HB160 2006

Accessed by SWINBURNE UNIVERSITY OF TECHNOLOGY on 07 Oct 2015 (Document currency not guaranteed when printed)
HB
HB 160—2006
Soils testing
HB 160—2006
Handbook
Soils testing
First published as HB 160—2006.
COPYRIGHT
© Standards Australia
All rights are reserved. No part of this work may be reproduced or copied in any form or by
any means, electronic or mechanical, including photocopying, without the written
permission of the publisher.
Published by Standards Australia, GPO Box 476, Sydney, NSW 2001, Australia
ISBN 0 7337 7494 6
HB 160—2006 ii
PREFACE
GENERAL
Since the early 20th century, the testing of soils for engineering purposes has played a very
important role in the construction of the infrastructure of Australia.
Standards Australia issued the first Methods of testing soils for engineering purposes in
1966 as the AS A89 series. These methods were revised in 1977 to take into account
metrication and presented as the AS 1289 A-F series. In the 1990s, the methods were
revised to take into account major changes in the soils testing industry and the series was
reissued as the AS 1289 series. Since that time, continual revision of the test methods has
taken place as technology and the construction industries have changed.
The Standards Australia committee responsible for the Methods of Testing Soils for
Engineering Purposes (CE-009) decided in 1999 that a handbook was required to—
• assist in the education of undergraduate engineers in the use and application of the
tests;
• assist geotechnicians seeking qualifications as soil testers; and
•
provide background material and useful tips to practicing geotechnicians in over 400
Australian laboratories that perform the test.
Subsequently, Standards Australia appointed a subcommittee of soils practitioners involved
in both geotechnical engineering design, contract management, engineering education and
soils testing to compile the information presented in this Handbook. It is not comprehensive
in that all methods are not fully covered due to limited time and resources; however, the
subcommittee recommended that it be published in its present form to achieve the initial
objectives. It was also considered that, since a limited number of practitioners and experts
had been involved in its preparation, there would be a need for the document to be
discussed, debated and revised in a few years.
This Handbook is intended to be of assistance to users of the 1289 series of tests, Methods
of testing soils for engineering purposes. Users are encouraged to participate in any future
revisions by offering constructive suggestions and criticism.
The terms used in this document are defined in AS 1289.0, Methods of testing soils for
engineering purposes, Part 0: General requirements and list of methods.
THE NEED FOR TESTING

Soils testing is called for in the following documents:
• AS ISO/IEC 17025—General requirements for the competence of testing and
calibration laboratories
• AS 1726—Geotechnical investigations
AS 1726 provides a guide to the samples and tests that are required for a
geotechnical engineer to be able to provide adequate information about the soils at a
proposed construction site. The nature and location of the sampling will depend on
what engineering information is required.
• AS 2159—Piling—Design and installation
• AS 2870—Residential slabs and footings—Construction
• AS 2870 Supplement 1—Residential slabs and footings—Construction—Commentary
• HB 28—The design of residential slabs and footings—(Standards Australia)
• AS 3798—Guidelines on earthworks for commercial and residential developments
AS 3798 provides a guide to engineers as to what investigations are to be carried out
(Chapter 2) and the type of materials that are required (Chapter 4).
iii HB 160—2006
• AS 5100—Bridge design (all parts)

• Austroads (2004)—Pavement rehabilitation—A Guide to the design of rehabilitation
treatments for road pavements AP-G78/04
• Austroads (1998(a))—Guide to stabilisation in roadworks, Austroads, Sydney, 1998
• Austroads (2004)—Pavement design—A Guide to the structural design of road
pavements, AP-G17/04
• Austroads (2004)—Austroads pavement rehabilitation—A Guide to the design of
rehabilitation treatments for road pavements, AP-G78/04
• Austroads (2002)—Mix design for pavement materials stabilised in situ (flow charts)
AP-T16/02
• ISO 17011 (2004)—Conformity assessment—General requirements for accreditation
bodies accrediting conformity assessment bodies
The following situations should also be noted:
Local water and sewage authorities’ specifications will call for certain properties of soils that
will require testing to AS 1289.
Specifications for civil engineering works will typically contain requirements that certain
tests be performed to provide a measure of assurance that the construction materials and
processes meet the design requirements. The geotechnician will need to ascertain what
tests are to be performed, other specification requirements and the frequency of sampling
and testing prior to commencement of sampling.
Specifiers, designers, specification writers and geotechnicians will also need to be aware of
a number of tests that are covered in AS 1141, Methods for sampling and testing of
aggregates as this Standard may be called up in certain job specifications (see AS 1141.0,
List of methods).
HB 160—2006 iv
CONTENTS
Page
SECTION 1 SOIL SAMPLING AND PREPARATION

1.1 GENERAL.................................................................................................................... 1
1.2 TYPES OF SAMPLES ................................................................................................. 1
1.3 SAMPLING TECHNIQUES .......................................................................................... 2
1.4 FIELD TESTS PERFORMED WHILST SAMPLING ..................................................... 3
1.5 IDENTIFICATION AND EXAMINATION OF SAMPLES ............................................... 4
1.6 SELECTION OF SITES ............................................................................................... 5
1.7 SIZE OF SAMPLES ..................................................................................................... 6
1.8 PREPARATION OF DISTURBED SAMPLES (AS 1289.1.1) ....................................... 7
1.9 STABILISATION ........................................................................................................ 15
SECTION 2 SOIL MOISTURE CONTENT

2.1 GENERAL.................................................................................................................. 16
2.2 MOISTURE CONTENT TESTS (AS 1289.2.1.1 to AS 1289.2.1.6) ............................ 16
2.3 CORRELATION OF TEST RESULTS (AS 1289.2.3.1) .............................................. 17
2.4 TOTAL SOIL SUCTION (AS 1289.2.2.1) ................................................................... 17
SECTION 3 SOIL CLASSIFICATION TESTS

3.1 GENERAL.................................................................................................................. 20
3.2 THE ATTERBERG LIMITS ........................................................................................ 23
3.3 LIQUID LIMIT............................................................................................................. 24
3.4 PLASTIC LIMIT (AS 1289.3.2.1) ................................................................................ 29
3.5 PLASTICITY INDEX (AS 1289.3.3.1)......................................................................... 30
3.6 LINEAR SHRINKAGE (AS 1289.3.4.1) ...................................................................... 31
3.7 SOIL PARTICLE DENSITY (AS 1289.3.5.1 and AS 1289.3.5.2)................................ 32
3.8 PARTICLE SIZE DISTRIBUTION .............................................................................. 33
3.9 SAND EQUIVALENT OF A SOIL (AS 1289.3.7.1) ..................................................... 40
3.10 DISPERSION............................................................................................................. 41
SECTION 4 SOIL CHEMICAL TESTS

4.1 GENERAL.................................................................................................................. 43
4.2 ORGANIC MATTER CONTENT OF A SOIL (AS 1289.4.1.1) .................................... 44
4.3 SULFATE CONTENT OF A NATURAL SOIL AND GROUNDWATER
(AS 1289.4.2.1).......................................................................................................... 45
4.4 ALKALINITY OF A SOIL ............................................................................................ 46
4.5 SOIL RESISTIVITY .................................................................................................... 49
SECTION 5 SOIL COMPACTION AND DENSITY TESTS

5.1 GENERAL.................................................................................................................. 51
5.2 DEVELOPMENT OF LABORATORY COMPACTION TEST PROCEDURES ............ 52
5.3 BACKGROUND THEORY.......................................................................................... 52
5.4 LABORATORY COMPACTION TESTS ..................................................................... 54
5.5 FIELD DENSITY TESTS............................................................................................ 58
5.6 COMPACTION CONTROL ........................................................................................ 60
SECTION 6 SOIL STRENGTH AND CONSOLIDATION TESTS

6.1 CALIFORNIA BEARING RATIO (CBR) TESTS ......................................................... 66
6.2 VANE SHEAR TESTING (AS 1289.6.2.1).................................................................. 69
6.3 STANDARD PENETRATION TEST (AS 1289.6.3.1) ................................................. 72
6.4 DYNAMIC CONE PENETROMETER (DCP) AND PERTH SAND
PENETROMETER (PSP) (AS 1289.6.3.2 and AS 1289.6.3.3) .................................. 75
6.5 TRIAXIAL TEST......................................................................................................... 82
6.6 ONE-DIMENSIONAL CONSOLIDATION TEST (AS 1289.6.6.1) ............................... 90
v HB 160—2006
Page
6.7 RESILIENT MODULUS.............................................................................................. 98
6.8 CLEGG IMPACT VALUE ........................................................................................... 99
SECTION 7 SOIL REACTIVITY TESTS

7.1 GENERAL................................................................................................................ 103
7.2 METHOD 7.1.1—SHRINK SWELL........................................................................... 103
7.3 METHODS 7.1.2 AND 7.1.3—CORE SHRINKAGE INDEX ..................................... 104
APPENDICES
A APPROXIMATE TIME REQUIREMENTS FOR TESTS ........................................... 107
B BIBLIOGRAPHY ...................................................................................................... 109
HB 160—2006
vi
NOTES
1 HB 160—2006
STANDARDS AUSTRALIA
Australian Handbook
Soils testing
S E CT ION 1 SO IL SAM PLI N G AN D

P RE PAR AT IO N
1.1 GENERAL
Sampling is one of the most critical aspects of testing of soils and should be carried out by,
or under the supervision of, persons who are experienced in the actual sampling techniques
and have a considerable knowledge of soils and the applications for which the testing is
required. When preparing site condition information, the services of an experienced
engineering geologist or geotechnical engineer will be required. In other cases, where
routine sampling of construction works are involved, experienced geotechnicians will be
able to perform and supervise sampling.

Sampling must be carried out with the utmost care and integrity; otherwise the test results
may misrepresent the characteristics of the material.
Sampling operations should be conducted by means that ensure that the samples obtained
represent, as far as possible, the true nature and condition of the soil from which they were
drawn. The procedures described in Methods 1.2.1 and 1.3.1 are designed to ensure this.
If the circumstances of any specific operation are such that the procedures described are
not followed, full details of the procedure adopted should be included in the sampling
records and the test report.
The reason for sampling should always be borne in mind. The location and number of
samples and whether they should be combined or kept separate depends not only on the
quantity of material supplied or available but whether the testing is intended as an
assessment of the material overall or its variation.
1.2 TYPES OF SAMPLES

1.2.1 Disturbed samples (AS 1289.1.2.1)
Disturbed samples are used for moisture content, classification, density and those strength
tests for which a disturbed sample is permitted.
Sampling of disturbed samples is covered by Method 1.2.1. For sampling from stockpiles
and windrows, AS 1141.3.1 is cited. The latter provides a number of photographs of
sampling techniques, which are of considerable help to the sampler.
Other methods within the AS 1289 series of tests are also cited, e.g., Method 5.4.1 for
sampling from pavements and Method 6.3.1 for sampling using an SPT sampler.
1.2.2 Undisturbed samples (AS 1289.1.3.1)
Undisturbed samples are generally required for strength, consolidation and reactivity tests,
including:
Method 2.2.1—Soil moisture content tests—Determination of the total suction of a
soil—Standard method
Method 6.1.2—Soil strength and consolidation tests—Determination of the California
Bearing Ratio of a soil—Standard laboratory method for an undisturbed specimen
Method 6.2.2—Soil strength and consolidation tests—Determination of the shear
strength of a soil—Direct shear test using a shear box
www.standards.com.au  Standards Australia

HB 160—2006 2
Method 6.4.1—Soil strength and consolidation tests—Determination of the

compressive strength of a soil—Compressive strength of a saturated specimen tested
in undrained triaxial compression without measurement of pore water pressure
compressive strength of a soil—Compressive strength of a saturated specimen tested
in undrained triaxial compression with measurement of pore water pressure
Method 6.6.1—Soil strength and consolidation tests—Determination of the one-
dimensional consolidation properties of a soil—Standard method
permeability of a soil—Constant head method
Method 7.1.1—Soil reactivity tests—Determination of the shrinkage index of a soil—
Shrink-swell index
Loaded shrinkage index
Core shrinkage index
Sampling of disturbed samples is covered by Method 1.3.1.
1.3 SAMPLING TECHNIQUES

1.3.1 Disturbed samples
Disturbed samples are usually obtained incidental to the drilling and excavation operation
(when sampling for design considerations) and consequently the structure of the natural soil
is destroyed. This is no disadvantage for general identification and those tests (e.g., particle
size distribution or compaction) in which the soil will be broken up in any case. However, to
provide any reliable information the disturbed sample must have the same composition as
the soil and come from a specific known location and depth; if mixed, this must be
deliberate and systematic.
1.3.2 Undisturbed samples
No soil sample can be completely undisturbed. An attempt should always be made to
assess what disturbance has occurred during the sampling operation and in transit to the
laboratory, and what effect this disturbance will have on results of particular laboratory
tests. In general, those tests in which the sample suffers gross distortion (e.g.,
determination of peak strength) are less sensitive to disturbance than those tests where
little distortion is caused (e.g., pre-consolidation pressure or permeability).
Disturbance during the sampling operations is not limited to gross physical distortion of the
sample. The almost inevitable release of in situ stress at some stage is a form of
disturbance that will affect the stress/strain properties to some extent, particularly with
natural unsaturated soils where irreversible changes may take place.
Not only must great care be taken in the actual sampling operation, but particular emphasis
should also be placed on the careful handling of the sample on site, during transport and in
the laboratory.
Undisturbed samples may be classified according to the degree of care taken to minimise
disturbance as follows:
(a) Simple class (representative samples). Whilst the samples must not be seriously
disturbed, primary consideration is given to the use of simple apparatus and to
avoiding delay in boring to keep costs as low as possible. Sampling at this level
would employ robust open-drive samplers, driving under impact of a sliding hammer.
 Standards Australia www.standards.com.au

3 HB 160—2006
(b) Standard class. A good quality of sample but with some attention paid to keeping
both the equipment fairly simple and the time of operation reasonably short to avoid
excessive costs. This level of sampling would employ well-designed, thin-walled open
and piston samplers, pushed not hammered, into the soil under reasonable close
supervision.
(c) Special class. Highest possible quality of sampling with less regard to cost (e.g.,
sampling for research projects or for very important or expensive foundation
structures). Techniques used would include special-purpose samplers or careful
hand-cut sampling from trial pits.
1.4 FIELD TESTS PERFORMED WHILST SAMPLING

1.4.1 Penetration tests
Cohesive soils are described in terms of strength and it is also useful to note whether
slaking occurs on immersion of air-dry material in water. Undrained shear strength can be
assessed using a pocket penetrometer for firm to very stiff soils and a hand vane for soft to
firm soils.
Non-cohesive soils are inherently difficult to assess; a number of penetration tests are
described in AS 1289, which, when used in conjunction with published tables, will assist in
the classification of such soils. Penetration tests are normally performed prior to removal of
the sample from the ground or bore hole. Penetration tests include:
(a) Standard penetration test (SPT) (AS 1289.6.3.1).
(b) 90 kg dynamic cone penetrometer test (DCP) (AS 1289.6.3.2).
(c) Perth sand penetrometer test (PSPT) (AS 1289.6.3.3).
(d) Cone penetrometer test (CPT) (AS 1289.6.5.1).
1.4.2 Field identification procedures
Whilst sampling, a number of field identification tests can be performed to assist in the
description of the soil when logging the information from a bore or excavation. These tests
are currently not included in AS 1289 and relate only to fine-grained soils or the fine fraction
of soils. These and other field identification tests are identified in AS 1726, Appendix A. The
descriptions of the soil detailed in AS 1726 need to be recorded during sampling and the
additional tests described below may be of some assistance.
(a) Dilatancy (reaction to shaking):
(i) Remove particles larger than 0.2 mm and prepare a pat of soil with a volume of
about 10 mL. Add enough water, if necessary, to make the soil soft but not
sticky.
(ii) Place the pat in the open palm of one hand and shake horizontally, striking the
hand vigorously against the other hand several times. A positive reaction
consists of the appearance of water on the pat that changes to lively
consistency and becomes glossy. When the pat is squeezed between the
fingers, the water and gloss disappear from the surface, the pat stiffens, and
finally it cracks or crumbles. The rapidity of appearance of water during shaking
and of its disappearance assists in identifying the character of fines in the soil.
Very clean sand gives the quickest and most distinct reaction whereas plastic
clay has no reaction. Inorganic silts, such as a typical rock flour, show a
moderately quick reaction.

HB 160—2006 4
(b) Dry strength (crushing characteristics):

Remove particles larger than 0.2 mm and prepare a pat of soil to the
consistency of putty, adding water if necessary. Allow the pat to dry completely
in the sun or by air-drying. Test the strength of the pat by breaking it and
crumbling it between the fingers. This strength is a measure of the character
and quantity of the colloidal fraction in the soil. The dry strength increases with
increasing plasticity.
High dry strength is characteristic of class in the clays of high plasticity (CH)
group. Typical organic silt possesses only very slight dry strength. Silty fine
sands and silts have about the same slight dry strength but can be
distinguished by feel when powdering the dry specimen. Fine sand feels gritty
whereas typical silt has the smooth feel of flour.
(c) Toughness (consistency near plastic limit):
(i) Remove particles larger than 0.2 mm and prepare a pat of soil about 10 mL in
volume to the consistency of putty. If it is too dry, add water but if it is sticky
spread the pat out in a thin layer and allow it to dry by evaporation. Roll the
specimen by hand on a smooth surface or the palm of the hand into to a thread
about 3 mm in diameter. Fold the thread and re-roll it repeatedly. When the
thread crumbles, its plastic limit has been reached.
(ii) After the thread crumbles, lump the pieces together and continue to knead the
soil until the lump crumbles.
The tougher the thread near the plastic limit and the stiffer the lump when it
finally crumbles, the more potent is the colloidal fraction in the soil. Weakness
of the thread at the plastic limit and quick loss of cohesion of the lump below
the liquid limit indicate either inorganic clay of low plasticity or materials such
as kaolin-type clay and organic clay.
Highly organic clays have a very weak and spongy feel at the plastic limit.
1.5 IDENTIFICATION AND EXAMINATION OF SAMPLES

1.5.1 Disturbed samples
Samples should be placed in suitable containers immediately after sampling for delivery to
the laboratory. If the moisture content of the material is to be measured, there should be no
delay in placing the sample in a sealed container. Sealed plastic bags are sometimes used
as containers for samples where moisture content is to be determined, but this practice
should be avoided whenever possible as moisture is lost after a few hours storage and the
moisture is often left on the sides of the bag.
The disturbed sample is usually examined immediately after excavation and its colour,
texture and other properties identified in Clause 1.4.2 noted on the sampling sheet.
The records required are detailed in AS 1289.1.2.1.
1.5.2 Undisturbed samples
The tube containing undisturbed samples should be examined at each end prior to sealing
to determine colour, texture and other properties as described in Clause 1.4.2. The field
tests performed on the soil are often performed on the material scraped from the tube when
preparing it for sealing. By pressing a finger into the end of the sample an assessment of
the stiffness of the soil can be made.
Details obtained from the examination should be noted on the sampling sheet.
The records required are detailed in AS 1289.1.3.1.

5 HB 160—2006
1.6 SELECTION OF SITES

1.6.1 General
AS 1289 provides two methods (Methods 1.4.1 and 1.4.2) that relate to random sampling.
Other than when soils have been processed to produce relatively uniform material, these
methods may not be applicable, and detailed site investigations will need to be performed.
1.6.2 Site investigations
A site investigation will generally involve a number of procedures such as—
• determination of engineering objective;
• determination of information required;
• collection and study of existing information;
• site inspection;
• topographic survey;
• terrain evaluation;
• engineering geology study; and
• geophysical surveys.
The information gathered in this process will enable a plan for direct subsurface
investigation and sampling to be established. Normally this plan includes an exploratory
phase followed by a more detailed phase of sampling and testing.
The required depth of sampling and testing will depend on the type of structure to be built
and the character and sequence of subsurface strata. Factors that influence the depth of
sampling include:
• The depth of stratum of adequate bearing capacity for the imposed loads. The
exploration should penetrate all deposits that are not suitable for foundations (e.g.,
unconsolidated fill, peat, organic silt, and very soft and compressible clay).
• The consolidation properties of soils which may have adequate strength but deform
under load.
• The possibility of holes and cavities under the structures (e.g., limestone deposits).
• The presence of rock or other sufficiently strong material and whether the thickness
of the layer of that material is adequate (i.e., not just boulders).
• The effect of moisture due to seasonal changes in water content.
At least one sample should be taken for each stratum encountered during the excavation or
boring. When design parameters are to be determined, it may be necessary to take five or
more representative samples of a stratum. When critical parameters are to be identified, a
larger number of samples may need to be taken and statistical techniques used to
determine site locations and to analyse the test results obtained from the samples.
1.6.3 Random sampling
Soil normally occurs naturally in strata. Random sample selection techniques may not be
applicable to these circumstances except when a significant number of samples is taken to
determine the properties of a particular stratum for critical parameters.
Random sampling site selection is generally used when statistical based sampling and
testing schemes are specified.
Random sampling techniques generally apply to soils that have been processed to provide
a relatively uniform mixture of soil such as mixing by road-building machinery on site or by
mixing of soils from various strata when building stockpiles of material from borough pits.
Crushed rock that is processed in a quarry is therefore often considered to be a soil, as the
test detailed in AS 1289 usually apply to the use of this type of material.

HB 160—2006 6
AS 1289 provides two methods for selection of test or sampling sites using random number
techniques. Each of these methods relies on the fact that prior to the selection of any one
site, each and every site in the related construction area or pit has an equal chance of
being selected.
The first method (Method 1.4.1) enables each site to be selected using the random number
tables provided or by use of a random number generated by a calculator. This method is
generally applicable to obtaining a single sampling or test site without bias to the presence
of particular properties or application process at the site. It is generally used for selection of
loads of soil delivered on site or for sampling from stockpiles.
One disadvantage of this method is that there is a possibility that large volumes of material
may not be selected in the process. Therefore, a second method (Method 1.4.2) is available
which relies on dividing the stockpile, or dividing the constructed area into strata, so that
samples are taken throughout the stockpile or constructed area.
The random number methods may be used for selection of sample sites for other
construction materials such as asphalt, aggregate, bricks, concrete, etc.
1.7 SIZE OF SAMPLES

The size of samples will depend on the examination and testing to be performed. In the
case of undisturbed samples taken using a boring device, the volume of material for
examination and testing is limited. When sampling disturbed samples, the amount of
material required may be estimated from the amount of material required for testing as
described in AS 1289.1.1. Generally, additional volumes of material should be sampled in
case additional or repeat testing is required.
When estimating the size of sample to be taken, consideration should be given to the
amount of material that would be excluded from the test. If a plasticity index test is to be
performed on a sample that contains only small quantities of material passing a 0.425 mm
sieve, a larger quantity of soil will be needed than if only a particle size distribution test is to
be performed.
Table 1.8.1 provides a guide to the size of sample required. Care should be taken to ensure
that there is only a single sampling exercise, as soils at each site will be different to that at
another site and operators may not have the opportunity to obtain another sample from
exactly the same site.
TABLE 1.8.1
AMOUNT OF DISTURBED SOIL SAMPLE REQUIRED FROM
EACH SAMPLING LOCATION
Mass of sample
Purpose of sample Soil type
required (kg)
Soil identification and natural Cohesive soils and sands 1

moisture content chemical tests Gravelly soils 5
Particle size distribution All soils Table 2 of AS 1289.1.1
Cohesive soils and sands 20–80
Compaction tests and CBR tests
Gravelly soils 30–80
Comprehensive examination of Cohesive soils and sands 25–50

construction materials Gravelly soils 50–100

7 HB 160—2006
1.8 PREPARATION OF DISTURBED SAMPLES (AS 1289.1.1)

1.8.1 Equipment
Balances The requirements shown in Table 1.8.1 are for equipment used in testing in
accordance with the test methods. Less accurate balances may be used when preparing
samples.
Oven In some circumstances, where air-drying can be performed, a low temperature oven
is not required. Drying in the sun needs to ensure that the temperature in the soil does not
exceed 50°C.
FIGURE 1.8.1(A) ROTARY CONE SPLITTER (EXTERIOR)

HB 160—2006 8
FIGURE 1.8.1(B) ROTARY SPLITTER (INTERNAL)
FIGURE 1.8.1(C) SAMPLE DIVIDER

9 HB 160—2006
FIGURE 1.8.1(D) SOIL GRATER
1.8.2 Preliminary preparation

In international Standards and earlier Australian Standards, initial preparation of soils was
required to be done by screening and cleaning materials over a 2.36 mm or 4.75 mm sieve.
Such fine screening may not be required if the soil is not dried out completely, thus
retaining its natural moisture. If the soil can crumble, enough moisture will be present to not
require complete breaking down to small particles. Preparation for some tests would require
further screening, e.g., plasticity tests.
This initial preparation is called for by AS 1289.5.7.1. Common practice has been to
prepare the soil over a 20 mm screen. When the ‘Hilf rapid compaction’ test was developed,
preparation was performed over a 2.36 mm sieve. It should be noted that this fine screening
causes significant loss of moisture and that coarser screening is more practical. However, it
has been observed that when the 20 mm screen was used, the resultant moulds of soil
compacted using this method were inadequate to permit proper distribution of the added
moisture or for drying out, as required.
When soil graters are used, a thin slice of cohesive soil is produced. If the slice is very thin
but wider than 10 mm, there is no need to break the slice over the 10 mm screen. It is
important that the soil grater not be used to break down discrete soil particles and that
these be screened out prior to grating.

