution (1990). Methods of Test for Soils for Civil Engineering
Purposes. London, BS 1377.
British Standards Institution (1999). Code of Practice for Site Investigations.
London, BS 5950.
Carter, M. and Bentley, S. P. (1991). Correlations of Soil Properties. London:
Pentech Press.
de Graft - Johnson, J.W.S. and Bhatia, H.S. (1969). The Engineering Characteristics
of the Lateritic Gravels of Ghana. Proceedings of 7th Inernational Conference on
Soil Mechanics and Foundation Engineering, Mexico. August 28-29. Bangkok:
Asian Institute of Technology. Vol.2: 13 - 43.
62
National Cooperative Highway Research Program (2001) Guide for Mechanistic and
Empirical Design for New and Rehabilitated Pavement Structures, Final
Document. In: Appendix CC-1: Correlation of CBR Values with Soil Index
Properties. West University Avenue Champaign, Illinois: Ara, Inc.
Steve, L. W., Richard, H. G. and Thomas, P. W. (1992) Description and Applications
of Dual Mass Dynamic Penetrometer. Washington, DC: US Army Corps of
Engineers.
Terzaghi, K., Peck, R.B. and Mesri, G. (1996) Soil Mechanics in Engineering
Practice. 3rd ed. United States of America: John Wiley & Sons, Inc.
The Highway Agency (1994) Design Manual for Roads and Bridges. In: Volume 7:
Section 2 Part 2 HD 25/94. London: Stationery Ltd.
63
APPENDIX A
Atterberg limits test results for coarse-grained soils
Soil
No.
Liquid Limit
(%)
Plastic Limit
(%)
Plasticity Index
(%)
Soil Plasticity
Based on A-Line
C1
42
30
12
MI
C2
37
23
14
CI
NON - PLASTIC
C3
C4
35
24
11
MI
C5
31
22
ML
C6
38
25
13
MI
C7
43
25
18
CI
C8
68
38
30
MH
64
APPENDIX B
Atterberg limits test results for fine-grained soils
Soil
No.
Liquid Limit
(%)
Plastic Limit
(%)
Plasticity Index
(%)
Soil Plasticity
Based on A-Line
F1
39
26
13
MI
F2
31
22
ML
F3
45
28
17
MI
F4
49
31
18
MI
F5
36
25
11
MI
F6
42
28
14
MI
F7
43
29
14
MI
F8
39
27
12
MI
F9
44
28
16
MI
F10
38
25
13
MI
F11
45
31
14
MI
F12
70
42
28
MH
F13
56
37
19
MH
F14
53
26
27
CH
F15
42
25
17
CI
F16
60
31
29
MH
F17
64
33
31
MH
F18
87
33
54
CV
F19
33
22
11
CL
F20
62
39
23
MH
F21
39
24
15
CI
F22
71
40
31
MV
F23
55
23
32
CH
F24
39
28
11
MI
F25
33
22
11
CL
F26
25
17
CL
F27
37
23
14
CI
F28
37
25
12
MI
F29
27
19
CL
F30
53
21
32
CH
F31
42
22
20
CI
F32
44
30
14
MI
F33
61
27
34
CH
F34
37
25
12
MI
F35
33
21
12
CL
65
APPENDIX B
Atterberg limits test results for fine-grained soils (Contd)
Soil
No.
Liquid Limit
(%)
Plastic Limit
(%)
Plasticity Index
(%)
Soil Plasticity
Based on A-Line
F36
39
21
18
CI
F37
41
28
13
MI
F38
31
21
10
CL
F39
38
26
12
MI
F40
35
24
11
MI
F41
33
22
11
CL
F42
35
23
12
CI
F43
40
27
13
MI
F44
37
25
12
MI
F45
48
29
19
MI
F46
54
43
11
MH
F47
33
26
ML
F48
29
19
10
CL
F49
92
52
40
ME
F50
69
44
25
MH
F51
78
43
35
MV
F52
71
36
35
MV
F53
69
31
38
CH
F54
39
28
11
MI
F55
67
41
26
MH
F56
71
43
28
MV
F57
53
27
26
CH
66
APPENDIX C
Compaction test results for coarse-grained soils
Soil No.
Maximum Dry Density