HB 160—2006 10
1.8.3 Drying of bulk samples

It is neither necessary nor desirable to dry out the material to a completely dry condition in
the preparation stage. Milling and sieving may be done on an air-dry sample whilst splitting
is preferably done on a moist sample. Drying may be carried out in a low-temperature oven,
but temperatures in excess of 50°C must not be used, as this can change the properties of
the soil. Sunlight and forced draughts of warm air may also be used to accelerate drying.
Total drying times may frequently be shortened by first drying the bulk sample to a moist
condition, then splitting it and finally drying the reduced quantity to the air-dry state required
for milling and sieving.
Drying in the sun needs to ensure that the temperature in the soil does not exceed 50°C.
1.8.4 Quartering and splitting
1.8.4.1 General
The bulk sample should be split and quartered into representative samples for the various
tests. Material for testing should not be scooped or poured out, the use of a sample splitting
device is preferred; however, hand-quartering is acceptable if carefully performed.
1.8.4.2 Splitting bulk samples with cone-type splitter
The cone-type splitter is most convenient where a large quantity of material is to be
drastically reduced to the correct amount of representative sample required, or where a

number of representative samples of the same material are required. A typical pattern
splitter contains eight segments in the revolving drum, enabling a sample to be split into
eight parts in one pass through the splitter.
Precautions necessary to ensure efficient and accurate splitting are to—
• avoiding segregation when placing the sample in the hopper;
• not overfilling the hopper; and
• ensuring that the drum is rotating before opening the gate.
1.8.4.3 Splitting bulk samples with riffle-type splitters
The riffle-type splitter should have openings sufficiently wide to permit easy passage of the
largest particles in the sample and yet not so wide that a non-representative separation
occurs. Generally, the width of openings should be approximately 25% larger than the
largest particles in the sample to be split (see AS 1141.2).
The sample should be thoroughly mixed and spread evenly across the pan or hopper. The
pan or hopper used with a riffle-type splitter should be equal in width to the overall width of
the chutes and it should be used so that the material will pour in equal amounts into each
chute.
1.8.4.4 Hand-quartering bulk samples over 50 kg mass
The recommended procedure is:
(a) Mix and pile the sample on the quartering sheet. Shovel the material into the centre to
form a cone. In order to mix the sample, place each shovelful so that the material
spills over the cone equally in all directions. Before proceeding with the following
steps, dampen samples that tend to segregate.
(b) Flatten the cone with a shovel, spreading the material into a circular layer of uniform
thickness.
(c) Insert a stick or pipe beneath the sheet and under the centre of the pile, then lift both
ends of the stick, dividing the sample into two equal parts. Remove the stick leaving a
fold of the sheet between the divided portions.
(d) Insert the stick under the centre of the pile at right angles to the first division and
again lift both ends of the stick, dividing the sample into four parts. (Alternatively, a
shovel may be used to divide the sample into four equal parts.)

11 HB 160—2006
(e) Remove two diagonally opposite quarters, being careful to clean the fines from the
sheet.
(f) Remix the remaining material by taking alternate shovelfuls from each remaining
quarter and placing each one in the centre so that a cone is formed as before. Repeat
the quartering process until the sample is reduced to the desired size.
1.8.4.5 Hand-quartering bulk samples from 10 to 50 kg mass
(a) Pile the sample on a quartering sheet and mix by alternately lifting each corner of the
sheet and pulling it over the sample toward the diagonally opposite corner causing
the material to be rolled.
(b) Dampen material that tends to segregate.
(c) Flatten and quarter as specified in Clause 1.8.4.4(c) above.
1.8.4.6 Hand-quartering bulk samples less than 10 kg mass
(a) Place the sample on a quartering sheet or on a clean sheet of heavy paper. Mix
thoroughly with a trowel and form the material into a conical pile.
(b) Dampen material that tends to segregate.

(c) Flatten the cone by pressing it down with a trowel.
(d) Separate into quarters with the trowel and remove diagonally opposite quarters.
(e) Repeat the above process until the sample is reduced to the desired size.
(f) After the required test samples have been removed, save the remainder of the
sample for possible future check tests.
1.8.5 Preparation procedures
1.8.5.1 Removal of coatings from coarse particles and breaking up of clods
Aggregates are usually clean and can be sieved without special preparation. On the other
hand, pavement materials generally include stone particles with a coating of fines, which
must be separated from the stones by use of the rumbler by brushing the stones or by
washing the sample and evaporating the wash liquor to dryness (these alternatives are
given in their order of relative convenience). Subgrade materials generally consist of clods
with some free fines.
1.8.5.2 Preparation by rumbler
(a) Inspect the rumbler and, if necessary, clean the cage and/or pan.
(b) Weigh the air-dry bulk sample and then place in the rumbler cage.
(c) If required, add the chains or balls. (The use of chains or balls is to be avoided
whenever feasible, but where the stones in the sample are coated with clay an
inspection of the sample after 5 minutes rumbling will show whether a charge must be
used.)
(d) Tightly secure the lid to the cage.
(e) If breakdown of individual particles starts to occur or, if after a considerable period in
the machine the coatings are not being adequately removed, discontinue the machine
treatment and substitute hand-brushing or washing. (The period of time required to
process the sample in the apparatus cannot be specified but must be sufficient to
ensure that the coatings have been cleaned off the stones and that all aggregations
of fine material have been broken down.)

HB 160—2006 12
(f) After an adequate time in the rumbler, remove the sample. Add the stones retained
inside the cage to the fine material in the pan by taking off the cage lid and rotating
the cage.
(g) Remove any stones actually caught in the cage sides by manipulating or gently
knocking them back into the cage.
(h) Carefully mix the fine and coarse portions of the sample before obtaining
representative samples for each test. Alternatively, the coarse and fine fractions may
be weighed and tested separately for sieve analysis, the results being combined
arithmetically. The portion of the fine fraction not required for mechanical analysis is
used to provide material for other tests.
1.8.5.3 Preparation by hand-brushing
Where rumbler equipment is not available and the action of sieving is insufficient to clean
the stones, an alternative is to pass the bulk sample through the nest of sieves and then
brush the stones retained on each sieve (using a stiff bristle brush for hard stone and a soft
brush for soft stone).
(a) Separate the air-dry bulk sample on the 4.75 mm sieve and put aside the portion
passing while the portion retained is passed through a nest of appropriate sized
sieves.
(b) Hand-brush the individual stones retained on each sieve and break down by hand any
aggregations.
(c) Retain the coatings loosened from the stones and the aggregations that are broken
down by passing them down through the sieve nest.
(d) Add all material passing the last sieve in the nest to the material that earlier had
passed the 4.75 mm sieve.
(e) Carefully mix the fine and coarse portions of the sample before obtaining
representative samples for each test.
1.8.5.4 Preparation by washing
Preparation by washing is the lengthiest method but, as it is probably the most accurate, it
may be used as a referee method. There is no need to dry the bulk sample before the
process commences.
(a) Soak the bulk sample for a sufficient time to soften lumps and coatings and then
hand-wash to clean the individual particles and to break and disperse the lumps.
(b) Pour the whole sample through the 4.75 mm sieve, wash the fines through and collect
the wash water. In the case of large samples it will be necessary to wash the sample
in several portions.
(c) Separate out the clean particles retained on the 4.75 mm sieve and oven-dry at
105°C.
(d) Evaporate the water from the residual material at a temperature not exceeding 50°C,
stirring occasionally during the drying process to prevent the formation of hard lumps.
(e) Carefully mix the fine and coarse portions of the sample before obtaining
representative samples for each test.
1.8.5.5 Preparation by mallet
The foregoing methods of preparation are not expeditious for subgrade materials, which are
usually fine-grained and sometimes contain some larger particles. By the use of a wooden
or rubber mallet, the large clods or aggregations can be broken down to a reasonable size
for splitting and further preparation.

13 HB 160—2006
The best results are obtained when the moisture content of the bulk sample at the start of
the preparation process is such that the sample is soft and crumbly. Care should be taken
to see that clay soils are not baked hard. Oven-drying temperatures in excess of 50°C must
not be used.
(a) Process the partially dry soil with the mallet, breaking down clods and aggregations
until they can be passed through a medium-sized sieve, such as the 13.2 mm. When
dealing with samples containing soft stone, the experience of the operator will dictate
how hard and for how long the mallet is used. If stone is retained on the sieve,
remove the coatings of fines by brushing and add to the sample.
(b) Carefully inspect the cleaned stone to see whether it is ‘foreign’ to the sample (e.g.,
stone from the pavement or seal which has been included by accident), in which case
it is discarded, or it is a naturally occurring part of the sample (e.g., gravel or
mudstone in a subgrade), in which case the process is continued in the usual
manner. With subgrade samples, where the amount of stone present is so small (up
to 5% of the sample) or so large in size compared to the sizes present in the
remainder of the soil that it is judged to have no effect upon the bearing value of the
soil, it may be discarded.
(c) From the material passing the sieve, obtain representative samples for each test.
1.8.5.6 Preparation by mortar and rubber-covered pestle

Where the amount of bulk sample is small or in the milling process when a lined mill is
unavailable, the mortar and pestle may be used as an alternative to a mallet. In both cases,
the aim is to break down all aggregations and remove all surface coatings from the particles
without grinding them down.
1.8.5.7 Preparation using a soil grater
When soil graters are used, a thin slice of cohesive soil is produced. If the slice is very thin
but wider than 10 mm, there is no need to break the slice over the 10 mm screen. It is
important that the soil grater be not used to break down discrete soil particles and that
these be screened out prior to grating.
1.8.6 Size of samples and test portions
The sizes of test portions detailed in the Standard are minimum sizes. In many cases,
additional material would be required; therefore, when planning a sampling scheme, the
minimum sizes of test portions should not be used for deciding the size of samples to be
taken.
1.8.7 Preparation of material for plasticity and shrinkage tests
1.8.7.1 Preparation of normal materials
(a) Air-dry and prepare the material as in Clause 1.8.3.
(b) Obtain a representative sample passing the 4.75 mm sieve as in Clause 1.8.4.
(c) Complete the breakdown of small clods or aggregations and remove the coatings
from small stones by the use of the lined mill. Since the presence of the larger stone
chips will interfere with the milling process and will also result in damage to the lining
of the mill, run the material through the 2.36 mm sieve before milling.
(d) Carefully clean any retained stones of all surface coatings by brushing and rubbing
on the sieve; any fines obtained must be included in the milled material whilst the
clean stones are discarded.
(e) Break down any aggregations.
(f) After the mill is started, observe it for a short time to see that it is operating correctly.

HB 160—2006 14
(g) Mill the sample for about 10 minutes (some heavy clays require much longer times).
When a mill is not available, a mortar and pestle may be used. Care must be taken
that all aggregations are broken down and surface coatings removed BUT no stone is
ground down.
(h) After removal from the mill, thoroughly sieve the sample in the nest of sieves
comprising a pan, the 425 µm sieve and a guard sieve a little coarser, generally a
1.18 mm.
(i) Take the material passing the 425 µm sieve (i.e., in the pan) for the plasticity tests.
(j) Inspect the material retained on the 425 µm sieve to ensure that no more than a
nominal amount of uncrushed aggregations of clayey particles remain. If a significant
quantity of clayey particles is found still uncrushed, the material retained on the
sieves should be milled again, perhaps in a mortar and pestle. [This is particularly
important when only a small amount of material passing a 425 µm sieve has been
produced and also where the main part of the material passing 425 µm is of markedly
different nature to the aggregations.] After remilling, sieve the material and add the
rest of the material in the pan and mix well with the material obtained earlier. In all
cases the final product passing the 425 µm sieve must be thoroughly mixed. In
difficult cases, wet preparation may be used.
1.8.7.2 Wet preparation

(a) Sieve the sample on a 425 µm sieve, suitably protected by larger sieves.
(b) Retain all material passing the 425 µm sieve.
(c) Wash the material retained on the 425 µm sieve and all protecting sieves over that
sieve and collect all wash water.
(d) Evaporate the water at a temperature not exceeding 50°C, rub the dried material in a
mortar with a pestle and sieve through the 425 µm sieve.
(e) Add the material passing the 425 µm sieve to that obtained earlier, mix thoroughly
until a sufficient sample is obtained by riffling or quartering.
NOTE: An alternative wet preparation method, which has been found satisfactory for some
coarse-grained and medium-grained soils, is to use a fine water-spray for washing of the dry-
sieve-retained 425 µm sieve material, keeping the wash water to the minimum consistent with
washing the material. Add the dry-sieved material passing 425 µm sieve material to the wash
water and mix thoroughly. Some air-drying of the material may be required if there is too much
wash water, but usually it is possible to proceed with the plasticity tests after a 24 hour curing
period.
1.8.7.3 Preparation of organic soils
With certain highly organic soils, oven-drying, milling or sieving produces a considerable
alteration in the plasticity characteristics of the material, which can lead to misleading
results.
In these cases, the soil is tested in the state in which it was sampled except that it is
squeezed through a 425 µm sieve by means of a spatula to eliminate any grit. Any drying
out needed during the tests is achieved by allowing the wet soil to stand in air. In other
respects the plasticity tests are carried out as usual.
The plasticity results of highly organic soils are characterised by very high plastic limits,
irrespective of the actual plasticity index. Such soils are dark in colour, generally black, and
frequently malodorous. Certain soils have a texture that can be adequately described, in a
self-explanatory manner, as ‘bird’s nest’. An occasional soil is so fibrous in this fashion that
the step of squeezing through the 425 µm sieve has to be omitted.

15 HB 160—2006
1.9 STABILISATION
Soils and aggregates may be stabilised with binders, such as lime, cement and bitumen, to
improve the mechanical properties of the quarried and in situ pavement and earthwork
materials. Small amounts of binders mixed with soils tend to modify the material and an
increasing content of a binder, such as cement, will create a bound material with structural
stiffness. For more information on which binders are used refer to Austroads (1998(a)) and
Austroads (2002).
Stabilised materials may be tested for pavement design or construction compliance. Loose
mixed samples taken from the construction site are compacted and cured in a laboratory.
Similarly, materials extracted from the field (referred to as the parent material) are usually
dried, mixed with a binder and water added to optimise the density of the material.
To minimise the amount of laboratory testing, it is recommended that the reference OMC be
taken as the OMC of the parent material only (i.e., without binder). For heavily bound
materials it is recommended that testing be carried out in the range of 0% to +2% of the
OMC of the parent material, and for lightly bound materials at OMC of the parent material
(refer to Austroads 2002, for the definition of lightly and heavily bound materials). For
modified materials, it is recommended that the range be –1% to 0% of the OMC of the
parent material.
Unconfined compressive strength testing to AS 1141.51 allows both standard and modified
levels of compaction of the specimen. Whilst it is recognised that various regions of

Australia specify either test compaction energy, as there is no definitive guide for the
selection of standard or modified compaction for bound materials, it is recommended that
all laboratory testing be carried out using standard compaction.
When using cementitious binders, the compaction should be completed before the setting of
cemented material is well underway as compaction will reduce the chemical bonds formed
from the hydration of the cemented materials. As a guide, the compaction is carried out by
establishing the working time of the stabilised material. For more information for the
laboratory determination of working time refer to Austroads (1998a) and VicRoads (2000).
Hydrated lime and quicklime are typically used for improving the PI and strength of the soil.
The timing of the mixing of lime and quantity is a function of the lime used and the supplier.
For more information refer to AustStab (2002).
Preparation of stabilised soils is covered by AS 1141.3.3.*
* Additional tests are currently being documented by Standards Australia and will be released in the future.

HB 160—2006 16
SEC T ION 2 S O IL M O IST U R E C O NT ENT
2.1 GENERAL
This Section covers the tests and procedures set out in the three parts into which the
AS 1289.2 series is divided, as follows:
• Moisture content tests.
• Establishment of a correlation between different moisture content tests.
• Total suction of a soil.
2.2 MOISTURE CONTENT TESTS (AS 1289.2.1.1 to AS 1289.2.1.6)

2.2.1 Applications
The measured properties of a soil sample are dependent on its moisture condition at the
time of test. The moisture content test is, therefore, one of the most widely used of the
series of soil tests covered by AS 1289. It is used in tests such as moisture-density
relationship tests, field density, CBR, soil classification, and many others tests when the
moisture condition needs to be known.

Moisture content test results may also be used to determine when to commence
compaction of the soil, when to place other layers of material on top of the soil or when soil
is so saturated that moisture needs to be drained from it.
2.2.2 Limitations
The test is designed to determine the amount of free water in a soil that will affect its
engineering properties.
As the standard test operates at just above the boiling point of water, soils that contain
bound moisture, e.g., water of crystallisation, which is lost below 105°C will lose both free
water and bound water, resulting in a misleading test result. Organic soils and soils
containing gypsum are examples of materials that lose bound water at temperatures below
the boiling point of water. Such materials should be dried at a lower temperature for much
longer periods or free water extracted using vacuum desiccation.
2.2.3 Drying ovens
It has been found that a drying oven complying with AS 1289.0 and not overloaded should
dry soil to constant mass in a reasonable time—usually overnight (12–15 hours).
When an oven is being checked for performance in relation to achieving constant mass, the
weighings should be after additional drying time, at intervals of 4 hours. It is also important
that the oven be loaded evenly and that test samples at each point used be measured for
constant mass.
Irrespective of the oven used and whether it meets the requirements of AS 1289.0 or not,
constant mass should always be checked and masses recorded.
2.2.4 Balances
Although the Standard specifies balances for particular soil sizes, balances may be used for
a finer soil size if the mass of the material is equal to or greater than the minimum mass
required for the larger soil size, e.g., fine-grained soil of mass greater than 300 g may be
weighed using a balance with a limit of performance not exceeding 1 ±0.1 µg.
2.2.5 Constant mass
The interval between mass determinations should ideally be 4 hours. However, in many
circumstances, when an oven is used within its limitations, 1 hour after the oven has
recovered its operating temperature and the test portion reaches the required temperature
will be sufficient. This can take up to 2 hours prior to commencement of the additional
1 hour of drying time.

17 HB 160—2006
2.2.6 Soils that absorb atmospheric water

Earlier versions of the moisture content test required the soils be placed in containers that
could be sealed immediately they were removed from the oven. This was to ensure that the
soil did not take up moisture from the atmosphere during cooling and weighing.
Where small quantities of soil are being tested, the effect of any amounts of moisture taken
up while the soil is being cooled could have a significant effect on the result, particularly
where the atmospheric humidity is high. In such cases, the soil needs to be dried (in its
container) in a desiccator containing a suitable desiccant and the container sealed
immediately it is removed from the desiccator. Calcium chloride should not be used as a
desiccant as certain clays are known to absorb water from it.
Generally, it has been found that if the dried soil is weighed as soon as possible after it is
removed from the drying oven, the final test result is not affected.
2.2.7 Subsidiary methods
The AS 1289.2.1 series provides for a number of subsidiary methods (Methods 2.1.2
to 2.1.6) for use when rapid test results are required or when a suitable oven is not
available.
The methods are considered subsidiary and any use of moisture content in other methods
should be related by establishing a correlation between the subsidiary methods and the
standard oven-drying method, AS 1289.2.1.1.

Particular care must be taken to ensure that the soil is stirred regularly to avoid excess
heating of particles. During stirring and heating, it is important to ensure that soil particles
are not lost.
2.3 CORRELATION OF TEST RESULTS (AS 1289.2.3.1)

The subsidiary methods for moisture content all involve heating soil to temperatures
exceeding 110°C. Although the overall temperature of the soil may not exceed this level,
spots within the sample may be very much hotter than 110°C. It is possible that the soil
structure will change at the higher temperature, causing loss of bound moisture as well as
loss of the free water within the soil. Therefore, it is necessary to carry out correlation tests
with each soil to ensure that the amount of bound moisture loss is known.
When carrying out correlation tests it is important to ensure that the sample is well
seasoned prior to performing the moisture content test (either by a subsidiary method or by
the standard method), particularly if moisture is added at the laboratory. Thorough mixing is
necessary to ensure that moisture is evenly distributed throughout the soil.
When obtaining samples for correlation tests, attempts should be made to cover the soil
moisture variation expected in practice.
It is necessary to report a different uncertainty of measurement when a correlation is not
used but it may also be necessary to report the measurement of uncertainty when
correlation standard error exceeds 2.
2.4 TOTAL SOIL SUCTION (AS 1289.2.2.1)

2.4.1 Water in suction
Partially saturated reactive clays shrink and swell with changes of moisture content (see
Figure 2.4(A)). The voids are filled partly with water and partly with air. The water is actually
in a state of very high tensile stress when dry and at low levels of saturation. Soil suction is
a measure of the logarithm of that tensile stress (commonly expressed as log negative
hydraulic head in centimetres or 1 + log i(kPa)i (and given the unit pF). The drier the soil,
the more tightly the water is held and the higher the suction. On the other hand, low
suctions are associated with wetter soils. The value in the field usually ranges between
extremes of 3 pF (wet) to 5 pF (dry). The soil can be much drier than this field state. When
oven-dried, the suction of highly plastic clays may be as high as 6.5 to 6.9 pF. Lower values
are appropriate for less plastic soils.

HB 160—2006 18
5
Heavy
SOIL SUCTION (pF)

clay
4
3
Light
clay
2
1 Silt
Sand
0
0 10 20 30 40
MOISTURE CONTENT (%)
FIGURE 2.4(A) RELATIONSHIPS BETWEEN SUCTION AND MOISTURE CONTENT
2.4.2 Measurement of suction

Suction is not simple to determine. Most measurements are based on the humidity
developed in an enclosed airspace in equilibrium with the soil sample. Total soil suction and
equilibrium humidity are directly related (see Figure 2.4(B)). A sample is placed in a small
chamber and allowed to come to equilibrium and the resultant high humidity (>95%) is
measured with a precision instrument such as a thermocouple psychrometer, transistor
thermistor or dew point hygrometer. Although these instruments are commercially available,
the taking of measurements is not simple, and requires both great care and good room
temperature control to obtain repeatable results. A standard procedure is provided in
Method 2.2.1 for the dew point method of measuring suction.
2.4.3 Components of suction
Total suction measured in this way can be divided into matrix suction (largely comprised of
capillary action) and solute or osmotic suction. For the purposes of estimating ground
movements, total suction is used.

19 HB 160—2006
100
Temperature
20˚C
RELATIVE HUMIDITY (%)

Typical range
for soils
50
0
3 5 7 9
SOIL SUCTION (pF)
FIGURE 2.4(B) RELATIONSHIP BETWEEN SUCTION AND

RELATIVE HUMIDITY—USED TO MEASURE SUCTION
The theory of the psychometer for measuring soil suction is described in Fredlund and
Rahardjo (1993).
To achieve an accuracy of 10 kPa in measurement of total suction, temperature control
within ±0.001°C is required (Fredlund and Rahardjo). Method 2.2.1 has a less stringent
temperature limit but a lower accuracy of measurement.
The thermocouple junction is a particularly delicate part of the instrument and care must be
taken to prevent damage and to keep the junction clean otherwise erroneous results will
occur.
Specimens should be handled with tweezers to prevent contamination with sweat, salt, etc
from the fingers. A cork borer of suitable diameter is a useful sub-sampling tool to produce
an appropriate size disc of soil for placing in the sample chamber.
It is essential to adhere to the details listed in the test method and the manufacturer’s
instructions, including calibration requirements.
NOTE: An alternative method of measurement of suction using the filter paper technique is
presented in RTA (NSW) Test Method T 1 69. However, because the filter papers are in contact
with the soil, the matrix suction is measured, not total suction. When the filter paper is suspended
above a soil specimen, but not in contact, the total suction is measured (Fredlund and Rahardjo).

HB 160—2006 20
SECTION 3 S OIL C LASSI F I C AT I ON T ESTS
3.1 GENERAL
Soil classification places a soil in a limited number of groups on the basis of the grading
and plasticity of the disturbed sample. These characteristics are independent of the
particular condition in which the soil occurs, and disregards the influence of structure,
including fabric, of the soil, but they can give a guide to how a disturbed soil will behave
when used as a construction material, under various conditions of moisture content.
Any soil can be classified into a number of soil groups on the basis of the grading of the
constituent particles, and the plasticity of the fraction of the material passing the 425 µm
sieve. This may be done on the basis of estimation (field or rapid method) or from
laboratory tests. The classification is carried out on material normally finer than 63 mm.
Coarse material, consisting of cobbles and boulders, is picked out and its proportion, as a
whole, is estimated and recorded.
Classification tests are performed on soil samples to indicate the type of soil and the
engineering category to which it belongs. The other main tests are those that measure
strength, compression, etc., for the purpose of determining the engineering properties of the
soil.
The parameters determined from laboratory classification tests, together with visual
descriptive data of the soil, are used by geotechnical engineers for many purposes, such
as:
• Identification of different soil strata in the process of a site investigation.
• Confirmation of assumptions made from previous experience.
• Criteria for the acceptance of soil for use in construction.
Various systems for describing and classifying soils have been devised, each with its own
special emphasis or application. For the field covered by AS 1289, the Unified Soil
Classification System has the greatest application. Table 3.1 presents a guide to the
description, identification and classification of soils in accordance with this system.

21 HB 160—2006
TABLE 3.1
UNIFIED SOIL CLASSIFICATION SYSTEM DATA
FOR DESCRIPTION IDENTIFICATION AND CLASSIFICATION OF SOILS
DESCRIPTION
MAJOR DIVISIONS
GROUP GRAPHIC
TYPICAL NAME DESCRIPTIVE DATA
More than 50% of coarse grains SYMBOL SYMBOL
GW Well-graded gravels and Give typical name, indicate

gravel-sand mixtures, little or approximate percentages of sand and
are greater than 2.0 mm
GRAVEL
no fines. gravel, maximum size, angularity,

More than 50% by dry mass less than 60 mm
surface condition and hardness of the

GP Poorly graded gravels and coarse grains, local or geological
gravel-sand mixtures, little or name and other pertinent descriptive
no fines. information, symbols in parenthesis.
is greater than 0.06 mm
GRAVELLY
COARSE-GRAINED SOILS
GM Silty gravels, gravel-sand-silt For undisturbed soils add information

SOILS
mixtures. on stratification, degree of

GC Clayey gravels gravel-sand- compactness, cementation, moisture
clay mixtures. conditions and drainage
characteristics.
grains are less than 2.00 mm
SW Well-graded sands and

More than 50% of coarse
EXAMPLE:
gravelly sands, little or no
SANDS
fines. Silty sand, gravelly, about 20% hard,

angular gravel particles, 10 mm
SP Poorly graded sands and
maximum size, rounded and sub-
gravelly sands, little or no

angular sand grains coarse to fine,
fines.
about 15% non-plastic fines with low
SM Silty sand, sand-silt mixtures. dry strength, well compacted and moist
SANDY
SOILS
in place, light brown alluvial sand (SM)

SC Clayey sands, sand-clay
mixtures.
Inorganic silts, very fine Give typical name, indicate degree and
More than 50% by dry mass less than 60 mm
ML sands, rock flour, silty or character of plasticity, amount and

Liquid limit less
clayey fine sands. maximum size of coarse grains, colour

than 50%
in wet condition, odour if any, local or

Inorganic clays of low to geological name and other pertinent
FINE-GRAINED SOILS
medium plasticity, gravelly descriptive information, symbols in

is less than 0.06 mm
CL
clays, sandy clays, silty clays, parenthesis.
lean clays.
For undisturbed soil, add information
Organic silts and organic silty on structure, stratification, consistency
OL
clays of low plasticity. in undisturbed and remoulded states,
Inorganic silts, micaceous or moisture and drainage conditions.
Liquid limit more
MH diatomaceous fine sands or EXAMPLE:

than 50%
silts, elastic silts.

Clayey silt, brown, low plasticity, small
CH Inorganic clays of high percentage of fine sand, numerous
plasticity, fat clays. vertical root-holes, firm and dry in
place, fill (ML)
OH Organic clays of medium to
high plasticity.
Peat muck and other highly

Pt
organic soils.

HB 160—2006 22
TABLE 3.1 (continued)
FIELD IDENTICATION LABORATORY CLASSIFICATION
GRAVELS AND SANDS

PLASTICITY Cc =
(D30 )2
GROUP % (2) D60

OF FINE Cu = D10 D 60
NATURE DRY SYMBOL <0.06mm D10
GRADATIONS FRACTION
Use the gradation curve of material passing 60 mm for classification of fractions according to the criteria given under m ‘Major Divisions’
OF FINES STRENGTH
Wide ‘Clean’ Between 1

GOOD range in materials GW 0–5 — >4
and 3
grain size (not
Predomin enough
Determine approximate percentages of material over 60 mm in size, maximum size, shape, surface texture, hardness of
None
antly one fines to
bind Fails to comply with
POOR size or GP 0–5 —
More than 50% of the material less than 60 mm is larger than 0.06 mm
coarse above
range of
sizes grains)
material, geological description identify on estimated percentage mass of various fractions.
Fines are
‘Dirty’ None to Below ‘A’
GOOD non-plastic GM 12–50 — —
materials medium line or /p<4
TO (1)
(excess
FAIR Fines are Medium to Above ‘A’
of fines) GC 12–50 — —
plastic (1) high line or /p<7
COARSE-GRAINED SOILS
0.06 mm is about the smallest particle visible to the naked eye
Wide ‘Clean’ Between 1

GOOD range in materials SW 0–5 — >6
and 3
grain size (not
Predomin enough None

antly one fines to
bind Fails to comply with
POOR size or SP 0–5 —
coarse above
range of
sizes grains)
Fines are Below ‘A’ — —

‘Dirty’ None to
GOOD non-plastic SM 12–50 line or /p<4
materials medium
TO (1)
(excess
FAIR Fines are Medium to Above ‘A’ — —
of fines) SC 12–50
plastic (1) high line or /p<7
SLIT AND CLAY FRACTION
Fraction smaller than 0.20 mm AS sieve size
TOUGH-
DRY STRENGTH DILATANCY
NESS
mass less than 60 mm is smaller than
More than 50% of the material by dry
Quick to Below ‘A’

None to low None ML
slow line
More than 50% passing 0.06mm
None to very Above ‘A’

FINE-GRAINED SOILS
Medium to high Medium CL

slow line
Below ‘A’
0.06 mm
Low to medium Slow Low OL*

line
Low to Below ‘A’

Low to medium Slow to none MH
medium line
Above ‘A’
High to very high None High CH
line
None to very Low to Below ‘A’

High to very high OH*
slow medium line
Readily identified by colour, odour, spongy feel and generally by fibrous Pt*
texture
* Effervesces with H 2 02

23 HB 160—2006
NOTES TO TABLE 3.1:
1 The above table follows the original Unified Classification System (USBR Earth Manual) and ASTM D2487 except that it
adopts the particle size limits given in AS 1289 and other standards, viz.:
Gravel .............................................................................................................................................. 2 to 60 mm.
Sand ............................................................................................................................................. 0.06 to 2 mm.
Silt and clay ........................................................................................................................................ <0.06 mm.
The system excludes the boulder and cobble fractions of the soil and classifies only the material less than 60 mm in size.
2 As 60 mm, 2 mm and 0.06 mm sieve sizes are not normally used, the percentages passing these sizes can be obtained
from a particle size distribution curve determined from a laboratory test. Alternatively, the percentages passing may be
estimated in the field.
3 For field identification procedures for fine-grained soil or fractions see AS 1726 Appendix D, Paragraph D6.
4 Identify fines by the method given for fine-grained soils.
5 Borderline classifications occur when the percentage of fines (fraction smaller than 0.06 mm size) is greater than 5% and
less than 12%. Borderline classifications require the use of dual symbols, e.g., SP-SM GX-GC.
The soil classification tests as described in AS 1289.3 include:

• Liquid limit (LL)
• Plastic limit (PL)
• Linear shrinkage (LS)
•
Soil particle density

• Particle size distribution
• Emerson class number
• Percent dispersion
• Pinhole dispersion.
It is often desirable and sometimes necessary to supplement the visual and manual
examination of cohesive soils by means of three simple classification tests measuring the
liquid, plastic and shrinkage limits.
These limits are useful as an aid to classification though it must be emphasised that they
are measured for fully remoulded samples and will not reflect differences in natural soil
structure. They are extremely useful as indicators of the variability of conditions over a
large site, as aids to selecting representative sites and layers for undisturbed sampling and
as a means of exchanging information on soil types.
3.2 THE ATTERBERG LIMITS

The liquid limit (LL) and plastic limit (PL) are generally known as the Atterberg limits, after
the Swedish scientist Dr A. Atterberg, who first defined them for the classification of
agricultural soils in 1911. The procedures were defined more precisely by Professor
Casagrande in 1932 for engineering purposes. The mechanical device he designed for
determining the liquid limit is known as the Casagrande apparatus. The device has been
superseded in many countries by various types of cone penetrometer.
For cohesive soils there is a range of moisture contents within which the soil (silt or clay) is
of a plastic consistency. The Atterberg limits provide a means of measuring the plastic
range of a cohesive soil in numerical terms. Water can be added and mixed into a cohesive
soil until the soil becomes a slurry and behaves as a viscous fluid. This is defined as the
‘liquid’ state. As the cohesive soil is dried back slowly and evenly it begins to gain strength
and to offer resistance to deformation. This condition is known as the ‘plastic’ state. Further
reduction of moisture in the soil will cause it to shrink and become stiffer until it shows little
plastic condition and the cohesive soil becomes brittle. This is defined as the semi-solid
state. With further drying, the soil will continue to shrink until no further moisture can be
removed. At this point further drying has no more effect on volume change and the soil is
defined as being in a solid state. These four states are shown diagrammatically in
Figure 3.2.

HB 160—2006 24
PHASE
Solid Semi-solid Plastic Liquid Suspension

state state state state
Water content decreasing
Plasticity Index
Ip
Dry Shrinkage Plastic Liquid

soil limit limit limit
SL PL LL
Sticky
limit
FIGURE 3.2 PHASES OF SOIL AND THE ATTERBERG LIMITS
The change from one phase to the next does not occur at a precise time but takes place as
a transition over a period of time. In AS 1289, tests are defined to establish only two of the
three boundaries, namely the liquid limit (LL) and the plastic limit (PL). The shrinkage limit
(SL), which is the boundary between the solid and semi-solid state, can be measured, but is
not a test or limit that is generally used for classifying soils for engineering purposes in
Australia.
The moisture content range of the cohesive soil between its liquid limit (LL) and its plastic
limit (PL) is the measure of the plasticity of the soil. The higher the plasticity, the higher the
plasticity index (PI).
3.3 LIQUID LIMIT

3.3.1 General
In Britain, the cone test method is preferred to the Casagrande method. Results from the
cone method are considered more consistent, and less liable to experimental and personal
errors than those obtained from the Casagrande method; however, trials have shown that
for liquid limits up to 100% there is little difference between the results obtained by the two
methods. Above 100% the cone method tends to give slightly lower levels.
In Australia, there tends to be a greater use of the Casagrande apparatus compared to the
cone device. Use of the Casagrande apparatus has been an acceptable procedure for
determining the liquid limit but experience has shown that it is difficult to maintain the
apparatus in accordance with the Standard and the results are subject to the judgement of
the operator. For these reasons, the method using the cone penetrometer device is to be
preferred, but provided that care is taken to ensure that the Casagrande apparatus is
correctly maintained, and the test procedure is strictly followed, satisfactory results can be
obtained.
In AS 1289 a distinction is made between the liquid limit (LL) and plasticity index (PI)
determined from the cone device and that from the Casagrande device. They are termed
the cone liquid limit (WCL) and cone plasticity index (ICP).
3.3.2 Cone penetrometer (AS 1289.3.9.1 and AS 1289.3.9.2)
The liquid limit determined using a cone device is defined as the moisture content at which
the soil passes from a plastic state to the liquid state. It is based on the measurement of
penetration into the soil of a standardised cone of specified mass. At the liquid limit the
cone penetration is 20 mm. The method in the Australian Standard is based on the British
Standard.
Cone penetrometer apparatus consistence with the Australian Standard is illustrated in
Figure 3.3.2.

www.standards.com.au
25
FIGURE 3.3.2 CONE PENETROMETER DEVICE
 Standards Australia
HB 160—2006
HB 160—2006 26
The cone test can be performed either as a one-point method or a standard four-point
method. The one-point method is based on a statistical analysis of experimental data.
Greater accuracy will be achieved from the four-point method; however, if the amount of
available sample is limited or the accuracy of the determination is not critical, then the
one-point method may be acceptable. It is important when reporting the results from the
one-point method that this be clearly noted. For both methods, the general test procedure
and apparatus is the same. The difference being that for the four-point method four
separate cone tests are performed on the same sample at progressively incremental
moisture increases and the data plotted on a graph of each cone test sample moisture
content versus penetration of the cone. The cone liquid limit (WCL) is taken as the moisture
content corresponding to a cone penetration of 20 mm.
For the one-point method, the sample needs to be prepared to a moisture content such that
a cone penetration between 15 mm to 25 mm is obtained. The moisture content of the
sample tested is then multiplied by a factor depending on the plasticity range into which the
soil falls, i.e., high, medium or low. These factors are predetermined from historical
experimental data and presented in a table in the test method.
3.3.3 Casagrande methods (AS 1289.3.1.1 and AS 1289.3.1.2)
The results obtained using the Casagrande apparatus may differ slightly from those using
the cone penetrometer method but in most cases these differences will not be significant
and will be less than the normal variations likely to be obtained using the Casagrande
apparatus.
An example of the Casagrande testing apparatus is shown in Figure 3.3.3(A).
FIGURE 3.3.3(A) CASAGRANDE DEVICE AND GROOVING TOOLS

27 HB 160—2006
The height to which the cup is lifted must be exactly 10 mm above the base. This should be
regularly checked with the spacer gauge and the gap adjusted when necessary.
Figure 3.3.3(A) illustrates the curved grooving tool, which is typically used in more sandy,
silty or low plasticity soil samples where there is a risk of the sample slipping in the bowl, or
of the soil tearing on grooving with the flat tool.
Figure 3.3.3(B) illustrates the apparatus set up with a soil sample after grooving but before
testing; Figure 3.3.3(C) shows the test after the required number of blows to close groove
with a bulge, whereas Figure 3.3.3(D) shows slippage only without a bulge. Note that the
end of the grooving tool is used as a gauge of when the two divided parts of the soil come
into contact along the bottom of the groove for a distance of 10 mm.
Figure 3.3.3(E) illustrates soil as separated by grooving before test and change following
the test.
FIGURE 3.3.3(B) GROOVING SOIL BEFORE BLOWS

HB 160—2006 28
FIGURE 3.3.3(C) GROOVING SOIL AFTER BLOWS SHOWING A

‘LIQUID’ FLOW OCCURRING IN THE FORM OF A BULGE
FIGURE 3.3.3(D) GROOVING SOIL AFTER BLOWS SHOWING SLIPPAGE

IN THE CUP, WITH NO ACTUAL FLOW OR BULGE OF THE MATERIAL

29 HB 160—2006
As for the cone penetrometer method for the determination of liquid limit, there is both a
four-point and one-point method for the Casagrande procedure. The one-point method
provides a quicker means of determining the liquid limit of a soil, because only one moisture
content measurement is needed; however, the result is likely to be less reliable than either
the four-point Casagrande method or the cone penetrometer method. The one-point method
is useful when only a very limited quantity of soil is available for testing or the determination
is not critical.
Divided soil cake

before test
Soil cake after test
(Reference: Federal Works Agency, USA (1943) Principles of Highway Construction as

Applied to Airports and Other Landing Areas for Aircraft, Public Roads Administration)
FIGURE 3.3.3(E) TYPICAL SOIL ARRANGEMENT BEFORE AND AFTER TEST
3.4 PLASTIC LIMIT (AS 1289.3.2.1)

This standard method covers the determination of the plastic limit of a soil, i.e., the lowest
moisture content at which a soil is plastic; it is often referred to as the soil ‘toughness’. To
3
determine the soil’s plasticity, a moistened specimen about 10 cm in size is moulded to the
consistency of putty. If too dry, water should be added, and if sticky (too wet), the specimen
should be spread out in a thin layer and allowed to lose some moisture by evaporation. The
specimen is then rolled out by hand on a smooth frosted glass plate into a thread about
3 mm in diameter. The thread is then folded and rerolled repeatedly. During this
manipulation, the moisture content is gradually reduced, the specimen stiffens, and finally it
loses its plasticity and crumbles. At this point, its ‘plastic limit’ is reached (WP).
Soil with low plasticity usually indicates a high content of silt, while high plasticity
corresponds to high clay content.
Figure 3.4(A) illustrates how the soil threads are prepared and the distinction between the
‘too wet’ soil and soil at the ‘plastic limit’. It should be noted that with some soils it is not
possible to perform the plastic limit test, usually due to the lack of clay. Such soils are
reported as non-plastic. Figure 3.4(B) shows a sample rolled to the plastic limit, where
crumbling occurred at 3 mm.

HB 160—2006 30
Soil thread above the plastic limit
Cr umbling of soil thread below

the plastic limit
(Reference: Federal Works Agency, USA (1943) Principles of Highway Construction as

Applied to Airports and Other Landing Areas for Aircraft, Public Roads Administration)
FIGURE 3.4(A) PREPARATION OF SOIL THREADS

FIGURE 3.4(B) ROLLED SAMPLE TO PLASTIC LIMIT,

WHERE CRUMBLING OCCURRED AT 3 mm
3.5 PLASTICITY INDEX (AS 1289.3.3.1)

The plasticity index is not determined by a test; it is a measure of the plasticity of a soil as
the difference between the liquid limit (WC) and the plastic limit (WP).
In general, high plasticity soils, i.e., those with a high plasticity index, tend to resemble
plasticine or putting clay. Low plasticity soils have a chalkier feel and are less pliable to
handle. The properties of high and low plasticity soils, given in Table 3.5, provide an
indication of why delineation of plasticity may be important in engineering.

31 HB 160—2006
TABLE 3.5
GENERALISED SOIL PROPERTIES RELATED TO PLASTICITY
High plasticity soils Low plasticity soils

Usually exhibit large volume change Exhibit small volume change with
with change in moisture content changes in moisture content
Generally have very low substance May have higher substance
permeability permeability
Have high clay content Have low clay content
It should be noted that when the liquid limit of a soil is determined using the cone
penetrometer, then the plastic index is specifically recorded as the cone plasticity index
(ICP).
3.6 LINEAR SHRINKAGE (AS 1289.3.4.1)

The liquid linear shrinkage is defined as the decrease in length expressed as a percentage
of the original length when a sample of soil is oven-dried from a moisture content of about
the liquid limit. It can be used as a control medium for road construction and also as an
index of susceptibility to volume change.
Method 3.4.1 nominates a trough length of 250 mm (internal); however, when the quantity
of material under test is limited, shorter moulds may be used but they should be not less
than 100 mm in length. Proficiency tests have shown that the length of the mould can affect
the final test result.
One of the difficulties with this test is that for highly plastic soils curling and cracking of the
test specimen can occur. The degree of curling of a specimen can generally be controlled
by extremely slow drying. Alternatively, when excessive curling is expected, the specimen
may be air-dried for 24 hours and then loaded in three places in such a manner as to
prevent curling but to allow moisture loss.
Figure 3.6 illustrates the linear shrinkage apparatus used in Method 3.4.1.
FIGURE 3.6 LINEAR SHRINKAGE APPARATUS

HB 160—2006 32
3.7 SOIL PARTICLE DENSITY (AS 1289.3.5.1 and AS 1289.3.5.2)

AS 1289 includes two methods for the measurement of soil particle density, Methods 3.5.1
and 3.5.2. Method 3.5.1 covers the determination of the soil particle density of a soil using
a density bottle for the fine fraction of a soil and weighing in water for soil particles retained
on a 2.36 mm sieve. The soil particle density of the total soil is the weighted average of the
two values. Method 3.5.2 covers the determination of the soil particle density of the
combined coarse, medium and fine fractions using a vacuum pycnometer.
Density can be expressed in many different ways. Methods 3.5.1 and 3.5.2 relate to the
3
measurement of ‘mass density’, which is mass per unit volume and expressed in g/cm or
3
kg/m . It should not be confused with ‘weight density’. Weight density relates to the
gravitational force (the weight) acting on the mass, and is expressed as force per unit
3
volume (e.g., kN/m ).
Particle density is defined as the average mass per unit volume of the solid particles in a
sample of soil, where the volumes include any sealed voids contained within solid particles.
Soil consists of an accumulation of particles, which may be of a single mineral type, such as
clean quartz sand, or more usually a mixture of a number of mineral types, each with a
different particle density. For a single mineral type, the particle density of the solids
comprising a mass of the soil is that of the mineral itself (e.g., 2.65 for quartz). But for a soil
consisting of a variety of minerals, only the mean particle density of the mass as a whole is
significant; this is the sense in which the term is used here.

The particle density of most soils lies between 2.60 and 2.80, and for sands that consist
wholly or mainly of quartz it is often sufficient to assume that the particle density is 2.65.
It is rarely possible to use particle density as an index for soil classification. But knowledge
of the particle density is essential in relation to some other soil tests, especially for
calculating porosity and voids ratio, and is particularly important when compaction and
consolidation properties are considered. The particle density must also be known for the
computation of particle size analysis from a sedimentation procedure.
The second part of Method 3.5.1, involving the weighing in water for the coarser soil
particles retained on the 2.36 mm sieve, requires relatively simple weighing apparatus, and
can be used for determining the density of quite large samples. It makes use of the principle
of Archimedes for the measurement of volume. Essentially, if a solid body is wholly or
partially immersed in a liquid, the upthrust force (or buoyancy force) acting on the body is
equal to the weight force of the liquid displaced by the body, and acts vertically upwards
through the centre of gravity of the displaced liquid. Therefore, the apparent mass of a solid
body immersed in water is equal to the mass of the body less the mass of water displaced.
This is the basis of the water immersion method for measuring density.

33 HB 160—2006
FIGURE 3.7.1 THE IMMERSION IN WATER APPARATUS

USED IN AS 1289.3.5.1
Method 3.5.2 covers the determination of the soil particle density of the combined coarse,
medium and fine fractions of the soil using a vacuum pycnometer. The size of the required
pycnometer is dependent on the combined soil fractions being used. For soil samples
containing coarse, medium and fine fractions, the minimum pycnometer volume required is
5 L. If a soil contains only medium and fine fractions, the minimum pycnometer volume is
1 L. Both typical density bottles and a large pycnometer are shown in Figure 3.7.2.
FIGURE 3.7.2 DENSITY BOTTLES AND A LARGE PYCNOMETER
3.8 PARTICLE SIZE DISTRIBUTION

3.8.1 General
Soils are made up of a combination of discrete particles having a range of sizes and
shapes. The purpose of the particle size distribution tests is to group these particles into
separate ranges and sizes, and hence determine the proportions of each size range by dry
mass.