(kN/m3)

(%)
C1
12.5
11.2
C2
8.8
9.8
C3
13.4
22.9
C4
12.7
23.8
C5
13.6
13.8
C6
14.6
23.0
C7
10.0
17.9
C8
11.9
14.2
67
APPENDIX D
Compaction test results for fine-grained soils
Soil No.
Maximum Dry Density

(kN/m3)

(%)
F1
19.0
14.4
F2
19.3
10.6
F3
17.4
16.3
F4
18.3
15.0
F5
19.3
12.4
F6
17.2
18.6
F7
18.6
12.4
F8
17.9
10.8
F9
19.2
14.4
F10
18.9
14.5
F11
15.8
24.8
F12
17.1
15.0
F13
16.6
19.1
F14
18.3
13.3
F15
19.2
14.8
F16
17.7
18.2
F17
18.3
15.0
F18
18.1
14.8
F19
18.8
13.0
F20
18.6
13.8
F21
17.7
16.0
F22
17.6
15.4
F23
19.3
12.7
F24
19.5
9.8
F25
20.5
9.9
F26
18.0
16.2
F27
19.4
13.6
F28
19.4
13.7
F29
19.4
13.4
F30
18.0
17.5
F31
20.0
10.9
F32
18.3
13.1
F33
18.2
14.0
F34
17.6
17.4
F35
18.5
15.4
APPENDIX D
Compaction test results for fine-grained soils (Contd)
68
Soil No.
Maximum Dry Density

(kN/m3)

(%)
F36
18.5
15.8
F37
18.3
15.8
F38
19.2
13.4
F39
17.1
17.3
F40
19.0
13.4
F41
19.5
12.2
F42
20.3
10.4
F43
18.8
14.7
F44
19.2
13.6
F45
18.6
12.4
F46
18.0
15.4
F47
19.2
11.4
F48
19.7
11.0
F49
14.5
24.8
F50
17.8
14.6
F51
17.8
14.8
F52
17.6
18.6
F53
16.8
18.2
F54
18.7
13.0
F55
15.4
24.4
F56
16.3
20.4
F57
19.5
12.8
69
APPENDIX E
Measured laboratory CBR values for coarse-grained soils
Soil
No.
Top Face Sample
Bottom Face Sample
CBRTOP
(%)
CBRTOP(3%)
(%)
CBRBOTTOM
(%)
CBRBOTTOM(3%)
(%)
C1
22
C2
13
34
C3
25
33
C4
28
16
39
22
C5
10
C6
19
C7
40
10
79
18
C8
22
28
12
70
APPENDIX F
Measured laboratory CBR values for fine-grained soils
Soil
No.
Top Face Sample
Bottom Face Sample
CBRTOP
(%)
CBRTOP(3%)
(%)
CBRBOTTOM
(%)
CBRBOTTOM(3%)
(%)
F1
13
29
F2
10
33
F3
F4
F5
15
10
F6
10
F7
15
11
F8
F9
10
F10
20
F11
F12
17
F13
12
11
F14
18
38
13
F15
17
F16
20
F17
11
F18
10
21
F19
22
F20
18
F21
21
42
F22
25
12
F23
30
17
F24
13
20
12
F25
18
F26
F27
F28
F29
F30
11
18
12
F31
20
49
11
F32
15
F33
23
11
30
12
F34
APPENDIX F
Measured laboratory CBR values for fine-grained soils (Contd)
Soil
No.
Top Face Sample
71
Bottom Face Sample
CBRTOP
(%)
CBRTOP(3%)
(%)
CBRBOTTOM
(%)
CBRBOTTOM(3%)
(%)
F35
10
F36
F37
10
F38
F39
F40
F41
F42
10
30
F43
19
F44
F45
12
F46
13
F47
15
52
12
F48
32
F49
F50
15
F51
16
13
F52
39
14
F53
F54
10
15
F55
F56
11
F57
12
18
12
APPENDIX G
Measured soil index properties required for NCHRPs correlations
Soil
No.
Diameter at 60% Passing from