HB 160—2006 34
Method 3.6.1 covers the quantitative determination by sieving analysis of the particle size
distribution in a soil, down to the 75 µm sieve size (0.075 mm). By using this method, the
total combined silt and clay fraction (i.e., the proportion of soil particles less than 0.075 mm
in size) can be obtained by difference. For particle sizes smaller than 75 µm, the
sedimentation method as described in Method 3.6.3, using a hydrometer, is required. When
insufficient material is available for separate samples for this method to be used in
conjunction with Method 3.6.3, then Methods 3.6.2 and 3.6.3 apply.
These alternative test methods (sieving and sedimentation) are required in order to span
the very wide range of particle sizes that are encountered. The sieving method is used on
the gravel and sand size particles, which can be separated into different size ranges with a
series of sieves of standard aperture openings. For the fine silt and clay fraction sieving is
not possible, so the sedimentation procedure is used instead. The particle size limits for
defining the various fractions are given in Table 3.8.1.
TABLE 3.8.1
PARTICLE SIZE LIMITS
Equivalent AS sieve
Fraction Particle size limits
apertures for separation
Boulders >200 mm —
Cobbles 200 mm to 60 mm —
Coarse gravel 60 mm to 20 mm —
Medium gravel 20 mm to 6 mm —
Fine gravel 6 mm to 2 mm 19 mm to 6.7 mm
6.7 mm to 2.36 mm
Coarse sand 2 mm to 600 µm 63 mm to 19 mm
Medium sand 600 µm to 200 µm 2.36 mm to 600 µm
Fine sand 200 µm to 60 µm 600 µm to 212 µm
212 µm to 75 µm
Coarse silt 60 µm to 20 µm —
Medium silt 20 µm to 6 µm —
Fine silt 6 µm to 2 µm —
Clay <2 µm —
The particle size distribution is one of the most important physical characteristics of a soil; it
is the main basis for classification of a soil. Many geotechnical and geophysical properties
of a soil are related to the particle size distribution.
The particle size distribution provides a description of the soil based on a subdivision in
discrete classes of particle sizes. The size of each class can be determined by sieving
and/or sedimentation.
Sieving is the process whereby the soil is separated in particle size classes by the use of
test sieves. Sedimentation is the process of the settling of the soil particles in a liquid. The
difference in settling rate enables the particle size classes to be separated.
Depending on the purpose for the determination of the particle size distribution,
pretreatment or correction for calcium carbonate, dissolved salts and/or organic matter may
be required.
Modern methods that incorporate detection systems using x-rays, laser beams, density
measurements and particle counters can by used but are not covered under these test
methods.

35 HB 160—2006
3.8.2 Sieving method (AS 1289.3.6.1)

Test sieves of metal wire cloth or perforated metal plate together with appropriate receivers
are used. The aperture of the sieves needs to adequately cover the range for the particular
soil tested. The smallest test sieve has an aperture of 0.075 mm and the largest is usually
75 mm, although 125 mm sieves are available. The number of sieves used needs to be
sufficient to ensure that any discontinuities in the grading curve are detected.
The general procedure for sieving is shown in Figure 3.8.2(A). It essentially involves the
oven-drying of a soil sample then sieving the dried material down to 0.075 mm. The soil
fraction retained on each test sieve is then weighed. To avoid the overloading of particular
sieves, maximum limits are set and nominated in the test method depending on the sieve
diameter or area. Sieving can be undertaken by hand-shaking the sieve sets or by using a
vibrating sieving machine.

HB 160—2006 36
m
6 8
2 mm
63 µm
b
LEGEND
1 = Sample 7
2 = Riffling or quartering
3 = Drying
4 = Weighing
5 = Stirring
6 = Separating
7 = Drying
8 = Dry separation (alternative method)
9 = Sieving
10 = Weighing 10
9 m1
11 = Computation
m2
a = Dispersing agent
b = Clear water
mn
11
(Reference: ISO/TS 17892-4:2004)
FIGURE 3.8.2(A) SIEVING PROCEDURE

37 HB 160—2006
FIGURE 3.8.2(B) SOIL TESTING SIEVES
3.8.3 Hydrometer method (AS 1289.3.6.3)

The hydrometer test method uses a density hydrometer of special design to measure the
density of a soil in a suspension of water at various intervals in time. From these
measurements, the distribution of particle sizes in the silt range (75 µm to 2 µm) can be
assessed. The test is not usually performed if the soil contains less than about 10% fines.
The hydrometer is torpedo in shape and made of glass. The glass needs to be resistant to
chemicals and also be well annealed.
Calibration of the hydrometer is critical before first use. Corrections are also required to
take into account—
• volume and geometry;
• scale calibration;
• meniscus correction;
• temperature correction; and
• dispersing agent correction.
Specimen preparation is also an important element of the hydrometer test. Pretreatment of
a sample is required if organic material and/or carbonate compounds are present in
significant quantities. Dispersion of the soil is also required before testing, otherwise
particles may stick together to form aggregates and, therefore, not provide a true indication
of the particle size distribution under sedimentation. Dispersion is achieved by the inclusion
of a dispersing agent coupled with agitation of the suspension using mechanical shaker or
stirrer for a specific period.
Following pretreatment, the soil’s suspension is passed over the 0.075 mm sieve to recover
the fine fraction for the sedimentation test.

HB 160—2006 38
The general procedure for the hydrometer analysis is outlined in Figure 3.8.3(A). It involves
the soil suspension (which includes the fine soil, the dispersant solution, and water to make
up the balance to exactly 1000 mL) being placed in the sedimentation cylinder. The
suspension in the cylinder is then agitated vigorously until full suspension is obtained. The
cylinder is then placed down and a timer started. The hydrometer is then immersed in the
suspension so that it floats freely. Hydrometer readings are then taken at the upper rim of
the meniscus at specific intervals for several minutes. The hydrometer is then removed
slowly and rinsed. It is then returned to the soil suspension only at suitable intervals to
allow for subsequent readings to be taken. Readings are typically continued for a period of
up to about 24 hours. The temperature of the suspension is also recorded at the same time
as each reading to allow for subsequent temperature corrections during the calculations.

39 HB 160—2006
2
1
3 3
mw
a 4
5
6
c 7
63 µm
500 mL
500 mL
8
9 10
LEGEND
1 = Sample
2 = Quartering
3 = Weighing
4 = Drying 1000 mL
5 = Stirring
6 = Separating
7 = Drying
8 = Evaporating and decanting
9 = Sieving
10 = Weighing 11 12
11 = Agitation
12 - Sedimentation
13 = Computation
a = Dispersant
13
b = Water content
c = Evaporation
(Reference: ISO/TS 17892-4:2004)

FIGURE 3.8.3(A) GENERAL PROCEDURE FOR SEDIMENTATION/HYDROMETER
TEST

HB 160—2006 40
The test method relies on Stokes’ Law, which applies to a single sphere falling through a
large body of water. Following the application of a number of corrections, the equivalent
particle diameter corresponding to specific hydrometer readings can be calculated using
Stokes’ Law. From this the fraction of the material smaller than the corresponding
equivalent particle diameter is derived for the soil specimen tested.
The results of the particle size distribution tests are presented on a semi-logarithmic plot,
an example of which is presented in Figure 3.8.3(B). When the results are obtained by
sieving and sedimentation, the results are combined on the one graphical plot. If there is a
discontinuity between the sieving and sedimentation curves, priority is usually given to the
sieving data, because it is generally considered to be more reliable.
FIGURE 3.8.3(B) EXAMPLE OF PARTICLE SIZE DISTRIBUTION PLOT
3.9 SAND EQUIVALENT OF A SOIL (AS 1289.3.7.1)

The sand equivalent test was first introduced in California in 1950 by F. N. Hveen as a
measure of the quality and quantity of fine aggregate, particularly crushed rock.
The CRB Research Memorandum details the research work that was performed to apply
this test to crushed rock materials in Victoria. The aim of the test is to provide a quick
determination of the amount of clay-like material. This correlation is reasonably good;
however, it cannot be applied as a broad specification test but is useful when the
correlation is known for a particular material from the quarry.
The test is also useful for checking the effectiveness of washing of concrete and asphalt
sands that are combined with clay-like materials in the natural state. The value obtained will
depend on the general particle size of the sand, i.e., coarse sand will have a much higher
value than fine sand.
It is sometimes found that the clay fraction level is not easily determined due to the
flocculation of the clay materials; therefore, AS 1289.3.7.1 is not suitable for such materials.

41 HB 160—2006
3.10 DISPERSION
3.10.1 General
Dispersive soils are clayey soils that break up (slake) in contact with water and all or part of
the clay goes into colloidal suspension (disperses), creating cloudiness in the water.
The degree of dispersion is affected by the—
• mineralogy of the clay;
• chemistry of the soil moisture; and
• chemistry of the water that comes in contact with the soil.
The more dispersive a soil, the more likely it is to be subject to rill or gully erosion or
internal erosion of earth dams and similar water-retaining structures.
A crack in an earth dam constructed of dispersive soil can be eroded by the stored water
flowing through it, possibly leading to breaching of the dam.
The following three tests classify the degree of dispersion:
• Emerson class number of a soil (AS 1289.3.8.1).
• Determination of the percent dispersion of a soil (AS 1289.3.8.2).
• Determination of pinhole dispersion classification of a soil (AS 1289.3.8.3).

Since the chemistry of the water affects the dispersion (purer water results in greater
dispersion), the Emerson and pinhole test procedures allow for water other than distilled
water to be used where appropriate (e.g., reservoir water or other water with which the soil
will come in contact). The type of water to be used should be nominated by the person
requesting the test or a sample of water supplied for use in the test. In simulating erosion
caused by rainwater, distilled water is normally used.
Crouch (2000) suggests that gypsum or hydrated lime stabilisation of soils can overcome
dispersive erosion.
3.10.2 Emerson class number (AS 1289.3.8.1)
The determination of the Emerson class number test was originally devised by Emerson
(1967); the test results in the sample are classified Class 1 to Class 8.
The first step involves observation of the behaviour of air-dried crumbs of soil when placed
in water (normally distilled). Where the crumbs or aggregates slake, the soil is placed in
Class 1 to Class 6. The aggregates that do not slake but swell are Class 7 and those that
do not swell are Class 8.
Those soils whose air-dried crumbs disperse are classified as Class 1 or Class 2 for
complete or partial dispersion respectively.
For samples that do not disperse on the first step, the soil is remoulded just wet of the
plastic limit and is also immersed in water. When remoulding the soil it is important to not
contaminate the soil with salt such as sweat from the hands as this will adversely affect the
result. Soils that disperse at this step are Class 3.
If the remoulded ball does not disperse, the presence of carbonate or gypsum is
determined. Where either is present the soil is Class 4.
Where either is not present, the air-dried soil is mixed with 5 times its mass of water and
shaken in for 10 minutes. After standing for 5 minutes, the soil that has dispersed (i.e., has
a cloudy suspension) is Class 5 and the soil that completely flocculates (the soil suspension
separates with clear water above) is Class 6.
The test can be conducted using the expected or actual reservoir or stored water where
appropriate, or using distilled water.
Class 1 to Class 4 are usually considered to be dispersive, with the degree of dispersion
reducing as the class number increases.

HB 160—2006 42
Loveday (1974) describes the use of a dispersion score based on the following
subclassification of Class 2 and Class 3:
• Slight dispersion, recognised by a slight milkiness (cloudiness), immediately adjacent
to the aggregate and sometimes by a narrow edging of dispersed clay to part of the
aggregate.
• Moderate dispersion, clearly visible.
• Strong dispersion, considerable milkiness and about half of the original aggregates
dispersed away.
• Complete dispersion, leaving only sand grains in a cloud of clay.
It should be noted that Loveday (1974) extends the time for observation of dispersion to
2 hours and 20 hours, which is considerably longer than the time adopted by the Standard.
3.10.3 Percent dispersion of a soil (AS 1289.3.8.2)
The percent dispersion of a soil test determines the ratio of the soil material less than 5 µ
after limited mechanical dispersion without dispersing agents (a non-standard hydrometer
test), compared with the total soil material less than 5 µ size using the standard hydrometer
test (AS 1289.3.6.3). This test is sometimes referred to as the double hydrometer test. The
higher the ratio, the more dispersive is the soil.
The test is used to determine the risk to tunnelling or dispersive piping failure; further
details are contained in Crouch (2000). As a broad rule, soils with less than 30% dispersion
have a low risk of failure by tunnelling or piping.
3.10.4 Pinhole dispersion classification (AS 1289.3.8.3)
The pinhole test was developed by Sherrard et al (1976) during investigations of piping
failures of earth dams. The classification has been based on extensive testing and case
histories of dam failures in the USA.
The test classifies soils based on visual assessment of the outflow of water from a 1 mm
diameter hole formed in a specimen of remoulded soil. The water percolated through a soil
specimen that is dispersive will be cloudy and contain colloidal clay in suspension. The
greater the cloudiness, the greater the dispersion. Soils that are not dispersive have a
crystal clear effluent.
The specimen is prepared at about plastic limit and 95% of standard maximum dry density
(see AS 1289.5.1.1) in a cylindrical mould. A 1 mm diameter pinhole is formed through the
axial centre of the specimen. The test is conducted with water flow through the specimen at
heads varying from 50 mm to 1000 mm and over times of 5 minutes or 10 minutes. The
outflow rate is also measured. On completion, the specimen is split open and the hole size
is measured. The classification of the soil is based on the head and time at termination of
the test, the final outflow rate and the ratio of the final to initial hole diameter. The
classification ranges from D1 (highly dispersive) to ND1 (completely erosion resistant). Full
details are contained in the Standard.
The test can be conducted using distilled water or other appropriate water such as the
expected reservoir water.

43 HB 160—2006
SE C T ION 4 S O IL C H EM I C AL TESTS
4.1 GENERAL
4.1.1 Overview
In general, detailed knowledge of the chemical composition of soil is not needed in civil
engineering. The presence of certain constituents in soils may be of significant importance,
e.g., organic matter may affect the engineering characteristics of a soil; similarly, the pH
value of a soil may influence the choice of the chemical stabilisation method and affect the
corrosion potential of the construction material used.
The AS 1289.4 series on soil chemical tests is subdivided into four parts:
• Organic content (1289.4.1.1).
• Sulfate content (1289.4.2.1).
• pH value (1289.4.3.1).
• Electrical resistivity (1289.4.4.1).
These soil chemical tests are primarily intended for the determination of the organic content
and corrosion potential of a soil. It should be noted that good practice in chemical analysis
requires the testing of duplicate specimens, which applies to the above chemical tests.
Chemical tests for the presence of other substances would normally be carried out in a
specialist chemical testing laboratory by a chemist or by a technician who has been
specially trained in chemical testing procedures.
4.1.2 Corrosion
Corrosion can be defined as the deterioration of a metallic structure due to its interaction
with the surrounding environment. This phenomenon occurs through an electrochemical
process. In order for corrosion of an underground metallic structure to occur, the following
conditions must be met:
• A difference in electric potential must exit between two points on the metallic
structure (which can be caused by strains, contact with different soil types,
non-homogeneity in metal, etc.). This potential difference causes the development of
anodes and cathodes at the metal surface.
• A surrounding electrolyte, which is the moisture within the soil structure, must be
present. The moisture may contain dissolved salts from the hydroxides and metal
salts. Availability of oxygen (aeration) is also essential to the corrosion process. The
presence (or absence) of these ions determines the electrical conductivity, or
resistivity, of the electrolyte. The most corrosive soils have the least resistivity.
Under the conditions described above, metal ions will migrate from the anodic locations to
the cathodic locations. It is the loss in metal at the anodic locations that results in the
degradation of the underground metallic structure.
4.1.3 Role of soil in corrosion
Soil chemistry, particularly that of pore water, can have a substantial influence on the
corrosivity of soil. Carbonates, chlorides, sulfates and sulfides have been identified as
being particularly important. Corrosion of metals can take place in neutral or alkaline
conditions if sulfate-reducing bacteria are present (these anaerobic bacteria are
characterised by the presence of sulfides and sulfates in the soil). Aqueous solutions of
chlorides cause corrosion of iron and steel. The presence of chlorides can accelerate the
corrosion process, even in alkaline conditions.

HB 160—2006 44
Some soil types are more corrosive than others. The physical and mineralogical
composition of soils, which results from their origin, decomposition and deposition, plant
life, and topography of the land, all influence soil corrosivity potential. Silty, loamy and
clayey soils tend to be more corrosive than granular soils.
4.2 ORGANIC MATTER CONTENT OF A SOIL (AS 1289.4.1.1)

4.2.1 Overview
Many important soil properties, including absorption and retention of water, compressibility,
shear strength, and stability of soil structure are dependent to some degree on the quantity
of organic matter present. Organic matter levels are commonly highest in surface soils, but
wide variations are possible.
Highly organic soils and peats (soils of predominantly organic matter) can easily be
recognised by their colour (from dark grey or brown to almost black), and frequently by the
odour of decaying matter when inspected fresh. Nevertheless, analysis of soils for their
organic content is frequently performed in connection with various kinds of soil
investigations and in soil testing laboratories.
Carbon is the chief element of soil organic matter that can be quantified. Hence, estimates
of organic matter are frequently based on the quantitative measurement of organic carbon.
Carbon occurs in soils in the following forms of mineral and organic matter:
• Inorganic carbon in carbonate minerals of soils — occurs as various sparingly soluble
alkaline earth compounds such as CaCO 3 (calcite) and CaCO 3 MgCO 3 (dolomite). A
2− 2−
small quantity of inorganic carbon also occurs as CO 2, and HCO 3 and CO3 ions of
more soluble salts.
• Fresh organic residues of plants and animals, and living and dead microorganisms —
could be subject to rather rapid decomposition in soils under favourable conditions.
• Altered organic residues, sometimes termed ‘humus’ — represent the vast bulk of
resistant organic matter, having high adsorptive capacity for water and cations.
• Inert forms of nearly elemental organic carbon, such as charcoal, coal or graphite,
which are present occasionally in appreciable quantities in soils.
The total carbon of soils obviously includes all the above forms. The last three cover
organic carbon of soils.
4.2.2 Methods used to determine organic matter content
Organic carbon can be determined mainly by two types of methods:
• Method based on quantitative combustion procedures with a measurement of the
release of carbon dioxide to estimate carbon quantity. (For the determination of
organic carbon, carbonate is removed from soils prior to combustion to eliminate the
inorganic carbon component.)
• Method based on wet oxidation of the oxidisable matter (related to organic carbon) in
soils using chromic acid under elevated temperature. (Titration is then used to
measure the quantity of reduced chromic acid.)
Method 4.1.1 is based on the second type—a rapid titration procedure originally proposed
by Walkley and Black (1934) and subsequently modified by Walkley (1946). In this
2−
procedure, the oxidisable matter in a soil sample is oxidised by Cr 2O 7 , and the reaction is
facilitated by the heat generated when the concentrated sulfuric acid is mixed with the
2−
potassium dichromate solution. The excess Cr2 O7 is determined by titration with standard
FeSO4 solution, and the quantity of substances oxidised is calculated from the amount of
2−
Cr 2O 7 reduced.

45 HB 160—2006
With the sodium diphenylamine-sulfonate as indicator (blue-to-green colour change), the

orthophosphoric acid is used to sharpen the colour change at the endpoint. The elevated
temperature (110°C to 120°C) attained by the heat-of-dilution reaction, produced upon
addition of the sulfuric acid, is sufficient to oxidise the active forms of soil organic carbon
but not the more inert form of carbon that may be present. Walkley and Black (1934)
reported the oxidised carbon in the organic matter ranging from 60 to 80%. They
recommended a multiplying factor of 1.3 for their organic carbon estimation, which assumed
that 77% of the organic matter is oxidised in the process. This figure has been adopted in
the equation used for the calculation, as described in AS 1289.4.1.1. It is important to
understand that this figure varies slightly for different groups of soils. This limitation should
be noted when using this rapid titration method.
The method is affected by the presence of readily oxidisable substances other than carbon.
For example, chlorides, sulfides, higher oxides of manganese, and reduced iron may
interfere under certain conditions, but their effects can be eliminated or greatly reduced by
the exercise of proper precautions (Walkley 1946).
The organic carbon of a soil may be reported directly as the percentage of carbon, or
calculated as organic matter by multiplying by a conversion factor. The conventional carbon
to organic matter conversion factor (sometimes referred to as the ‘Van Bemmelen factor’) is
1.724, which is based on the assumption that soil organic matter contains an average of
58% carbon by mass; however, studies have found that it varies slightly for different groups
of soils. In Method 4.1.1, the above figure has been included in the equation used for the
calculation of organic matter from organic carbon.
This method, because of its simplicity and convenience, has been used extensively in lieu
of the quantitative combustion methods, which require more-complicated apparatus and a
longer determination time. However, because of the limitations mentioned above, the
results may vary slightly from figures obtained by other methods.
4.3 SULFATE CONTENT OF A NATURAL SOIL AND GROUNDWATER (AS 1289.4.2.1)

4.3.1 Overview
2−
Sulfate ions (SO4 ) are commonly found in natural and underground water, usually as the
result of leaching from gypsum and other mineral deposits. Naturally occurring sulfates in
soils are those of calcium, magnesium, sodium, and potassium. The presence of calcium
sulfate is fairly common in many clay soils. Sulfates may also occur as the final oxidised
2− 2− 2−
stage of sulfide ions (S-), sulfite ions (SO 3 ) and thiosulfate ions (S 2O 3 ). The SO4 levels
in Australian surface waters are generally lower than those found in other parts of the
world.
Soils with some or all of the following properties are most likely to have high sulfate
contents:
• Wet
• Fine-grained
• Black or grey colour
• High organic content
• Highly acidic or highly alkaline
Sulfate concentration may be high at sites that were formerly used for agricultural purposes,
as some fertilisers have a high sulfate content. The same is true for sites associated with
industrial wastes, as sulfates may be present in the form of gypsum plaster in many
commercial building products. Seawater also has a high concentration of sulfate (about
2300 ppm).

HB 160—2006 46
4.3.2 Engineering implications

Soluble sulfates react with certain constituents of Portland cement to form sulfoaluminate
compounds that inhibit the hardening process and cause damage to concrete. Solid salts do
not attack concrete; only when present in solution can they react with hardened cement
paste.
Soluble sulfate ions can react with Ca(OH) 2 and calcium aluminate hydrate. The products of
the reaction are gypsum and calcium sulfoaluminate (ettringite) crystals, which have
considerably greater volumes than the compounds they replace. As these crystals grow and
expand, the concrete disintegrates and becomes porous; this allows further ingress of
groundwater and attack, usually leading to corrosion of steel reinforcement.
A soil's susceptibility to sulfate attack can be evaluated by measuring the concentration of
sulfates in the soil and in the groundwater and comparing them with soils and groundwater
that have had problems with sulfate attack.
Sulfate concentrations in groundwater are commonly expressed as the number of parts by
2−
mass of SO 4 per million (ppm) or in grams per litre. The rate of attack increases with
increases in the strength of the solution. 1000 ppm is considered moderately severe and
2000 ppm is very severe. The parallel values of soluble sulfate in soil are about 0.2% and
0.5%. Soils can be classified according to their sulfate contents. Precautions and
recommendations, for various types of cement to be used for concrete in each particular
class of soil, can be found in concrete design guidelines and handbooks.

4.3.3 Method used to determine sulfate content
The total sulfate content of a soil is commonly determined by first obtaining a solution (or
extract) of the salt by the addition of dilute hydrochloric acid. The acid is added to enhance
solubility of the sulfates. The adding of acid also serves to remove carbonates that would
otherwise cause unwanted precipitation in the later ion-exchange process. For this reason,
it is advisable to wait for any effervescence to cease before proceeding with the extract.
After filtering to obtain the extract, a solution of barium chloride is slowly added; the
ion-exchange will produce a precipitate of insoluble barium sulfate. The precipitate is then
obtained by filtering. It is ignited at a high temperature and its mass is determined by the
loss of mass after ignition. The water-soluble sulfate content of soil (expressed in
percentage of soil by mass) can then be calculated based on the mass of barium sulfate
precipitated.
Similarly, the determination of sulfate in groundwater follows the above procedure using the
salt extract. Hydrochloric acid is added to groundwater samples to serve the same purpose
of carbonate removal. The concentration of sulfate in groundwater is expressed in grams
per litre.
The sulfate content of groundwater gives the best indication of the likely concentration of
sulfate in the ground, particularly that resulting from soluble sulfates. It is important to note
that the result obtained gives the sulfate content at the time of sampling only, and it is
subject to seasonal fluctuations. If possible, the groundwater should be sampled after a
long period of dry weather. Groundwater flow across a sloping site through sulfate bearing
ground results in the highest concentration on the downhill side of the site. In some cases,
the flow may continue into permeable soil deposits that are not naturally sulfate bearing.
4.4 ALKALINITY OF A SOIL

4.4.1 Overview
The extent of acidity or alkalinity of a soil is commonly expressed by its pH value.
Excessive acidity or alkalinity of the pore water in soils can have detrimental effects on
engineering structures buried in the ground. The pH level can also affect the nature of
microbiological activity.

47 HB 160—2006
Soil acidity is produced by mineral leaching, decomposition of acidic plants (for example
coniferous tree needles), industrial wastes, acid rain and certain forms of microbiological
activity. Highly acidic soils obviously represent a serious corrosion risk to common
construction materials such as steel, cast iron and those with zinc coatings. Alkaline soils
tend to have high sodium, potassium, magnesium and calcium contents. The latter two
elements may form undesirable calcareous deposits on buried structures.
In many soils, the pH encountered will be approximately near neutral (pH 5–8); however,
there may be unusual chemical conditions which would cause pH variations. Most
detrimental are low pH values (<4), which enhance metal attack. This is typical of peat
soils. Soils of high pH (>8) are not particularly aggressive to steel.
Soil pH can be determined on groundwater. If this is not possible, the soil may be made into
suspension and the pH then measured using either indicator papers or pH meters, as
described below.
The concentration specified in Method 4.3.1 (10%) must be used for consistency, as other
concentrations will change results.
4.4.2 Determination of pH using indicator papers
For quick, general determination of a solution’s pH, strips of specially impregnated paper
are often used. The indicator strips change colour when a certain pH level is reached.
Phenolphthalein is a common indicator that is colourless in solutions with pH of less than

8.3. It turns bright pink-red when pH is greater than 8.3. Although the test is simple to
perform, the results can be used only as a rough estimate of pH values. Its use in soil
testing is not supported by Australian Standards.
To test the pH of a soil:
(a) Place a quantity of the soil in a test-tube, add distilled water and shake vigorously
until all solids are in suspension.
(b) Dip a strip of the indicator paper into the water and lay it on a white tile or similar
white non-absorbent surface.
(c) After 30 seconds, compare its colour with the colour chart on the packet or dispenser.
The number on the colour that matches the test strip most closely is the pH value of
the water. Note that if the water is turbid, it is better to place a drop of water on one
side of the paper and observe the same spot on the reverse side for comparison with
the colour chart.
If the approximate pH of the water is not known to begin with, a wide-range paper should be
used first, followed by a narrow-range paper appropriate to the approximate value
indicated. It should be possible to assess the pH to 0.5 with a narrow-range paper.
4.4.3 Determination of pH by electrometric method (AS 1289.4.3.1)
4.4.3.1 Apparatus
Modern determination of pH relies upon pH meters. The electrometric determination of pH
value of a soil suspension in water, or of groundwater, may be the most accurate method
that gives a direct reading to 0.05 pH unit (some instruments have a resolution of 0.02 pH
unit).
The electrometric pH probe should be thought of as a battery whose voltage changes as the
pH of the solution in which it is inserted changes. The probe consists of two parts (some pH
measurements are made with two separate probes): first is the hydrogen-sensitive glass
bulb and second the reference electrode. The special glass of the bulb has the ability to
absorb H+ ions on the outside of the thin glass in proportion to the concentration in the bulk
solution, and the charge on the inside influences the charges on the inside surface, which
determines the electrical potential on the internal wire. The bulb is a half-cell and requires a
companion to function. A reference electrode surrounded by a solution provides the other
half cell (the most commonly used reference electrode is the saturated calomel type). This
electrode remains at a constant voltage with respect to the solution and is unaffected by
changes to the pH of the test solution. In most instruments, the voltage indicator is
calibrated to read directly in pH units.

HB 160—2006 48
The equipment is relatively expensive, and the electrode response must be calibrated
before each set of tests, and sometimes for each sample. The manufacturer’s directions
should always be consulted for specific calibration instructions. The most correct calibration
procedure is to choose buffer solutions, which bracket the anticipated pH of the sample
(see Note in Clause 4.4.3.3). The pH strips are quite useful in this situation.
4.4.3.2 Procedure
The requirements for the calibration and checks of pH meters are detailed in NATA
Technical Note 21 Laboratory pH meters—Calibration and electrode performance checks.
Details of how to perform these checks are given in Clause 4.4.3.3.
4.4.3.3 Calibration of pH meter
(a) Wash the electrodes with distilled water.
(b) Set the electrical control as necessary.
(c) Immerse the electrodes and a thermometer in the buffer solution of pH 4.0.
(d) Adjust the buffer controls so that the pH scale reads 4.00. If the temperature of the
solution is not within a range of 15°C to 25°C, correction tables or graphs may be
needed to correct the reading.
(e) Wash the electrodes with distilled water.

(f) Immerse the electrodes in the buffer solution of 9.2 pH, as a check on the reading
before use.
(g) Wash the electrodes with distilled water before using in the soil suspension.
NOTE: Buffer solutions can be prepared from powders as supplied by the manufacturer.
Alternatively, they can be made up as follows:
(a) pH 4.0 buffer solution: Dissolve 5.106 g of potassium hydrogen phthalate in distilled water
and make up to 500 mL.
(b) pH 9.2 buffer solution: Dissolve 9.54 g of sodium tetraborate (borax) in distilled water and
make up to 500 mL.
4.4.3.4 Precautions
• To avoid faulty instrument calibration, working buffer solutions should be replaced at
least once per week with commercially prepared stock buffer or freshly prepared
buffer solutions. Fresh stock buffer solutions should be prepared at least every four
weeks.
NOTE: For best results buffer solutions should be discarded after each use.
• If possible, pH meter, buffer solutions, and samples should all be at the same
temperature, because extreme temperature variations can give erroneous results.
• Care must be taken to keep the electrodes' relatively thin, delicate surfaces from
striking another surface and getting scratched, gouged or broken. After rinsing with
distilled water, carry-over/contamination may be reduced to a minimum by blotting
(never wiping!) the electrodes with a clean, non-abrasive paper or a clean cloth towel.
• Erratic results or sluggish response may result from cracked electrodes, faulty
connections, or fouling of electrodes with grease and/or solids residue. The two-point
calibration can help detect a faulty electrode. If a reasonably accurate value for the
second buffer cannot be obtained after the meter has been standardised against the
first buffer, then it is likely that one, if not both, of the electrodes is faulty and must be
replaced.
4.4.3.5 Interferences
The glass electrode is relatively free from interferences due to colour, turbidity, colloidal
matter, oxidants, reductants, or high salinity, except for a sodium error at high pH. This
error at a pH above 10 may be reduced by using special ‘low sodium error’ electrodes.

49 HB 160—2006
4.4.3.6 Precision and accuracy

The precision and accuracy attainable with any given pH meter will depend on the type and
condition of the instrument and the care used in standardisation and operation. With proper
care and use, a precision of +0.02 pH unit and an accuracy of +0.05 pH unit can be
achieved with many of the new models; however, +0.1 pH unit represents the limit of
accuracy under normal conditions. For this reason, pH values should be reported to the
nearest 0.1 pH unit.
4.5 SOIL RESISTIVITY

4.5.1 Overview
Soil resistivity has historically been used as a broad indicator of soil corrosivity. Since ionic
current flow is associated with soil corrosion reactions, high soil resistivity will arguably
slow down corrosion reactions. Soil resistivity generally decreases with increasing water
content and the concentration of ions. It is by no means the only property affecting the risk
of corrosion damage. High soil resistivity alone will not guarantee absence of serious
corrosion.
Soil resistivity surveys introduce an electrical current into the soil and measure the ease (or
difficulty) with which this current flows within the soil. Resistivity surveys respond to a
combination of soil moisture, soluble ion concentration, and soil type. Moist soils have lower
resistivity than dry soils. Fine soils (clay) have lower resistivity than coarse soils. Soils with
high salinity have low resistivity.
From experience, it has been found that soil resistivity may serve as an indicator of the
corrosivity of the soil; therefore, field soil resistivity surveys (typically by using the Wenner
four-pin method) are often conducted to obtain estimates of soil corrosivity. Typical values
of soil resistivity are shown in Table 4.5.1(A); potential corrosivity levels are shown in
Table 4.5.1(B). It must be stressed that soil resistivity may vary widely within very short
distances. Soil resistivity also changes with depth below the ground surface.
TABLE 4.5.1(A)
SOIL RESISTIVITY VALUES
Soil resistivity
Soil type
(Ohm.m)
Surface soil, loam, etc. 1–50
Clay 2–100
Sand and gravel 50–1,000
Limestone 5–4,000
Shales 5–100
Sandstone 20–2,000
Granite, basalt, etc. 1,000
Decomposed gneiss 50–500
Slate, etc. 10–100

HB 160—2006 50
TABLE 4.5.1(B)
AMOUNT OF DISTURBED SOIL SAMPLE REQUIRED
Soil resistivity
Soil corrosivity
(Ohm.m)
<5 Very corrosive
5–10 Corrosive
10–20 Moderately corrosive
20–100 Mildly corrosive
>100 Progressively less corrosive
(Reference: Roberge P.R., (1999) Handbook of Corrosion
Engineering, McGraw-Hill)
4.5.2 Method used to determine soil resistivity (AS 1289.4.4.1)

In the laboratory, the test (soil box method) is normally used for the quality control of
bedding and backfilling materials for buried metallic structures, including metal reinforcing
in soils. The method of measurement is essentially the four-pin method; however, the test
result is not likely to represent true, in situ soil resistivity for the various reasons described
above.
4.5.3 Precautions
In carrying out laboratory resistivity testing, note the following:
• Each addition of water should be dispersed evenly throughout the sample.
• If a direct current source is used in the test, measurement should be made as quickly
as possible after the current is turned on, to keep polarisation of soil box electrodes
as low as possible. The Standard specifies that readings should be taken 1 minute
after connection.
• The resistivity readings may decrease for several readings before an increase is
noted.
• The resistivity for sandy soils is generally higher than for clayey soils. Sandy soils
may contain higher levels of soluble salts and readings may not increase after having
decreased.
• For sandy soils, the reading used to calculate the resistivity value will be the point at
which total saturation occurs. This happens when water is observed rising to the
surface during compaction of the sample.

51 HB 160—2006
SE C T ION 5 S O IL C O MPAC T IO N AND

DE N SI T Y T EST S
5.1 GENERAL
Civil engineering works often involve the use of soils as foundations, fill or, in the case of
pavements, structural layers. When used in any of these circumstances it is generally
necessary to compact soil to a dense condition to achieve satisfactory engineering
properties. Compaction at the work site is usually carried out using mechanical rollers,
vibration or impact loading. Control of compaction is necessary to ensure adequate density
is achieved at a reasonable cost. This usually involves laboratory compaction and field
density tests.
Laboratory compaction tests provide the basis for control of the compaction procedures
used on site and provide the following information for soils:
• The relationship between dry density and moisture content for a given compactive
effort.
• The moisture content at which the maximum dry density is achieved for a given
compactive effort.
• The value of the maximum dry density achieved.
Field density tests provide a measure of wet density and moisture content in situ and
enable dry density at the works site to be calculated.
In most situations, the required compaction is specified as a minimum relative compaction
and not as an absolute dry density, i.e., calculation of field compaction in relation to a
laboratory-determined maximum value in terms of density ratio or density index.
Previous Austroads pavement design guides provided detailed procedures for estimating
the design moisture content (DMC) from a comprehensive field and laboratory testing
program. For practical application, however, a compromise must be reached between the
level of precision to be obtained and the cost of obtaining the necessary input data to
determine the DMC. In all cases, the designer should ensure—either on the basis of
knowledge of moisture conditions likely to occur in the locality, or by means of detailed field
investigations—that the test conditions realistically represent service conditions. In many
situations, testing under soaked conditions may be warranted.
Several combinations of design moisture contents and laboratory CBR testing conditions
that may be appropriate are presented in Table 5.1.
TABLE 5.1
TYPICAL DESIGN MOISTURE CONTENTS AND LABORATORY
CBR SOAKING REGIMES
Median annual Design moisture content/testing condition

rainfall (mm) Excellent to good drainage Fair to poor drainage
<500 OMC/unsoaked OMC/1 to 4 day soak
500–800 OMC/unsoaked OMC/4 to 7 day soak
>800 1.15 x OMC/unsoaked 1.15 x OMC/7 to 10 day soak
From Austroads Pavement Design Guide 2004

HB 160—2006 52
5.2 DEVELOPMENT OF LABORATORY COMPACTION TEST PROCEDURES

A test to provide information on the compaction characteristics of soils was introduced by
R.R. Proctor in the USA in 1933 to provide a means for controlling compaction in dams.
This test used a hand rammer and cylindrical mould and became known as the standard
‘Proctor’ compaction. AS 1289.5.1.1, using standard compactive effort, is based on this
test.
At the time of its introduction, the ‘Proctor’ test represented, in the laboratory, the degree of
compaction that could reasonably be achieved in the field with the then-current construction
equipment. As heavier compaction equipment was introduced, higher densities became
achievable in the field and the ‘modified AAHSHO’ test was introduced using the same size
mould and a heavier rammer plus more layers and hence a greater number of total blows of
the rammer. AS 1289.5.2.1, using modified compaction, is based on this test.
A test using vibration was later developed for use with cohesionless soils. Cohesionless
soils are more easily compacted with vibration and AS 1289.5.5.1 uses a vibrating table to
achieve maximum density. The test also provides for the determination of minimum density
using a pouring method.
Field densities measured on compacted soil in situ are commonly expressed as a ratio or
percentage of the maximum densities achieved by one of these standard laboratory
compaction tests for use in compaction control.
5.3 BACKGROUND THEORY

Compaction of a soil is the process by which the soil particles are packed together more
closely using mechanical means to reduce the air voids in the soil without reducing the
moisture content. This is different from the process of consolidation in which excess water
is forced out of the soil by static loading.
The presence of a limited amount of water in soil assists compaction, although at very low
moisture contents soil grains are surrounded by a thin film of water that tends to keep them
apart. As moisture content is increased, the additional water acts as a lubricant and enables
the grains to be more easily compacted together so that the air voids are reduced and the
density increased. This process continues to a point, with increasing moisture content, until
the soil grains become as closely packed as possible and maximum density is achieved for
the compactive effort used (e.g., Method 5.1.1, standard compaction). If the moisture
content is increased beyond requirements of Method 5.1.1, the excess water pushes the
soil particles apart and dry density decreases.
The change in dry density with moisture content for a given compactive effort may be
graphed to establish a moisture/dry density relationship curve (see Figure 5.3). The
moisture content at which maximum density is achieved for a given compactive effort is
termed the optimum moisture content. If a heavier compactive effort is used (e.g.,
Method 5.2.1, modified compaction), higher densities will be achieved at lower moisture
contents. The optimum moisture content will reduce and the maximum dry density will
increase. Each different compactive effort on a particular soil results in a different
compaction curve, with each curve having unique values of maximum dry density and
optimum moisture content. Construction equipment used in the field may have different
moisture/dry density relationships than either the standard or modified laboratory
compaction.
It is common to include an air voids line on a graph of a moisture/dry density relationship
curve to assist drawing and interpreting the latter (see Figure 5.3). This is a line showing
the moisture/dry density relation for soil containing a constant percentage of air voids. A set
of air voids lines can be drawn for a range of air voids; however, it is most common to draw
an approximation of the zero air voids line or the saturation line. It is impossible for a point
on a compaction curve to lie to the right of this line, i.e., with a moisture content higher than
saturation.

53 HB 160—2006
The following equation defines the zero air voids line:

ρ d = 1/(1/ρs + w/100)
where
ρ d = dry density of soil, in tonnes per cubic metre
ρ s = soil particle density, in tonnes per cubic metre
w = moisture content, expressed as a percentage of the mass of dry soil
Note that the dry density of soil specimens prepared using the AS 1289 series is calculated
using the following equation:
ρ d = 100ρ / (100 + w)
where
ρ = wet density of soil in tonnes per cubic metre (other symbols as before).

HB 160—2006 54
Modified maximum
dry density
Zero air void line
DRY DENSITY OF SOIL, t/m 3

10% 5% 0% Modified compaction
cure
2.3 Standard
maximum
dry density
Standard compaction
curve
2.2 Standard
optimum
moisture
content
2.1
Modified
DRY DENSITY OF SOIL, t/m 3
optimum
2.0 moisture
content
1.9
MOISTURE CONTENT, percent

Maximum dry
density
1.8
Compaction
curve
1.7
1.6 Optimum
moisture
content
1.5
1.4
5 10 15 20 25 30
MOISTURE CONTENT, percent
FIGURE 5.3 DEFINITIONS OF TERMS USED IN COMPACTION TESTS
5.4 LABORATORY COMPACTION TESTS

5.4.1 General
The Australian Standard laboratory compaction test procedures for soils comprise:
• AS 1289.5.1.1, Soil compaction and density tests—Determination of dry
density/moisture content relation of a soil using standard compactive effort.
• AS 1289.5.2.1, Soil compaction and density tests—Determination of dry
density/moisture content relation of a soil using modified compactive effort.
• AS 1289.5.5.1, Soil compaction and density tests—Determination of the minimum and
maximum dry density of cohesionless material—Standard method.

55 HB 160—2006
The methods basically involve application of a specified compactive effort to samples of soil
at prepared moisture contents in moulds of known volume so that the dry density of the
compacted samples can be determined. The standard and modified compactive efforts are
imparted through hand rammer or mechanical compactor. Samples are tested over a range
of moisture contents so that optimum moisture and maximum densities can be estimated
from plots of moisture and dry density. The test for cohesionless soils involves a vibrating
table and surcharge system to impart compaction and enable determination of maximum
density.
These compaction tests are relatively complex and require properly trained operators.
Typically, the man-hours required to complete testing on a single soil sample using any of
the methods will be in the range 3 to 4 hours, spread over at least two days.
Working time is defined as the time from the commencement of mixing the binder until full
compaction has been completed. The concept of working time arose from the introduction
of the construction of deep-lift in situ stabilisation. The deeper stabilised pavement layers
required additional time to achieve compaction, even with heavier vibratory rollers. It was
also identified by engineers that this form of construction would need the use of stabilising
binders that did not hydrate rapidly, such as GP cement, would limit compaction to within
2 hours.
Slow-setting binders were introduced in the early 1990s by the Australian cement industry.
Products combining cement and fly ash, lime and ground blast furnace slag were blended in
the cement works and spread on the surface in two passes to effectively mix the binder into
the pavement material. Timing of the addition of water into the second pass assisted in
maintaining the duration of hydration such that compaction equipment would not break the
chemical bonds created during the chemical reaction (i.e., hydration of cementitious
materials).
When compacting stabilised soils, care needs to be taken to ensure that compaction is
performed within the working time of the stabiliser additive used. Working time varies
according to the binder and pavement material type; typical values for slow-setting binders
range from 8 to 24 hours. Test procedures developed by RTA NSW and VicRoads allows
engineers to derive the working time in the laboratory.
5.4.2 Applicability—Oversized material
Methods 5.1.1 and 5.2.1 are applicable only to those portions of soil passing the 37.5 mm
sieve. Both tests use a 105 mm diameter mould for soils containing no more than 20%
material retained on the 19.0 mm sieve and a 152 mm diameter mould for soils with a
higher percentage retained on this sieve.
Method 5.4.1 may be used to make corrections for soils containing not more than 20% of
material (material retained on the 37.5 mm sieve on a wet basis). The AS 1289 series is not
applicable to cohesive materials containing more than 20% retained on the 37.5 mm sieve;
alternative methods of compaction quality control are required in these circumstances.
Method 5.5.1 is applicable only to cohesionless soils but does cater for materials with
maximum particle size up to 75 mm, oversized material is not therefore normally an issue.
The method is applicable to soils that may contain up to 5% by mass of material passing a
0.075 mm sieve, except that silty sands with non-plastic fines may contain up to 12%
passing the 0.075 mm sieve.
5.4.3 Apparatus
Details of the apparatus required are provided in the Standards, which should be referred to
for full specifications. The first two methods include details for hand-operated rammers but
allow use of mechanical compactors as an alternative.

HB 160—2006 56
Mechanical compactors are usually proprietary items manufactured for sale by national or
international companies. They must comply with the essential specifications of the hand
rammers in respect of matters such as drop-height and mass energy input. The use of a
well-constructed and maintained mechanical compactor has a number of advantages over
hand compaction, which requires a well-trained and skilled tester, and involves substantial
physical effort. Hand compaction is tiring and may be subject to considerable operator
variability. A mechanical compactor can eliminates tiredness as a significant factor and
provide greater accuracy of blow placement and drop control. This normally improves
precision and minimises the need for repeat testing; however, mechanical compactors are
complex items of equipment and can malfunction. Their operation should be observed
throughout compaction so that faults such as missed blows or uneven operation can be
corrected at the end of a layer cycle.
5.4.4 Procedures for standard and modified compaction (AS 1289.5.1.1 and
AS 1289.5.2.1)
Standard and modified compaction procedures essentially involve the following key steps:
(a) Sample preparation according to size fractions and division into test portions.
(b) Addition of moisture to the test portions so the range of moisture contents covered
straddles the expected optimum moisture content.
(c) Curing of test portions.

(d) Mould selection according to size fractions and mould preparation.
(e) Compaction of specimens in specified number of layers with specified number of
blows of appropriate rammer.
(f) Trimming of specimens to mould size.
(g) Determination of mass and moisture content of each specimen.
(h) Calculation of wet and dry densities, air voids line.
(i) Plotting of dry density, moisture content and air voids lines.
(j) Reporting of results.
The precision of the methods depend on accurate application of the specified compactive
effort, accurate measurement of mould volume and the mass and moisture content of the
compacted mould specimen. The test procedures are directed at ensuring precision.
A number of techniques or practices that are consistent with the specified test procedure
but not included in the method, may be used to facilitate more-efficient and accurate
determination of the moisture density relationship.
One such practice relates to moisture content selection. Wetting of the test portions so that
the highest and lowest moisture content specimens straddle the optimum moisture content
of the sample can be difficult to judge and requires experienced operators or prior
knowledge of the likely characteristic of the material. Allowance must be made for the initial
moisture before addition of water. It may be necessary to prepare extra test portions to
allow for compaction of additional specimens at higher or lower moisture contents than first
selected if a material is unknown. Use of the following moisture increments is
recommended:
• Crushed rock .......................................................................................... 0.5%–1.0%.
• Gravels ............................................................................................................... 1%.
• Sand ................................................................................................................ 1.5%.
• Clays................................................................................................................... 2%.

57 HB 160—2006
The plastic limit of fine-grained soils gives a useful approximation of their optimum moisture
content. It is preferable that the highest moisture content specimen does not exceed the
optimum moisture content excessively, as pore pressure can cause marked sample
distortion and variable results. With low plasticity high permeability soils there may also be
loss of water during compaction, while with some materials points wetter than optimum
moisture may not be achieved.
It is extremely important that the moisture of each test portion is evenly distributed
throughout to avoid variable test results. This can be achieved only by thorough mixing, and
curing of each test portion for an adequate period of time in an airtight container. Heavy
plastic bags sealed with an elastic band make a convenient airtight container. The time
required for curing depends on the nature of the soil and its moisture content prior to the
addition of water. This may vary from 2 hours for low-plasticity permeable gravels to more
than a week for heavy clays. It is preferable to cure for longer than essential than to have to
repeat tests. If the soil contains significant coarse particles, it can be advantageous to
screen and wet the fraction retained on a 4.75 mm sieve before remixing with the passing
fraction and the remaining water, to ensure the coarse particles are properly wet prior to
curing.
Compaction using a hand rammer or mechanical device is relatively straightforward but
attention must be paid to the uniform distribution of blows around the mould on each layer
to avoid variable test results. Many mechanical compactors use air pressure and a rotating
turntable. Increasing pressure beyond the specified range does not improve the speed of
operation and is unnecessary. Problems can be caused by soil lodging under the turntable;
thus it is important that the turntable be kept clean. Similarly, the rammer may slip and miss
a blow and require correction before the next layer is placed.
The thickness of layer specified must be adhered to; with clayey samples it will be
necessary to clean the face of the rammer during compaction to avoid build-up of soil,
which will interfere with the energy input. Compacting in dry, high-temperature conditions
can cause samples to dry out during compaction if care is not taken to minimise the time
taken to complete the test.
Similarly, in hot dry conditions, samples for moisture determination after compaction should
be taken quickly, and preferably include the complete specimen. This is particularly
important for well-graded materials such as crushed rock base that have a low optimum
moisture content, where a small loss of moisture mass can result in a relatively large
change in the percent moisture measured.
It is important that the curves of moisture content versus wet density are plotted carefully
and the curve between the points is representative of the points. Methods to assist plotting
curves include:
3
• Using standard scales on the axes 2 mm per 0.002 t/m for the horizontal scale. This
will assist the operator in assessing an appropriate shape of the curve for each
different type of material.
• Using a computer-plotting program that is based on a non-linear relationship,
preferably a cubic spline curve, which assists in the selection of the peak point.
• Using a plotting method as described in RTA NSW Test method TIII. The 3 points
used should include 2 points on the dry leg and 1 point on the wet leg [possibly the
midpoint of the two points on the wet side of the peak point (i.e., wetter than OMC)].
5.4.5 Procedure for cohesionless material (AS 1289.5.5.1)
Method 5.5.1 differs from the previous two methods, as it involves determination of both a
maximum and minimum dry density. It is applicable only to cohesionless soils that do not
exhibit a clear maximum density or optimum moisture content when tested in accordance
with Method 5.1.1 or 5.2.1.