Grain Size Distribution, D60
(mm)
Percentage Passing 0.075mm

Sieve, w
(in Decimal)
C1
2.8
0.36
C2
2.4
0.25
C3
1.1
0.21
C4
4.6
0.1
C5
0.22
C6
0.91
0.34
C7
5.2
0.31
C8
4.2
0.33
F1
0.043
0.71
F2
0.23
0.48
F3
0.046
0.64
F4
0.05
0.63
F5
1.5
0.42
F6
0.85
0.43
F7
0.2
0.51
F8
0.5
0.71
F9
0.03
0.73
F10
0.3
0.48
F11
0.013
0.88
F12
0.0085
0.93
F13
0.028
0.83
F14
0.15
0.48
F15
0.11
0.54
F16
0.06
0.62
F17
0.35
0.52
F18
0.8
0.48
F19
0.3
0.47
F20
0.34
0.47
F21
0.7
0.61
F22
0.015
0.78
F23
0.028
0.73
F24
0.034
0.78
F25
0.23
0.51
F26
0.25
0.37
72
73
APPENDIX G
Measured soil index properties required for NCHRPs correlations (Contd)
Soil
No.
Diameter at 60% Passing from

Grain Size Distribution, D60
(mm)
Percentage Passing 0.075mm

Sieve, w
(in Decimal)
F27
0.08
0.59
F28
0.7
0.41
F29
0.18
0.52
F30
0.024
0.77
F31
0.35
0.43
F32
0.027
0.71
F33
0.19
0.5
F34
0.04
0.68
F35
0.16
0.52
F36
0.16
0.53
F37
0.31
0.56
F38
0.25
0.5
F39
0.19
0.54
F40
0.21
0.51
F41
1.7
0.43
F42
0.32
0.42
F43
0.7
0.44
F44
0.24
0.48
F45
0.39
0.67
F46
0.022
0.69
F47
0.24
0.43
F48
0.23
0.4
F49
0.0044
0.85
F50
0.006
0.87
F51
1.6
0.5
F52
0.25
0.56
F53
0.0031
0.82
F54
0.37
0.44
F55
0.0075
0.9
F56
0.0067
0.81
F57
0.13
0.58
APPENDIX H
Estimated CBR values from NCHRPs correlation for coarse-grained soils
Soil No.
Estimated CBR Value

(%)
C1
41
C2
38
C3
29
C4
49
C5
46
C6
27
C7
51
C8
47
74
APPENDIX I
Estimated CBR values from NCHRPs correlation for fine-grained soils
Soil No.
Estimated CBR Value

(%)
F1
10
F2
18
F3
F4
F5
17
F6
14
F7
12
F8
10
F9
F10
14
F11
F12
F13
F14
F15
10
F16
F17
F18
F19
16
F20
F21
10
F22
F23
F24
10
F25
15
F26
24
F27
11
F28
16
F29
19
F30
F31
10
F32
F33
F34
11
F35
14
75
76
APPENDIX I
Estimated CBR values from NCHRPs correlation for fine-grained soils (Contd)
Soil No.
Estimated CBR Value

(%)
F36
F37
12
F38
16
F39
13
F40
15
F41
17
F42
16
F43
15
F44
14
F45
F46
11
F47
24
F48
19
F49
F50
F51
F52
F53
F54
17
F55
F56
F57
APPENDIX J
Estimated CBR values based on Agarwal & Ghanekars Correlation
Soil No.
Estimated CBRA&G Value

(%)
F1
F2
F3
F4
F5
F6
F7
F8
F9
F10
F11
F12
F13
F14
F15
F16
F17
F18
F19
F20
F21
F22
F23
F24
F25
F26
F27
F28
F29
F30
F31
F32
F33
F34
F35
77
78
APPENDIX J
Estimated CBR values based on Agarwal & Ghanekars Correlation (Contd)
Soil No.
Estimated CBRA&G Value