HB 160—2006 58
Minimum density is measured by pouring a dry prepared sample through a funnel or

transferring with a scoop to a mould selected according to the maximum particle size in the
sample. Every care must be taken to achieve the minimum density possible. Maximum
density is measured by adding a saturated sample to a mould selected according to
maximum sample particle size, then vibrating and surcharging until maximum density is
reached. It is usually necessary to try a range of vibration amplitudes until an optimum is
found and maximum density is achieved. Heavy field compaction equipment may be able to
achieve densities that are higher than laboratory-measured values.
5.5 FIELD DENSITY TESTS

5.5.1 General
The Australian Standard methods for field density determination for soils comprise:
• AS 1289.5.3.1, Soil compaction and density tests—Determination of the field density
of a soil—Sand replacement method using a sand-cone pouring apparatus.
of a soil—Sand replacement method using a sand pouring can with or without a
volume displacer.
of a soil—Water replacement method.

and field moisture content of a soil using a nuclear surface moisture-density gauge—
Direct transmission mode.
The methods using replacement procedures are often termed direct measurement, while
the nuclear moisture-density gauge method is considered indirect.
The direct measurement replacement methods have been used since the 1930s. They are
well developed but slow and labour intensive. Typically, testing at a single site takes about
an hour in the field with laboratory calibration and calculation time involving another hour.
Testing at a series of sites normally allows more efficient use of time.
The nuclear moisture-density gauge method is relatively new. It was largely developed into
a practical tool only over the last 20 years. Density measurement essentially involves direct
transmission or backscatter attenuation of gamma radiation from a small, low-level source
by the soil being tested. Correlation relationships of direct transmission or backscatter
attenuation with materials of known density using the same source are used to arrive at an
indirect measure of wet soil density. Only the direct transmission mode of operation is used
in the AS 1289 series.
Nuclear moisture measurement relies on fast neutrons from a small, low-level source
colliding with hydrogen atoms present in the soil (largely in the form of water) and slowing
to become thermal neutrons. Measurement of the thermal neutrons in the backscatter mode
is used with a correlation relationship to indirectly determine the moisture content of the
soil. This is used to adjust the wet density so as to determine the dry density of the soil.
The nuclear moisture-density method of measurement is rapid and, in the case of
backscatter measurement, non-destructive. The gauges available commercially are
manufactured by a number of private companies and vary in source type, geometry,
detectors and other details such as on-board computing power.
Nuclear testing is usually carried out in sets on a lot-by-lot basis. Typically, a set of 9 tests
on a lot involves 1 to 2 hours in the field plus calibration and calculation time. Full
calibration of a gauge typically takes about 3 days, but is normally required only every
2 years.
More information is available in NATA’s Technical Note 32, Use of nuclear gauges for
testing soils and asphalt.

59 HB 160—2006
5.5.2 Replacement methods

Test procedures for replacement methods are provided in the AS 1289 test method series
and should be closely followed. Basically, the replacement methods involve careful
excavation of a mass of soil from a test site to the required depth, then measurement of the
volume of the hole created by replacement with sand of known pour density, or water. The
excavated soil is dried and its dry mass determined, then the density is calculated based on
this mass and the measured volume of the hole.
The methods in the AS 1289 series include replacement with sand and water and provide
alternative means of sand replacement. The sand pour methods rely on the density of the
sand poured into the excavated hole being consistent and accurately known; careful
adherence to the test procedures is necessary to ensure this is so. The water displacement
method uses a plastic sheet to ensure the water is contained in the excavated hole and not
absorbed into the surrounding material. It is most suitable for use on soils containing
appreciable proportions of cobbles or boulders (e.g., rock fill).
5.5.3 Nuclear moisture-density method (AS 1289.5.8.1)
Method 5.8.1 provides a generic description of the test procedure but does not detail the
operation of the gauge as this varies according to the manufacturer. The manufacturer’s
handbook must be followed in conjunction with the Method 5.8.1 procedure.
The equipment used includes radioactive materials that may be hazardous to health unless
proper precautions are taken. Statutory regulations covering the use and transport of
nuclear gauges must be complied with. Operators must be properly trained and all safety
measures strictly adhered to. Construction-site safety usually includes the isolation of the
test site with radiation warning signs and a vehicle-mounted amber flashing strobe light.
The Method 5.8.1 procedure is based on the direct transmission mode of operation using
the count ratio method and consists of the following key steps:
(a) Gauge calibration and operational checks.
(b) Determination of the standard counts for the gauge on the manufacturer’s reference
block in the test area.
(c) Test-site preparation including the formation of a hole to accommodate the gauge
source rod.
(d) Seating of the gauge and insertion of the source rod into the hole prepared.
(e) Determination of the test site counts for density (direct transmission).
(f) Determination of the test site counts for moisture (backscatter).
(g) Calculation of field moisture contents and dry density using the calibration equations
for the particular meter being used.
To minimise error and obtain consistent results from the gauge extra care should be taken
to—
• allow a sufficient warm-up period by switching the gauge on at least 30 minutes prior
to the commencement of testing;
• ensure the base of the gauge and the reference block are clean (brushed) before
placing the gauge on the manufacturer’s reference block;
• ensure the gauge is always placed on the manufacturer’s reference block in the
orientation indicated in the manufacturer’s handbook;
• ensure the gauge and the manufacturer’s reference block are always paired;
• minimise air gap between the test material and the base of the gauge by ensuring site
preparation is carried out in accordance with the Standard;
• minimise the air gap between the sides of the access hole and the test probe by
pulling the gauge forward towards the operator, after the probe has been inserted.

HB 160—2006 60
The accuracy of the method depends heavily on correct calibration of the gauge and
operation in accordance with the test method. The calibration procedures are detailed in
Methods 5.8.4 to 5.8.9).
Direct readings of density from the nuclear gauge may be affected in some nuclear gauges
when the difference between the displayed moisture content and the actual field moisture
content are significantly different.
Correct operation of the gauge largely involves avoiding situations that could distort
radiation counts such as other radiation sources, anomalies in background radiation, trench
or structure effects, rocks or oversized material and the presence of neutron moderators or
absorbers other than free water in the layer being measured.
The use of nuclear gauges to determine moisture content is less reliable than their use for
density determination because soils can contain variable percentages of chemically bound
water or other neutron moderators, while shallow water tables can distort results. In
addition, the volume of soil being assessed for moisture may differ from the volume being
assessed for density, and such volumes are not precisely known. When the moisture
content is measured using a gauge, the intercept of the moisture calibration equation needs
to be determined for each material being tested. Method 5.8.1 includes a procedure for this
operation. Moisture determinations by the gauge should be used only for test depths as
recommended in the manufacturer’s instructions. The most accurate operating moisture
depth is generally between 100 mm and 200 mm. It is recommended that regular
gravimetric moisture contents be taken to ensure confidence in the moisture content
indicated by the gauge. If a maximum dry density value cannot be obtained, then it is not
recommended that a moisture density gauge be used.
It is acceptable and relatively common to use alternative means of measuring in situ
moisture contents. It is important that sampling for moisture content determination by
gravimetric means be carried out on uncontaminated soil taken by excavating a hole with
vertical sides to the depth at which the probe was located during the test. The centre of the
hole should be at the centre of the gauge base.
5.6 COMPACTION CONTROL

5.6.1 Compaction criteria
5.6.1.1 General
As mentioned earlier, soil is compacted to a dense condition to achieve satisfactory
engineering properties, primarily a required strength. To achieve this, the soil is usually
compacted to a high-density ratio at or near optimum moisture content. Compaction criteria
are selected accordingly. In some applications such as clay liners for dams or land fill sites,
low hydraulic conductivity is the primary aim and a different approach is required.
In most situations, the required compaction should be specified as a minimum relative
compaction and not as an absolute dry density. The means by which such minimum relative
compaction may be specified may vary for each project. Projects involving large areas or
volumes often use ‘statistical’ acceptance criteria. This approach is also common with main
roads and highways; its use is likely to further increase as this method of quality-
acceptance testing becomes more familiar. In other instances, it is still common to adopt
no-value-to-be-less-than acceptance criteria. It should be appreciated that the minimum
value of relative compaction specified may be different for each type of acceptance criteria
for the same expected overall level of compaction.
The designer should be conscious of not over-specifying the level of compaction required,
simply for the sake of expediency. In some soils, e.g., reactive clay soils, over-compaction
may actually impair subsequent performance. Furthermore, increasing the required relative
compaction may result in substantial cost increases and possible delays in completion.

61 HB 160—2006
It should be appreciated that in clay soils, large modern compaction equipment will often
operate more effectively at moisture contents substantially lower than the optimum moisture
content determined in the standard compaction test (AS 1289.5.1.1). Many types of clays
that are placed in this manner may swell significantly and lose strength if they are wetted up
after compaction. In some circumstances, it will also be necessary to specify a range of
moisture content within which the fill is to be placed and compacted. This will be restricted
to potentially reactive soils or moisture-sensitive soils, where their subsequent engineering
performance may be dependent upon their placement moisture.
Where it is required to reduce the potential for swelling, reactive clays should be placed
close to their equilibrium moisture content (not a soil property, but dependent on the soil
and environment). In temperate climates, the equilibrium moisture content is often close to
the optimum moisture content (standard compaction). In arid and semi-arid environments,
the equilibrium moisture content may be considerably drier than in temperate climates.
Material in borrow areas at or about the depth of seasonal influence is often close to the
equilibrium moisture content and therefore the potential for problems may be minimised by
placing material directly from cut to fill.
Where clay soils are being compacted for clay liners for dams or landfill sites, it is
necessary to compact the soil so that both micro-voids and macro-voids are minimised, to
reduce hydraulic conductivity. When compacting clay soils at or near even the relatively
higher optimum moisture content of the standard compactive effort, difficulties are often
experienced when breaking down soil clods to minimise the macro-pores that ultimately
control soil hydraulic conductivity. The strength of a clay soil depends on its moisture
content and the soil clods can be more easily broken down at high-moisture contents. For
this reason, it is common for compaction criteria for this application to require the
compaction moisture content to be higher than the standard compactive effort optimum
moisture content.
5.6.1.2 Guidelines
The chemical tests of this Section covers soils that contain less than 20% by mass of
particles coarser than 37.5 mm after field compaction. Soils containing more than 20% of
particles coarser than 37.5 mm cannot be tested for density ratio using the procedures in
AS 1289. Clause 5.6.1.3 covers this situation.
Guidelines for minimum density ratio and density index values for different applications of
various projects are given in Table 5.6.1.2. These values are typical but should be varied to
suit the soil type and other requirements of a project. The values are for a specification of
the no-value-to-be-less-than type; they are not applicable where statistical control criteria
are applied. In the latter circumstances, the minimum acceptance criteria will be different
and dependent upon the scheme adopted.

HB 160—2006 62
TABLE 5.6.1.2
TYPICAL MINIMUM DENSITY RATIO CRITERIA
Density Ratio
Item Application Minimum density ratio % Minimum density index

(Cohesive soils) (Cohesionless soils)
(see Note 1) (see Note 2)
1 Residential—lot fill, house sites 95 std 70
(see Note 3)
2 Commercial—fills to support minor loadings, 98 std 75
including floor loadings of up to 20 kPa and (see Note 4)
isolated pad or strip footings to 100 kPa
3 Pavements (see Note 5)
(a) Foundation to support fill or 95 std 65
embankment
(b) Fill or embankment to support 95 std 70
pavement
(c) Subgrade (to a depth of 0.3 m) 98 std 75
(d) Sub-base courses 95 mod Not usually applicable

(e) Base course Not usually applicable
Heavily loaded 98 mod
Other 95 mod
NOTES:
1 For standard compactive effort (std) see AS 1289.5.1.1 and for modified compactive effort (mod) see
AS 1289.5.2.1.
2 Density index as a means for control of achieved density ratio may be difficult to use and interpret.
Local correlations with other methods may exist and can be used where these are well established.
3 Development on this fill will be restricted to single- and some double-storey houses, with floor slab
average imposed bearing pressure not exceeding 20 kPa, and strip or pad footings not exceeding
imposed bearing pressures of 100 kPa. Residential developments imposing higher pressures than
these are considered as commercial. A minimum dry density ratio of 98% or higher may need to be
considered if collapse on saturation or excessive settlement is likely to occur.
4 Commercial developments are likely to impose loads on fills that will have a more severe effect than
those of houses, even where contact pressures are limited to those stated. The engineer should
assess the load-carrying capacity and expected deformations associated with proposed filling and be
satisfied the fill can perform its required function. Where highly loaded fills are proposed, the
minimum relative compaction may need to be increased.
5 Where pavements will be required to carry a significant volume of heavy vehicles, the minimum
compaction criteria for the upper levels of the fill and pavement materials may need to be reviewed.
For all pavements, it is essential that the specification for compaction of subgrade materials reflect
the condition under which tests carried out for pavement thickness design are conducted.
6 Local practice in some areas for well-known materials may rely on acceptance criteria based on other
parameters, e.g., penetrometer tests, Method 6.3.2 or Method 6.3.3.
It is recognised that in some parts of Australia, such as Central Australia and parts of
Western Australia, the use of modified compactive effort is preferred, because in the natural
moisture content it more closely approximates the modified optimum rather than the
standard optimum. This Handbook does not give guidance on minimum density ratio for
modified compaction except in relation to pavements.
Notwithstanding the need for moisture control in some circumstances, it is not necessary to
routinely specify and enforce moisture control in non-critical areas, or in materials where
subsequent engineering performance is unlikely to be adversely affected by placement
moisture. This varies according to the situation.

63 HB 160—2006
In the case of fill material, moisture control may often be unnecessary; however, if
specified, the moisture content after compaction will commonly be required to be in the
range 85% to 115% of the optimum moisture content, determined in the standard
compaction test (AS 1289.5.1.1); however, the designer should assess the applicability of
this moisture range for the particular application being considered.
It is more common to specify moisture control in pavements where minimum relative
densities are high and compaction is difficult if the material is not near optimum moisture
content, determined in the modified compaction test. It is also common to require many
unbound pavement materials be dried back after compaction to ensure adequate strength.
It should be appreciated that some materials degrade rapidly during compaction. Care will
be required to avoid over-compaction of such materials, e.g., some weak or weathered
rocks. The engineer may need to consider revision of the relative compaction guidelines of
Table 5.6.1.2, to take account of the associated changing compaction characteristics. This
will be particularly important where further breakdown occurs during laboratory compaction.
NOTES:
1 Density ratio may be either dry density ratio (see AS 1289.5.4.1) of Hilf density ratio (see
AS 1289.5.7.1) as applicable. These test methods require reporting to the nearest 0.5% and
this is assumed in these values.
2 If further breakdown occurs during laboratory compaction, the resulting moisture density
relationship may not satisfactorily relate to that which existed at the time of field compaction.
As a consequence, comparison of the field dry density to the laboratory maximum dry density
(MDD) may not adequately reflect the design intent (if MDD reduces with breakdown), or may
unfairly penalise the constructor (if MDD increases with breakdown). Where such variations
are likely, testing should be carried out at the time of design and realistic requirements that
reflect the design intentions should then be specified. It is not appropriate in these
circumstances to sample material for laboratory testing before field compaction.
5.6.1.3 Coarse material
Where ripped rock or coarse material is used for filling, the after-compaction quantity of
material coarser than 37.5 mm may exceed 20%. With such material, the test procedures
for in situ determination of dry density ratio specified in the relevant parts of AS 1289 are
not applicable and special consideration should be given to alternative methods of testing
for compaction. In such circumstances, it is common to adopt a method that includes test
rolling.
A method specification will usually include required moisture conditioning, a minimum
number of passes of a given sized roller and a maximum layer thickness. Such method
specifications should be decided upon once the nature of the material to be placed is known
and after consideration of the proposed use of the fill.
5.6.2 Compaction control tests
5.6.2.1 General
Compaction control tests combine laboratory compaction tests and field density tests to
provide an effective means of assessing the quality of field compaction.
The Australian Standard methods for compaction control test comprise:
• AS 1289.5.4.1, Soil compaction and density tests—Compaction control tests—Dry
density ratio, moisture variation and moisture ratio.
• AS 1289.5.4.2, Soil compaction and density tests—Compaction control tests—
Assignment of maximum dry density ratio and optimum moisture content values.
• AS 1289.5.6.1, Soil compaction and density tests—Compaction control tests—Density
index method for a cohesionless material.
• AS 1298.5.7.1, Soil compaction and density tests—Compaction control tests—Hilf
density ratio and Hilf moisture variation (rapid method).

HB 160—2006 64
The procedures rely on testing in accordance with the laboratory compaction and field
density/moisture tests discussed earlier and are essentially calculation methods.
The selection of the most appropriate laboratory compaction test to be used as a basis for
assessing relative density will depend on the project and factors such as soil type, layer
and the service loading concerned. It is usual to select a laboratory compaction test such
that relative densities achieved using best practice construction techniques are generally
not greater than 100%.
5.6.2.2 Standard density and moisture ratio method (AS 1289.5.4.1)
Method 5.4.1 deals with the calculation and reporting of density ratio, moisture variation
and moisture ratio. These parameters are respectively the ratio (as a percentage) of field
dry density to maximum dry density, the difference between field moisture and optimum
moisture and the ratio (as a percentage) of field moisture to optimum moisture. The
calculations are detailed in the test method and allow use of assigned or determined values
of maximum dry density and optimum moisture content. They include a procedure to deal
with up to 20% material retained on the 37.5 mm sieve.
An important practical element of this procedure is sample selection. Except where an
assigned value is allowed, the standard procedure for compaction control requires sampling
at or around the site of every field density test, and that a maximum dry density is to be
determined for each field density determination. Some specifications may require sampling
from a stockpile or test area, prior to compaction when, it has been established that the
influences of field compaction do not affect the value of the laboratory reference density.
If in one-to-one testing assigned values or statistical procedures are not used the
uncertainty of measurement of the resultant density ratio will increase considerably, and will
affect the final result in relation to the specification.
5.6.2.3 Assigned value method (AS 1289.5.4.2)
Method 5.4.2 provides a means of reducing testing requirements by assigning values to
maximum dry density and optimum moisture content. The method is generally applicable
only to very uniform materials such as crushed rock base that has been produced under
controlled conditions.
The method is a statistical one and is based on taking six random samples from a number
of days’ production and determining the maximum dry density and optimum moisture
content of each. The mean of each parameter is calculated and, provided no value exceeds
specified limits, accepted as the assigned value. The assigned values are monitored and
updated by new test results until any individual value falls outside the specified limits. If this
occurs, a new assigned value must be developed.
The reliability of the method depends on active ongoing monitoring of the assigned values
and good process control by the producer.
5.6.2.4 Statistical method (AS 1289.5.4.3)
Method 5.4.3 also provides a means of reducing testing requirements by using standard
deviation limits for maximum dry density and optimum moisture content variations.
Additionally, this method also allows for pre- and post-compaction sampling for determining
maximum dry density and optimum moisture content.
This statistical method allows for the ratio of maximum dry densities to field densities to be
reduced from 1:1 to 1:3, provided all samples fit within the standard deviation limits for both
maximum dry densities and optimum moisture content determinations, as per Table 5.6.2.4.
Values for the standard deviations for both maximum dry densities and optimum moisture
content that fall outside the prescribed limits means automatic reversion to Method 5.4.1
(one-to-one testing).

65 HB 160—2006
TABLE 5.6.2.4
LIMITING VALUES OF STANDARD DEVIATION
3
Component Max. standard deviation (t/m )
Basecourse 0.030
Sub-base 0.045
Pre-compaction sampling can help achieve faster turnaround times for results, where it is
certain that the material does not break down further during the compaction process.
Generally, it is necessary to prove there is no material breakdown.
The reliability of this procedure and its statistical base is very dependent on initial lot
selection and the monitoring of material within the standard deviation limits.
5.6.2.5 Hilf (rapid) method (AS 1289.5.7.1)
Method 5.7.1 is a rapid method of determining compaction control parameters for soils. The
method is based on a correlation between wet density and optimum moisture content based
on experimental data collected by Hilf (1966). This enables a laboratory-determined
relationship between wet density and added moisture content to be used to derive a
moisture variation without the need to determine moisture contents.
The Hilf rapid method involves sampling the soil concerned at the time of determining the
field wet density and using the wet sample to compact specimens in the usual compaction
moulds. Samples are compacted using standard compactive effort at the field moisture
content, with added or removed water, and the wet densities determined. Converted wet
densities are calculated based on the added percentage of water and plotted against the
moisture increments added. The peak converted wet density is used to calculate the Hilf
density ratio, i.e., the ratio of peak converted wet density to the field wet density. The added
moisture at the peak converted wet density is corrected according to the Hilf correlation and
used to calculate the Hilf moisture variation.
The method is not as accurate as the standard method of compaction control because it is
based on a correlation and is not necessarily applicable to all soils or site conditions. It has
a number of limitations, viz.:
• The test is based on a standard compactive effort and experimental data must be
accumulated for the particular soil concerned if modified compactive effort is used.
• The method may not be accurate for soils that are much wetter or drier than optimum
moisture content because of the omission of proper curing.
• The method is limited to added moisture contents between –4% and +6%.
• The Hilf density ratio is a wet density ratio.
5.6.2.6 Density index method (AS 1289.5.6.1)
Method 5.6.1 is applicable to cohesionless soils where maximum and minimum densities
have been determined in accordance with Method 5.5.1. The density index is derived from
the ratio (as a percentage) of the difference between the maximum laboratory voids and the
field voids, and the difference between the maximum laboratory voids and the minimum
laboratory voids. This is calculated using an equation based on maximum and minimum
laboratory dry densities and field dry density. The method does not allow use of assigned
values.
Wherever possible, consideration should be given to determining maximum dry density
using Method 5.1.1 or 5.2.1 as these methods relate only to cohesionless materials such as
clean sands.

HB 160—2006 66
S E C T ION 6 SO IL ST REN G T H AN D
C ONS OL I D AT I O N TEST S
6.1 CALIFORNIA BEARING RATIO (CBR) TESTS

6.1.1 Overview
The CBR test is usually used as a design parameter for designing the thickness of road and
airport pavements (see Austroads Pavement Guide). It was introduced during World War II
for use in airport and major road pavement designs and has since been adopted throughout
the world for this purpose.
This design procedure has been proven to be relatively suitable but the test results need
careful interpretation by qualified pavement designers.
Some specifications require minimum CBR values of the specified field density ratio (e.g.,
CBR at 95% density ratio using standard compactive effort or CBR at 98% density ratio
using modified compactive effort). Due to difficulties in compacting test specimens to other
than 100% density ratio (i.e., 53 blows/layer) the variability of resulting CBR tests is
extremely high.
Measurement uncertainty estimates indicate that moisture content and density ratio affect
the test result significantly. Therefore, it is recommended that, if it is necessary to specify
the CBR to determine if suitable fill or new material is being supplied, the value of CBR at
100% density ratio be specified. If the design is at lower density ratio values, the CBR at
100% should be determined at the same time as the design values are determined.
The AS 1289.6 series is divided into three distinct parts:
• AS 1289.6.1.1, CBR—Remoulded specimen.
• AS 1289.6.1.2, CBR—Undisturbed specimen.
• AS 1289.6.1.3, CBR—Field in place method.
6.1.2 Determination from a remoulded specimen (AS 1289.6.1.1)
6.1.2.1 General
The CBR properties of a soil sample are usually measured in the laboratory on a remoulded
specimen. Samples are compacted dynamically using either standard or modified
compactive effort as defined in Methods 5.1.1 and 5.2.1. Vibratory compaction may be
substituted though this must be reported. The test method usually applies to both fine-
grained and medium-grained soils and, in particular, all soils with less than 5% passing the
19 mm sieve (see AS 1289.0 and Clause 6.1.2.2).
In addition to the above, the compaction regime is often specified to comply with a required
measured laboratory moisture ratio or laboratory density ratio. The values of these ratios
can be determined from the MDD/OMC graph calculated in accordance with Methods 5.1.1
and 5.2.1. Initial sample preparation and the seasoning to a particular moisture content is,
therefore, a critical function.
6.1.2.2 Preparation
It is important in the test to ensure that material used in both the initial compaction testing
and the CBR test is of a size passing the 19 mm sieve, and also that no material used in
the CBR test has been previously compacted.
NOTE: In some instances other preparation regimes, such as cyclic compaction and the use of
special additives (e.g., cement, lime) may be a client requirement.

67 HB 160—2006
It has been found that test results that do not match field performance can be obtained
when materials that break down readily under field compaction are used. In particular,
siltstone and mudstone soils that contain weakly cemented larger particles, which break
down into silty materials, will have significant lower CBR values after field compaction.
These materials should be broken down by repeated compaction prior to compacting in a
mould for CBR testing.
During the preparation of the sample, both in the mixing, curing and seasoning stages and
their storage, it is important that sealed containers be used. Knowledge of this process and
the times required to fully cure the specimens can be gained from the initial compaction
test. The mass required for the compaction of the CBR specimen to a measured density
can be calculated for the required number of layers. This sample will be subjected to a
variable number of blows to ensure the depth of each layer is even; the number of blows is
recorded and reported.
6.1.2.3 Compaction of test specimens to a specified density
Method 6.1.1 requires that 53 blows per layer be applied to achieve 100% maximum dry
density at optimum moisture content. When compaction to a specified density ratio is
required, the best results are obtained by compacting three specimens using 56, 25 and 10
blows per layer. The soaking and penetration test should then be applied to each specimen.
CBR is then plotted against the moulded dry density and the required CBR is read from the
plot at the desired density ratio.

Occasionally, varying the added moisture at compaction is specified to achieve the required
density ratio. This method can produce significant errors in the result as CBR is highly
dependent on moulding moisture content. Even though Note 7 to Method 6.1.1 limits
moisture for the normal test to ±1% moisture ratio from optimum moisture content, it is
recommended that it be held to ±0.5%.
In such cases, it is important that—
• the compaction is applied by allowing the hammer to fall over the total drop for each
blow;
• the number of blows per layer is uniform;
• the energy is applied evenly through the layers by ensuring the depth of layer
specified is adhered to; and
• the number of blows per layer be recorded and reported.
6.1.2.4 Swell measurement
For swell measurement, it is particularly important to utilise a designated laboratory ‘setting
piece’, on which to place the tripod and measuring gauge, both before and after measuring
the stem and perforated plate to ensure the dial gauge has not been altered. Whenever the
tripod and gauge are placed on the side of the mould, and a reading is recorded, it is also
important to mark the position on the mould so that no erroneous measurements are
recorded, particularly if the swell gauge is removed during soaking. The swell of samples
must be measured in the same place both before and after soaking.
6.1.2.5 Surcharges
The general mass of the added surcharges, used in both the soaking and testing of the
specimens, relates to the pavement at the shoulder at its weakest point, and is as stated in
the test method being usually 4.5 kg, which is also the value to be used when the amount of
the surcharge is not specified by the designer. In certain situations, the designer may
specify additional surcharges as shown in Figure 7 at the method when it is warranted, e.g.,
airport pavements. This is in addition to the mass of the stem and perforated plate (usually
around 1 kg). Therefore, the mass of the stem and perforated plate must be added to the
surcharge used when the specimen is tested. In this respect, when the standard surcharge
mass is used, during soaking, it is advised that when testing the specimen, a surcharge
mass, or 5.5 kg steel annulus is used instead of the surcharge masses.

HB 160—2006 68
The high swell materials referred to in the Austroads Pavement Design Guide are based on
the use of a single surcharge of 4.5 kg. If additional surcharges are used, erroneous
classification of high-swell materials will occur.
6.1.2.6 Penetration piston
When performing the penetration test, it is particularly important to mount the measuring
gauge for displacement measurement in reference to the mould and separate from the load
ring. The displacement gauge can be mounted on a ring attached to a piston, or a clamped
rod attached to the mould or baseplate. The test should be undertaken until a penetration
distance of 7.5 mm has been reached.
6.1.3 Determination in the field (AS 1289.6.1.2 and AS 1289.6.1.3)
The CBR of in situ material, either compacted in the field or in the natural state, can be
determined by either Method 6.1.2 or 6.1.3; however, in practice, penetration testing in
accordance with the simple procedure in Method 6.5.1 or 6.3.2 is commonly used in
surveying the variation in the bearing capacity of soils (see Scala (1956)).
This disadvantage of using the measurement of field CBR tests relates to the possibility of
there being larger stones immediately under the piston during the test either in the field or
in the laboratory.
The direct field test of CBR may give the CBR at field moisture content but it is not possible
to measure the CBR after soaking unless the test site has been inundated for some time.
As these tests are relatively uncommon, this Handbook does not cover them in detail.
Extension collar
2 inches (50.8) 2 lugs
Screwed socket fits post
Mould body
6 inches dia
(152.4)
7 inches
(177.8) 4.584 inches 2 lugs
high (116.4) Posts fixed corners
of square baseplate
Spacer disc
Cutting shoe
(Reference: Federal Works Agency, USA (1943) Principles of Highway Construction

as Applied to Airports and Other Landing Areas for Aircraft, Public Roads Administration)
FIGURE 6.1.2.1(A) CBR MOULD

69 HB 160—2006
(Reference: Federal Works Agency, USA (1943) Principles of Highway Construction

as Applied to Airports and Other Landing Areas for Aircraft, Public Roads Administration)
FIGURE 6.1.2.1(B) CBR MOULD, FITTINGS AND ACCESSORIES

6.2 VANE SHEAR TESTING (AS 1289.6.2.1)

6.2.1 Application
The vane shear test is used to determine the undrained shear strength of very soft to firm
fine-grained cohesive soils. It is particularly suitable for soft, saturated, sensitive clays that
are difficult to sample, and with which the disturbance by sampling may be significant. The
test is not always suitable for stiff to very stiff or sandy clays, or those that are not
saturated.
The vane test is normally performed in the bottom of a borehole.
6.2.2 Apparatus
The vane comprises four rectangular blades arranged in the form of a cross on the end of a
thin rod (about 12 mm in diameter), which is coupled to the bottom of a drill rod string. The
size of the vane is limited by the size of the casing that is used, i.e., down which it must
pass. A vane size of 130 mm long by 65 mm in diameter is common. This size of vane can
be inserted through standard NX drill casing.
Alternative systems that do not include testing in a borehole may be used. Such systems
are known as penetration vanes, which are pushed directly into the soil. Particular care
needs to be taken to sheathe the torque rods against the effects of friction from the soil.
For the conventional system, a device is used at the top of the borehole, which suspends
the vane and rods so that they do not sink into very soft ground beyond the depth at which
the test is to be carried out. A device is also used to rotate the rods and vane and to
measure the resistance of the vane to rotation in the soil (i.e., the torque on the rods). For
standard testing these devices will be incorporated in a testing head that clamps to the
casing and which in turn clamps the rods. The rods are turned by a worm drive. The
standard rate is 6° per minute, i.e., 1 hour for a full revolution.
For some purposes, a simpler system can be used. The rods are held on the end of the
winch line and are rotated with a torque wrench on which a dial gauge is fitted to give
readings of torque or resistance to rotation.

HB 160—2006 70
6.2.3 Procedure
The rod string, with the vane attached at the bottom, is run into the hole and the vane is
pushed 300 mm below the bottom of the bore. The rods are then clamped or hung to
prevent them sinking further and the rotation and torque measuring device that is being
used is attached. The initial dial gauge reading is taken (or reset to zero); the rods are then
rotated at the appropriate rate. The torque reading on the dial gauge will rise to a maximum
and then fall somewhat to a fairly steady figure, with minor fluctuations. The maximum
figure shown on the dial gauge is recorded and also the final steady figure.
Examples of the vane test apparatus for both borehole and penetration type devices are
shown in Figure 6.2.3(A). Details of the vane and rod mounting are shown in
Figure 6.2.3(B).
FIGURE 6.2.3(A) EXAMPLES OF VANE SHEAR DEVICES

71 HB 160—2006
FIGURE 6.2.3(B) EXAMPLES OF VANE AND ROD MOUNTING
Calibration charts accompanying the shear vane should allow the direct reading of shear
strength against torque gauge readings. The peak and residual vane shear strength are
noted—
• s v = (peak strength) kPa
• s t = (residual strength) kPa
Research has shown that the stress distribution at the ends of the shear vane is parabolic.
If the parabolic stress distribution is simplified to a triangular mobilization of shear strength
at the ends of the vane, the strength at failure (sv ), can be expressed as—
2 3
s v = T/π(d2h/ + d /8)
where
T = torque at failure
d = diameter of shear vane
h = height of the shear vane

HB 160—2006 72
Bjerrum (1973) has shown that the shear strength of clay assessed from shear vane testing
varies with the plasticity index (PI) of the clay. For that reason he has suggested the
following correction to assess a design shear strength:
C u(design) = λs v
where
λ = correction factor = 1.7 – 0.54 log (PI)
6.2.4 Hand-held test devices
For hand-held testing, a hand-held torque gauge is used. This connects directly to the vane
or extension rods. These devices are typically available with 33 mm vanes for measuring
shear strength up to about 25 kPa, and 19 mm vanes for measuring shear strength up to
300 kPa. They are also available with up to 6 m of extension rods for down-borehole
applications only.
6.3 STANDARD PENETRATION TEST (AS 1289.6.3.1)

6.3.1 Application
The standard penetration test (SPT) is used in many parts of the world and there is a large
volume of information relating the results of the test to various engineering properties.
The equipment is used in conjunction with a drilling rig. Drilling contractors use automatic
tripping drop-hammers as described in the test procedure. In summary, the test can be
used during auger drilling or rotary drilling and provides a numerical result (N value) and a
sample taken with a thick-walled split tube. The sample recovered is not suitable for
laboratory measurement of strength or consolidation properties but is suitable for
classification testing and identification purposes.
The test has its major application in non-cohesive soils (e.g., sands and gravels) but can
also be used in clay and weathered rock as a rough guide to strength.
6.3.2 General procedure
The penetration tool consists of a thick-walled split-spoon sampler (Raymond sampler)
50 mm OD and 35 mm ID. The sampler is attached to the bottom of the drill rods and
lowered to the bottom of the borehole. It is then driven 450 mm into the soil at the bottom of
the borehole with a 63.5 kg hammer falling freely through 760 mm. Only the number of
blows for the last 300 mm of driving is recorded as the standard penetration resistance ‘N’.
It is considered good practice to count the number of blows for every 150 mm of penetration
for the full 450 mm of driving.
After withdrawal from the borehole, the tube is split and the sample is examined. Various
suggestions have been made relating to the adjustment of the measured N values to allow
for the influence of effective overburden pressure and to the adjustment of the measured N
value in fine sand beneath the water table, but these adjustments are generally not
universally adopted or applicable. Where N values are being measured beneath the water
table, great care is needed to minimise errors caused by the disturbance of granular soils
by the inflow of water.
In particularly strong or dense soils, if the blow count exceeds 30 blows for any 150 mm
interval, the test should be stopped and the distance driven for the 30 blows recorded. Note
also that if the penetrometer tool is bouncing with no or very little penetration, the test
should be stopped after 5 blows and the penetrated distance recorded.
An example of a typical split tube sampler assembly is presented in Figure 6.3.2.

73 HB 160—2006
DIMENSIONS IN MILLIMETRES
FIGURE 6.3.2 SPT SPLIT TUBE SAMPLER ASSEMBLY

6.3.3 Interpretation of test result

Various correlations between SPT ‘N’ values and soil density and consistency have been
proposed. The relative density of granular soils according to SPT results are presented in
Table 6.3.3(A). The consistency of cohesive soils, according to the SPT N value, is
presented in Table 6.3.3(B).
TABLE 6.3.3(A)
SPT ‘N’ RELATIVE TO DENSITY
SPT ‘N’ blows Density index

Term Abbreviation Field guide
per 300 mm* (%)
0–4 0–15 Very loose VL Ravels
4–10 15–35 Loose L Shovels easily
10–30 35–65 Medium dense MD Shovelling difficult
30–50 65–85 Dense D Pick required
>50 85–100 Very dense VD Picking difficult
* Dependent on overburden pressure and equipment used

HB 160—2006 74
TABLE 6.3.3(B)
SPT ‘N’ RELATIVE TO CONSISTENCY
SPT ‘N’ blows Shear strength

Term Abbreviation Field guide
per 300 mm* (kPa)
0–2 <12 Very soft VS A finger can be pushed well into
the soil with little effort
2–4 12–25 Soft S A finger can be pushed into the
soil to about 25 mm depth
4–8 25–50 Firm F The soil can be penetrated about
5 mm with the thumb
8–15 50–100 Stiff St The surface of the soil can be
indented with the thumb, but not
penetrated
15–30 100–200 Very stiff VSt The surface of the soil can be
marked by the fingers, but not
indented with thumb pressure
>30 >200 Hard H The surface of the soil can be
marked only with the thumb nail
* Dependent on equipment used

6.3.4 Corrections to ‘N’ values
6.3.4.1 General
Measured SPT ‘N’ results may be subject to corrections for equipment type used and other
factors. Overburden pressure, rod length, bore diameter and groundwater corrections are
those most commonly applied.
6.3.4.2 Overburden
Gibbs and Holtz (1957) demonstrated that the standard penetration resistance depends not
only on the relative density for coarse-grained soils but also on the effective stresses at the
depth of measurement. Coarse-grained soils with the same relative density would give
different values of standard penetration resistance at different depths. There are a number
of correction factors for effective overburden pressure. One of the simplest to use was
proposed by Liao and Whitman (1986):
 95.76  1/ 2
C OP =  

 p o′ 
where
C OP = correction for overburden pressure
p o′ = effective overburden pressure (kPa)
6.3.4.3 Rod length correction
When sampling is performed with drill rod lengths less than 10 m, a reflected energy wave
reduces the net test energy. To accommodate this, the following correction factors are
recommended for tests performed with such rod lengths:
Drill rod correction factor,

Rod length (m)
CDR
0–4 0.75
4–8 0.85
8–10 0.95

75 HB 160—2006
6.3.4.4 Borehole diameter

Evidence suggests that lower ‘N’ values are obtained in oversize boreholes (150 mm to
200 mm). A borehole diameter correction factor can be applied in these cases, as follows:
Borehole diameter Blowhole diameter

(mm) correction factor, CBD
150 1.05
200 1.15
6.3.4.5 Groundwater
When testing is conducted in fine-grained soils, fine sands or silty sand below the water
table can generate excess pore water pressure (during driving), which cannot dissipate.
The excess pore water pressure will be realised by increased test resistance. This can be
corrected for ‘N’ counts greater than 15 using the following formula:
N′ = 15 + 0.5(N – 15)
6.4 DYNAMIC CONE PENETROMETER (DCP) AND PERTH SAND PENETROMETER

(PSP) (AS 1289.6.3.2 and AS 1289.6.3.3)
6.4.1 Application
DCP and PSP testing is used to assess the resistance of a soil to driving a cone-tipped or
blunt-ended rod. Results may be used to infer shear strength of clayey soil, California
bearing ratio (CBR) of clayey or sandy soil and compaction and relative density of sand.
6.4.2 Occupational health and safety
Improper use of a penetrometer can result in short- or long-term injury. The following safety
measures should be adhered to:
• The equipment is heavy. Always ensure correct lifting techniques are used and,
where possible, carry the components separately, and utilise manual lifting aids or a
second person.
• Prolonged use of a penetrometer may result in hearing damage. Hearing protection
should be worn when operating the apparatus.
• Good posture is important while operating a penetrometer. Maintain a straight back
and bend the knees when lifting and releasing the hammer.
• The most common injury when operating a penetrometer is a broken, crushed or
severely bruised finger caught between the anvil and the falling hammer. KEEP
FINGERS AND OTHER BODY PARTS CLEAR OF THE ANVIL WHEN RELEASING
THE HAMMER.
• When transporting a penetrometer in a vehicle, ensure that all components are
secured and do not move about in the back of the vehicle.

HB 160—2006 76
6.4.3 Apparatus
6.4.3.1 DCP
The essential components and dimensions of the DCP testing apparatus are shown in
Figure 6.4.3.1. In summary, the DCP test involves dropping a 9 ±0.1 kg hammer a distance
of 510 ±5 mm in order to drive a 20 mm diameter hardened steel cone ahead of 16 mm
diameter rods graduated in 100 mm intervals.
6.4.3.2 PSP
The PSP apparatus varies from the DCP apparatus as follows:
• The hammer drop is 600 ±5 mm.
• The cone is replaced with a 16 mm diameter blunt tip.
• The rods are graduated in 50 mm or 150 mm intervals.
Figure 6.4.3.2 shows the general arrangement of the PSP.

77 HB 160—2006
FIGURE 6.4.3.1 DYNAMIC CONE PENETROMETER ARRANGEMENT

HB 160—2006 78
FIGURE 6.4.3.2 PERTH SAND PENETROMETER

79 HB 160—2006
6.4.4 Testing procedure

The DCP testing procedure is as follows:
(a) Excavate to the top of the soil profile to be tested. Materials such as gravels, existing
pavements, crushed rock or hard components in fill are often difficult to penetrate and
could result in damage to the equipment. Excavation should extend below these
materials.
(b) Measure the depth from the surface to the top of the soil profile to be tested to the
nearest 10 mm and record.
(c) Hold the penetrometer, with rod attached vertically, and give the hammer a gentle tap
such that the widest part of the cone has started to penetrate the surface. This is the
start point for the test.
(d) Drive the penetrometer into the ground by lifting the hammer until it gently touches
the top nut and allowing it to fall freely onto the anvil.
CAUTION: KEEP FINGERS AND ANY OTHER BODY PART CLEAR OF ANVIL
(e) Record the number of blows required for each 100 mm penetration interval. If the
ground surface is uneven, any other fixed reference point may be used.
(f) If the desired test depth is not reached before the uppermost interval on the rod has
penetrated the test surface, attach another rod.

(g) Once the desired depth has been reached, remove the hammer and anvil assembly.
Place the base of the extractor against the protruding rod, place the lever on top of
the stand and the cam wheels over the rod (they fit only one way!). Extract the rods
by working the lever up and down and removing each rod when the joint reaches
about 200 mm above the surface (so there is enough exposed rod for the cam to
grip).
The procedure for the PSP test is similar to that for the DCP test, except that:
• The tip penetrates the surface until the first mark is reached (150 mm depth). This is
referred to as ‘seating’ the penetrometer.
• The number of blows per 150 mm interval is recorded. The number of blows required
to drive the rod from 150 mm to 450 mm depth is the penetration resistance. Blow
counts recorded at deeper intervals may need adjustment to remove the effects of
friction on the side of the rods.
It is important to be aware of the locations of all underground services in the vicinity. At
sites where service locations are not known, a ‘dial before you dig’ enquiry should be
undertaken prior to commencing the test.
The test should be stopped if—
• the penetration resistance exceeds 8 blows per 20 mm;
• more than 25 blows are recorded in a single interval; or
• the hammer is bouncing and no observable penetration is occurring.
If the test is stopped, record the number of blows and the penetration that has occurred
over the interval in which the test was stopped. Note the reason for stopping the test.
Remember also that if it takes a lot of effort to drive the rods into the ground, it may also
take a lot of energy to extract them!
If high penetration resistance is thought to be due to an isolated obstruction or hard
material, attempt a test in a nearby location. If after three attempts penetration cannot be
achieved, abandon the test and consider excavating the obstruction or hard material before
recommencing the test.

HB 160—2006 80
If during the test a softer layer is found underlying a harder layer, it is possible that the final
blow in the harder layer will result in considerable penetration into the softer layer. In this
event, the distance penetrated by the final blow should not be considered with the readings
for the hard layer. The approximate depth of the base of the harder layer should be
recorded.
If the test is conducted in soft or loose material, a small number of blows may result in
considerable penetration. Should this happen, it is possible that the final blow count in the
interval will result in considerable penetration beyond the end of that interval and the blow
count recorded may not be indicative. An alternative in soft ground is to place a vertical
metre rule adjacent to the penetrometer. A reference point is taken on the hammer (usually
a ring is engraved at the base of the hammer for this purpose) and a reading taken on the
ruler to the nearest millimetre. With each blow, the depth of penetration is recorded using
the reference point and the ruler. If when using this method the material becomes stiffer or
more dense and a penetration of less than 20 mm is recorded per blow, then record the
penetration for each eight blows, noting the number of blows and corresponding amount of
penetration.
Penetrometer testing will often be carried out in conjunction with test pits or boreholes. If
this is the case, conduct the penetrometer test prior to excavation or ensure that the test is
undertaken at least 1 m away from the test pit/borehole to avoid adverse ground-loosening
effects. Penetration resistance is a function of moisture content, material type and density
or stiffness. Accurate sampling from a borehole or test pit should allow good correlation
with the penetrometer test results.
6.4.5 Reporting
Typically, the results of DCP/PSP testing are presented by plotting penetrometer
information alongside a test pit or borehole log at the same location. Where DCP/PSP tests
are performed well away from the test pits (or no test pitting is carried out), or multiple
penetrometer tests are performed at a test pit location, the DCP/PSP results can be
presented on the field record sheets.
6.4.6 Interpretation
6.4.6.1 DCP
DCP results are often used to estimate the consistency or density of subgrade soil.
Tables 6.4.6.1(A) and 6.4.6.1(B) provide a guide to the interpretation of DCP results.
TABLE 6.4.6.1(A)
CORRELATION OF FINE GRAINED SOIL
CONSISTENCY TO DCP RESULTS
Consistency Blows/100 mm
Very soft to Soft <1
Firm 1–2
Stiff 3–4
Very stiff 5–10
Hard >10

81 HB 160—2006
TABLE 6.4.6.1(B)
CORRELATION OF SAND DENSITY
TO DCP RESULTS
Density Blows/100 mm
Very loose <1
Loose 1–2
Medium dense 2–3
Dense 4–8
Very dense >8
Note that the values in Tables 6.4.6.1(A) and 6.4.6.1(B) are a guide only and that these
correlations are depth and moisture dependent.
DCP results are also often used to estimate a CBR value. The listed references may be
useful for this purpose; however, when using the DCP extensively for assessment of CBR,
other testing alternatives should be used to confirm the validity of the CBR/DCP blow count
relationship adopted.
A guide to the conversion of DCP blow counts to CBR values is given in Table 6.4.6.1(C)
(adapted from Austroads, 2004).
TABLE 6.4.6.1(C)
CORRELATION OF DCP BLOW COUNTS TO CBR VALUES
Blows/100 mm CBR (%) Blows/100 mm CBR (%)

1 2 11 25
2 4 13 30
3 6 15 35
4 8 17 40
5 10 19 46
7 14 21 51
9 20 23 56
10 22 25 62
6.4.6.2 PSP
Results from the PSP are often used in the design of footings or as guide to relative
density. It should be noted that the PSP was developed for use in the sands of the Perth
area, Western Australia. Care should be taken when interpreting results in other material
types. The listed references may be useful in the interpretation of results.
Table 6.4.6.2 provides a guide to the correlation of sand density to PSP results for the
depth interval 150 to 450 mm (from Glick and Clegg, 1965).