(%)
F36
F37
F38
F39
F40
F41
F42
F43
F44
F45
F46
F47
F48
F49
F50
F51
F52
F53
F54
F55
F56
F57
APPENDIX K
Determination of the CBR value extracted from BS 1377 Part 4:1990
79
80
APPENDIX K
Determination of the CBR value extracted from BS 1377 Part 4:1990 (Contd)
81
APPENDIX K
82
APPENDIX K
83
APPENDIX K
84
APPENDIX K
85
APPENDIX K
86
APPENDIX K
87
APPENDIX K
88
APPENDIX K
89
APPENDIX K
90
APPENDIX K
91
APPENDIX K
92
APPENDIX K
93
APPENDIX K
94
APPENDIX K
95
APPENDIX K
96
APPENDIX K
97
APPENDIX K
APPENDIX L
Determination of the CBR value extracted from ASTM D 1883 - 92
98
99
APPENDIX L
Determination of the CBR value extracted from ASTM D 1883 - 92 (Contd)
100
APPENDIX L
101
APPENDIX L
102
APPENDIX L
103
APPENDIX L
104
APPENDIX L
105
APPENDIX L

CBR Soil Paper PDF

Uploaded by

Copyright:

Available Formats

You might also like

CBR Soil Paper PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CBR Soil Paper PDF

Uploaded by

Copyright:

Available Formats

PSZ 19:16 (Pind.

Universiti Teknologi Malaysia

BORANG PENGESAHAN STATUS TESIS

SESI PENGAJIAN : 2005 / 2006

MAK WAI KIN

Tesis adalah hakmilik Universiti Teknologi Malaysia.

(Mengandungi maklumat yang berdarjah keselamatan atau

(Mengandungi maklumat yang TERHAD yang telah ditentukan

404 , Blok A1 , Seksyen 2 ,

Potong yang tidak berkenaan.

PSZ 19:16 (Pind. 1/97)

Dr. Nurly Gofar

CALIFORNIA BEARING RATIO CORRELATION WITH

MAK WAI KIN

A project report submitted in partial fulfillment of

Faculty of Civil Engineering

MAK WAI KIN

To my beloved parents and sisters

I would like to take this opportunity to express my sincere appreciation to all

Secondly, I am thankful to my companys director and colleagues for their

California Bearing Ratio (CBR) is a commonly used indirect method to

Aim and Objectives of Study

California Bearing Ratio

Applications of California Bearing Ratio

In Situ Field Testing

Grain Size Distribution

Design Manual for Roads and Bridges (1994)

de Graft - Johnson and Bhatia (1969)

Agarwal and Ghanekar (1970)

Current Practice in Malaysia

Correlations between CBR and Soil Classification

RESULTS AND DISCUSSIONS

4.2 Particle Size Distribution

Relationship of CBR at Top Face and Bottom Face

Evaluation of Published Correlations

Agarwal and Ghanekars Correlation

Typical Range of CBR Values

4.6 Relationship of Maximum Dry Density with Optimum

Proposed Correlation for CBR Values

Evaluation of Published Correlations

Typical Range of CBR Values

CBR Correlation with Soil Index Properties for

CONCLUSIONS AND RECOMMENDATIONS

Recommendations for Future Research

Definitions of soils classified by grading according to British

Relationship of plasticity with liquid limit

Subgrade CBR estimation of British soils compacted at

Particle size distribution test results for fine-grained soils

Particle size distribution test results for coarse-grained soils

Dynamic cone penetrometer equipment

Test equipment for determination of CBR value in

Example of grading curves

Relationship between CBR and plasticity index at

Correction of CBR values for partial saturation

Relationship between suitability index and soaked CBR

Relationship between the ratio of maximum dry density

Flowchart of the study

Relationship between CBRTOP and CBRBOTTOM values