HB 160—2006 82
TABLE 6.4.6.2
CORRELATION OF PSP BLOW COUNT
WITH RELATIVE DENSITY
Density Blows/300 mm
Very loose ≤2
Loose 2–6
Medium dense 6–8
Dense 8–15
Very dense >15
Note that this correlation will vary where the test extends to greater depth. Care should be
exercised when interpreting density at depths greater than 450 mm.
Where the PSP is used as a guide to relative density (or degree of compaction), the blow
count should be correlated against density by establishing a trial pad where the relative
density and PSP blow count are measured against compactive effort.
6.5 TRIAXIAL TEST

6.5.1 Background
A widely used apparatus to determine the shear strength parameters and the stress-strain
behaviour of soils is the triaxial apparatus. The name is a misnomer since two, not three,
stresses can be applied on the soil specimen.
In the triaxial test, a cylindrical sample of soil, usually with a length to diameter ratio of 2, is
subjected to either controlled increases in axial stresses or axial displacements and radial
stresses (see Figure 6.5.1).
FIGURE 6.5.1 SCHEMATIC OF A TRIAXIAL CELL (after Bishop & Henkel, 1962)
Some of the first triaxial tests carried out were reported by Rendulic in 1936 on work that
was carried out under Karl Terzaghi in Vienna. Since that date the triaxial test, in all its
various forms, has become one of the most common tests carried out in soil mechanics.

83 HB 160—2006
The triaxial apparatus is versatile because it is possible to—

• independently vary the applied axial and radial stresses;
• alter the drainage conditions so that the soil may be tested to best represent the
condition existing in the field;
• apply either controlled displacements or stresses; and
• obtain important soil parameters such as Young’s modulus, Poisson’s ratio, etc.
In general, the diameter of the sample should be a minimum of 38 mm or 6 times greater
than the maximum particle size, whichever is the greater. This ratio becomes less critical
for specimens 100 mm and larger, where the ratio may be relaxed to 5. In the case of an
unconsolidated undrained (quick) triaxial test, the porous stones shown top and bottom of
the specimen can be replaced with impervious plastic discs to prevent drainage.
A wide variety of soil properties can be measured using triaxial testing equipment, including
consolidation parameters, permeability coefficient, as well as undrained and effective shear
strength. Triaxial testing equipment can also be used to test partially saturated soils to
measure negative pore pressure values and the coefficient of at-rest lateral pressure (K 0).
These complex testing procedures, however, are outside the scope of this Handbook.
6.5.2 Apparatus
The apparatus comprises:

• Triaxial compression machine (electrically driven strain-controlled machine is
recommended) with a load ring (or a load cell).
• Triaxial cell with a dial gauge for deformation measurement.
• Compaction mould (for cohesionless soil).
• Rubber membrane (0.3 mm–0.5 mm thick)—natural rubber for use in water or air,
synthetic rubber for use in oil.
• Membrane stretcher (for cohesive soil).
• O-rings or suitable rubber bands.
• Porous stones (for consolidated and/or drained tests).
• Plastic discs (for unconsolidated undrained test).
• Vacuum pump.
• Compressed air source.
• Wire saw and soil lathe (for cohesive soil).
6.5.3 Measurements
The measurements taken are:
• Diameter of soil specimen, D
• Length of soil specimen, L
• Initial moisture content, wi
• Final moisture content, wf
• Load ring constant, PRC
• Load ring reading at regular vertical deformation intervals
• Vertical deformations at certain intervals
• Angle of failure plane, θ

HB 160—2006 84
6.5.4 Specimen preparation

6.5.4.1 Undisturbed cohesive soil
Cohesive soils are usually fairly straightforward to prepare except when they are very soft.
In such cases a field and/or laboratory shear vane test can be carried out in conjunction
with the triaxial test. The specimen should ideally be prepared in a humid room, with every
reasonable precaution taken to minimise drying and disturbance of the soil. The sample
should then be extruded from a sampling tube and placed on a mitre box or soil lathe and
carefully trimmed off to the correct size using a wire saw.
Clay samples containing pieces of gravel can have voids in their sides after extrusion.
These voids may be carefully patched with sample trimmings. Gravel protruding from the
top and base may be removed and replaced with excess material obtained from the
trimmings.
6.5.4.2 Cohesionless soil
If specimens of cohesionless material have to be tested, then it is highly likely that they will
not be self-supporting when they are placed on the pedestal. To position the sample in the
triaxial cell, a rubber membrane is used to confine the specimen in combination with a split
former and the application of a low vacuum [Bishop & Henkel (1962)].
6.5.4.3 Compacted soil
Disturbed soil can be compacted into triaxial specimens of appropriate dimension. Care
must be taken to ensure that the soil sample has been properly treated to bring it to a target
moisture content. For specimens 100 mm or greater, split moulds are usually used. In
compacting the soil sample there should be not less than three layers. Five or more may be
more appropriate for samples that are difficult to compact. The surface of the intermediate
layers needs to be lightly scarified before adding the next layer to, prevent stratification.
Once compacted, the layers must be handled in the same manner as for undisturbed
specimens.
6.5.5 Terminology and principles
6.5.5.1 Stress system
In a triaxial test, compressive stresses are applied from three directions (see
Figure 6.5.5.1(A)). Axial stress σ1 is called the major principal stress in a compression test
but will be the minor principal stress in an extension test*. In a compression test, the lateral
stresses σ 2 and σ 3 are the intermediate and minor principal stresses, respectively, which
will become the major principal stresses in an extension test. The two stresses are applied
by pressurising a medium, which is normally water or, in some cases, oil or air, such that
σ 2 = σ3 = confining pressure.
During triaxial testing, the triaxial cell, which is in fact a pressure chamber containing a soil
specimen, is normally placed on a machine that applies axial loading to the specimen via a
loading ram (see Figure 6.5.1). The applied axial stress is called the deviator stress (σ d).
Thus,
applied axial load
σd =
average cross-sectional area
σ 1 = total axial stress = σd + σ3; or

σ d = deviator stress = (σ 1 − σ3 )
* Compression test is when the specimen is loaded under increasing axial stress while keeping the lateral
stresses constant; extension test is when the specimen is loaded by decreasing the axial stress while keeping
the lateral stresses constant. Only the compression test is described in this Handbook.

85 HB 160—2006
Triaxial testing commonly involves the application of axial load on a specimen at a constant
confining pressure. The compression machine used can be programmed such that the
specimen’s axial deformation rate (e.g., mm/min) is maintained throughout the test until the
specimen fails. Axial load is measured using a load ring (or, more conveniently, using a
load cell). Deformation of the specimen is measured using a dial gauge (or, more
conveniently, using a displacement transducer). The variation of deviator stress (σ d) (i.e.,
axial load divided by cross-sectional area of the specimen) is then plotted against the axial
strain (ε v) (i.e., axial deformation of the specimen divided by the original height of the
specimen). Figure 6.5.5.1(B) shows a typical plot of σ d vs. ε v .
σ1
σ2
σ3
FIGURE 6.5.5.1(A) THREE-DIMENSIONAL STRESS SYSTEM

DEVIATOR STRESS, kPa
AXIAL STRAIN
FIGURE 6.5.5.1(B) TYPICAL STRESS-STRAIN CURVE OF TRIAXIAL SPECIMEN
6.5.5.2 Mohr’s circles and soil parameters

The analysis theory behind the concept of Mohr’s circle is explained in detail in most
strength of materials or mechanics of solids textbooks and, consequently, will not be
addressed here beyond what is needed.
In applying Mohr’s circle, the value of σ3 is equal to the confining pressure σ c . The value of
σ 1 is found by adding the maximum value of deviator stress [Figure 6.5.5.1(B)], ∆σ max , to
the confining pressure (σ3):
σ 1 = σ3 + ∆σ max

HB 160—2006 86
The Mohr’s circle (only a semicircle is required) that intercepts the horizontal or σ n axis at
points equal to σ 3 and σ 1 is drawn as shown in Figure 6.5.5.2. The test is repeated at least
two more times, each trial at a different confining pressure. The resulting Mohr’s circles are
drawn on the same set of axes as the first circle. The horizontal and vertical scales used in
drawing Mohr’s circles must be the same. A straight line of best fit is then drawn so that the
curve is tangential to each of the Mohr’s circles. This line represents the Coulomb failure
envelope—
τ = c + σ tan φ
where
τ = shear stress
c = cohesion
σ = normal stress
φ = the angle of internal friction
Some graph-plotting software packages that are capable of plotting Mohr’s circles are
available; however, care should be taken to ensure that the software has been programmed
to accommodate the assessments on the correct position of Coulomb’s failure envelope.
SHEAR STRESS, T
n
(σ n Tf ) σ n ta
c +
T =
σn NORMAL STRESS, σn
σ3
σ1
FIGURE 6.5.5.2 MOHR’S STRESS CIRCLE
6.5.6 Types of triaxial test

6.5.6.1 Standard tests
Generally, the application of the confining pressure and of the deviator stress form two
separate stages of the test. Therefore, tests are classified according to the conditions of
drainage obtained during each stage (Bishop and Henkel, 1962):
• Test I [unconsolidated–undrained (UU) or Quick Test]: No drainage is allowed during
the application of confining pressure and during the subsequent application of the
deviator stress. The loading is applied to the specimen so rapidly that no water can
escape or drain from the test specimen during the test. This test will always produce
a value for φ of 0° for saturated cohesive soils, which is often described as ‘φ = 0°
conditions’. A variant of the test, which does not use confining pressure, is called the
unconfined compression test or uniaxial test.

87 HB 160—2006
• Test II [consolidated–undrained (CD)]: Drainage is allowed during the application of

confining pressure but no drainage is allowed during the subsequent application of
the deviator stress. The loading is applied rapidly, as in the UU test. This test will
usually produce c and φ values intermediate to those resulting from UU and CD tests
performed on the same soil.
• Test III [drained]: Drainage is permitted throughout the test such that no water
pressure should develop within the specimen. The loading rate must be sufficiently
slow to allow full dissipation of pore water pressure. The test produces drained
(effective) shear strength parameters (cd and φ d or c′ and φ′).
Figure 6.5.6.1 shows typical Mohr’s circles and failure envelopes for the three test types.
The choice of which conditions to use during testing is usually dependent on the drainage
conditions expected to exist at the site. If the soil is sandy, excess pore water pressures
resulting from new loads can usually be expected to dissipate rapidly and drained
conditions may be preferred. On the other hand, if the soil in situ is saturated clay and the
loading rate is reasonably quick, the UU test is the proper choice.

HB 160—2006 88
u 0°
Cu
σn
(a) Unconsolidated-undrained (UU) test
T
cu
C cu
σn
(b) Consolidated-undrained (CU) test
T d
Cd 0
σn
(c) Consolidated-drained (CD) test
FIGURE 6.5.6.1 MOHR’S CIRCLES AND FAILURE ENVELOPES FOR UU, CU

AND CD TESTS

89 HB 160—2006
6.5.6.2 Undrained test on saturated cohesive soils

Under undrained conditions, saturated soil will deform under pressure but the total volume
remains unchanged. Therefore, the assumption is made for laboratory calculations that it
remains a right cylinder although the cross-sectional area will change under pressure.
When calculating the average cross-sectional area (CSA) the following formula is adopted:
volume V
CSA = =
initial height − change in height h − ∆h
Note that the above is the assumed deformed shape that can be quite different in reality.
The three major types of failure are illustrated in Figure 6.5.6.2. It is important to make a
sketch of the specimen after failure and measure the angle of the failure plane (θ) and
report this information with the test results.
Brittle/shear Plastic/barrel Intermediate/

failure failure multi-shear plane
failure
FIGURE 6.5.6.2 POSSIBLE FAILURE SHAPES IN TRIAXIAL TESTING
Failure occurs readily on stiff to hard specimens at relatively low strain; therefore, the
readings can be taken at 0.5% intervals up to 5% strain and 1% intervals thereafter. On
very moist, highly plastic and soft samples the deviator stress will keep increasing for very
high strains. In such cases, failure should be taken at 20% strain.
6.5.6.3 Strain rates
Bishop & Henkel (1962) noted that a test should be complete in about 10 minutes; Akroyd
(1957) recommends about 2%/minute. AS 1289.6.4.1 cites a range of 0.05 to
10 mm/minute. This wide range cited by the Australian Standard is to accommodate the
various sizes of samples to be tested. Table 6.5.6.3 provides a guide for the strain rate that
should be adopted for various specimen heights.
TABLE 6.5.6.3
STRAIN RATES FOR VARIOUS
SPECIMEN HEIGHTS
Specimen height (mm) Rate of strain mm/min

76 1.5
100 2.0
125 2.5
150 3.0
200 4.0

HB 160—2006 90
6.6 ONE-DIMENSIONAL CONSOLIDATION TEST (AS 1289.6.6.1)

6.6.1 General
The two important soil properties used in the design of foundations are the bearing capacity
and the settlement characteristics. The first is determined by means of shear strength tests,
which are not discussed in this Handbook, whilst the settlement characteristics are
determined by means of a consolidation test. In the consolidation test the change in volume
of a saturated soil, due to the drainage of the water from a sample induced by a static load,
is measured.
When saturated soil is loaded, a state of stress is set up and nearly all the load is initially
carried by the water in the soil pores since the water can be considered to be almost
incompressible. The increase in pressure in the water causes it to drain away from the soil
pores and produces a change in water content of the soil. The rate at which the water can
escape depends upon the permeability of the soil. In a free-draining cohesionless soil, such
as medium and coarse sand or gravel, the drainage and consolidation are almost
instantaneous, so that very soon after its application the load is carried by the intergranular
pressure (i.e., effective soil pressure); however, with clay soils, the permeability is low and
there is a time lag between the application of load and the change in volume. During this
period, the soil remains fully saturated and the change in water content corresponds to a
reduction in the volume of the pores (the voids ratio) in the soil. Since the change may take
place over many years, it is important to measure the rate at which the voids ratio changes,
and the final amount of consolidation.
The consolidation test measures the rate and amount of voids change with load on the soil
in a laboratory specimen, the results of which are applied to engineering problems using the
theory of consolidation.
The process of consolidation is best illustrated by reference to the model analogy shown in
Figure 6.6.1, which has been employed by Terzaghi (1943) when demonstrating the
fundamentals of consolidation. This model consists of a cylinder with a closely fitting piston
perforated with a number of small holes resting on a number of compression springs. The
cylinder beneath the piston is filled with water. Consider what occurs when a vertical load is
applied suddenly to the piston. Instantaneously, the entire load is carried by the water.
Owing to the difference between the pressure existing in the water and that in the
atmosphere, water will commence to flow out through the holes in the piston, the rate of
flow depending on the difference of pressure and the size of holes (e.g., a soil of low
permeability would be represented by small holes). As water escapes through the piston, a
reduction in the water pressure will ensue and the load will be transferred gradually to the
springs. Eventually, the pressure difference of the water will reduce to zero, and the entire
load will be carried by the springs, the piston coming to rest in a new equilibrium position. It
will be seen that the speed with which the process takes place will depend on the number
and size of the holes in the piston.

91 HB 160—2006
Ver tical load on piston
Compression Hole in piston

spr ings
Final equilibr ium
position of piston
under ver tical load
Cylinder
Water
FIGURE 6.6.1 PROCESS OF CONSOLIDATION
A mathematical theory of consolidation was proposed by Terzaghi many years ago. The
theory is fairly complex and the final differential equation that he obtained for the process of
consolidation can be solved only by means of an infinite series.

The main assumptions Terzaghi made are:
• The soil is homogeneous and is completely saturated.
• The soil particles and water are incompressible.
• Darcy’s law and a constant coefficient of permeability apply.
• The time for consolidation is dependent upon the coefficient of permeability.
• All consolidation is due to change in the water content.
In the one-dimensional theory of consolidation, the soil is considered to consolidate only in
one direction, and to be laterally confined. In the application of the test results, it is
assumed that size effects do not apply and that the results may be used in direct
calculations.
From the various initial assumptions, it follows that the results represent only an
approximation to the correct solution since most soils are not homogeneous and their
permeability in a horizontal direction may be 10 or 100 times that in a vertical direction.
Also, as consolidation takes place, the voids ratio reduces and, consequently, the
coefficient of permeability diminishes, thus changing the rate of consolidation.
Soils can also contain materials that are soluble, compressible (apart from soil grains and
water) and decomposable, all of which can have a significant effect on their long-term
settlement characteristics. These properties may not be manifested during consolidation
tests or they may cause the soils to behave in a manner that makes the results difficult to
interpret.
In addition, secondary consolidation takes place, which is not due entirely to the change in
water content. Consolidation due to secondary causes can be so pronounced in certain
soils in laboratory results, that it is necessary to distinguish between consolidation that is
explained by the I-D consolidation theory and consolidation that is not explained by the I-D
consolidation theory. The change point, at which it is assumed that the consolidation due to
moisture content change has been completed, is known at the 100% primary consolidation,
and the voids ratio at this point is used in the calculations.
It is generally found that the voids ratio is directly proportional to the logarithm of pressure,
so this has been adopted as common practice to plot a curve of e or percent change in
thickness vs. log p, for laboratory results.

HB 160—2006 92
The rate at which the volume change takes place, the coefficient of consolidation, is
determined from the measurement of the volume change per unit time. The rate depends
upon the coefficient of permeability and is proportional to the square of the length of path
through which the water has to drain. Since the permeability of clay soils may be
10 −10 m/sec or less, it is necessary to reduce the path length to a minimum consistent with
reproducible results. By allowing the water to escape from both the top and bottom of the
sample (double drainage condition) during test, the path length can be reduced by a half.
Method 6.6.1 covers the measurement of one-dimensional consolidation of a soil, the
pressure voids ratio relationship, and the coefficient of consolidation at various
successively increasing pressures, whilst the soil specimen is laterally confined, subject to
a vertical axial load and allowed to drain.
6.6.2 Apparatus
The apparatus includes an odometer; a sampling ring, with inside surfaces polished to
reduce the skin friction effects; a flange with an aperture the same as the ring and a
consolidation pot consisting of a circular water container in which are placed the
consolidation ring; and porous discs held in place by the flange, as described in
Method 6.6.1.
To measure settlement, a dial gauge with at least 5 mm travel and readable to 0.002 mm,
or LVDT (linear variable differential transformer) of equivalent readability should be used. It
should be noted that this is an extreme level of accuracy. Toolmaker’s tolerance is typically
to the nearest 0.025 mm. In order for these measurements to be meaningful, much greater
attention must be given to detail than is usually required for other types of soil tests. The
system should be thoroughly checked to ensure it applies pressures to the accuracy cited in
Method 6.6.1.
6.6.3 Procedure
The procedure is as follows:
(a) Extrude an undisturbed or compacted sample into the appropriate size sampling ring,
the inner surface of which has been pre-greased using silicon grease.
(b) Carefully trim the sample until the ring just slides over the soil, the last fraction of soil
being pared away by the cutting edge of the ring as it is slowly and evenly pushed
down over the sample.
(c) Trim off the top of the specimen using a sharp, thin-bladed knife until it is flush and
level with the top of the ring.
(d) After trimming the first side, to keep it intact, slide it onto a glass plate of, say,
100 × 100 mm.
(e) Trim the base of the specimen in the same manner.
It is extremely difficult to cut soil to a smooth flat surface, especially if the material
contains pieces of coarse sand and or fine gravel proud of the surface after trimming
and or it is fissured. These hard particles must be removed in such a manner as to
create as little disturbance to the surrounding soil as possible. To overcome the
resulting pockmarks, it is common practice to carefully patch them with the pared off
material, free of hard particles, and ‘paste over’ the area with a straightedge. If
extreme care is not taken, a significant percentage of the sample could end up
remoulded. Remoulding and ‘pasting over’ can dramatically alter the settling, swelling
and permeability characteristics of material, especially considering the measurements
being taken to the accuracy 2 µ.
(f) Slide the sample and the ring off the glass plate and, determine its mass (from which
the wet density of the soil can be calculated). Moisture content can be determined
from the cuttings in accordance with Method 2.1.1.
(g) Place the specimen in the consolidation cell as per the procedure set out in
Method 6.6.1, and load it appropriately.

93 HB 160—2006
If the specimen swells, it can be loaded in the manner described in the Standard or by the
following procedure:
(i) When swelling is observed, keep loading the specimen in such a manner that it
balances the swelling, trying to ensure that the specimen neither swells nor
consolidates. This can be achieved by having a pre-determined mass of dry sand and
a receptacle such that their combined mass is equivalent to a 25 kPa surcharge
mass.
(ii) When swelling commences, place the empty vessel on the load arm of the
consolidation unit. As soon as additional swelling is observed, tip sand into it to
prevent swelling. When the sand has been used and swelling is continuing, replace it
with a 25 kPa surcharge mass and repeat the process with the empty vessel and
sand.
(iii) When swelling ceases, unload the specimen to the initial load, usually overburden
pressure. Determine the swelling surcharge mass and calculate the swelling
pressure.
(iv) Report the swelling pressure with the other results obtained from the test.
(v) Allow the specimen to swell for 24 hours, record the degree of swelling then load it
using the originally prescribed pressures.
When calculating the consolidation, add the degree of swell to the initial height of sample.
It is strongly recommended that all the readings cited in Method 6.6.1 be taken in order to
see how the specimen is behaving even though only mv may be required. It is a requirement
of Method 6.6.1 to ‘Continue the readings at least until the slope of the characteristic linear
secondary position of the thickness vs. log of time plot is apparent’. If dial gauges are used,
great care must be taken to correctly ascertain the correct number of revolutions because
the internal rev counter can sometimes be very difficult to read.
6.6.4 Unloading
Once the loading stages of the test are complete, as a minimum, the specimen should be
unloaded to overburden and the rebound recorded after 24 hours. If the unloading
characteristics are required, Method 6.6.1 states ‘unload by stages using every second
pressure from the load sequence’. Carry out cycles at 24 hour intervals using the same time
intervals as for recording the loading cycles.
At the completion of the test, remove the sample and determine its moisture content as per
Method 2.1.1.
6.6.5 Size of specimen
Method 6.6.1 states that the preferred minimum specimen diameter be 50 mm, the
thickness 15 mm (but not less than 10 times the maximum grain diameter) and the
diameter-to-thickness ratio 3:1. Specimens have been used with success on a uniform
cohesive material of 38 mm; however, smaller specimens should not be used because
research indicates that the wall friction on the specimen can become significant.
6.6.6 Porous stone
A careful check must be kept on the porous discs to ensure that their surfaces have not
become clogged after continued use. If water is not readily flowing through the stone, its
permeability may be restored by grinding the surface back by about 0.1 mm.
6.6.7 Final mass
The final mass determination of the sample is most important when voids ratio values are to
be used, since all these calculations depend upon this figure. In routine work, it is usual to
have it checked by a second person independently from the one carrying out the test. If the
result of the mass determination should be wrong, about one week’s work would be wasted.

HB 160—2006 94
6.6.8 Plotting of deformation gauge dial readings

6.6.8.1 General
It is good practice to plot the deformation gauge dial readings as work proceeds. A separate
graph is plotted for each loading and the graph is extended until the results show that the
primary consolidation has been completed. The results of the deformation gauge dial
readings may be plotted in two ways, i.e., the log-time and the root-time basis.
Method 6.6.1 clearly sets out these two methods of plotting.
6.6.8.2 Log time basis
The deformation gauge readings are plotted as ordinates on an arithmetic scale; the
corresponding times, in minutes, at which the readings were taken, are plotted as abscissae
on a logarithmic scale. The usual form of the graph is shown in Figure 1 of Method 6.6.1.
The graph consists of a curve followed by two straight portions connected by a curve. The
first straight line (the primary consolidation), is dealt with by Terzaghi’s theory of
consolidation. The second straight line is the secondary consolidation. The two straight
portions are extended to intersect; this intersection is the point of 100% primary
consolidation.
The corrected zero point is obtained on the assumption that the curved portion at the
beginning of the graph is a parabola. The difference in ordinates between the 7.5 second
point at the beginning of the graph and the 30 second point (whose times are in the ratio of
four to one) is marked off above the 7.5 second point to obtain the corrected zero.
An example of the first five readings is as follows:
Elapsed Deformation gauge

Elapsed time
time (sec) readings (mm × 10–3)
0 2128
7.5 2046 t1
15 2027
30 2005 4t 1
60 1960
Let y = deformation at t1 – deformation at 4t1 = (2046 – 2005) = 41

0% consolidation = deformation at t1 + y
i.e., 0% consolidation is at reading = 2046 + 41 = 2087
Other texts use 10 and 40 seconds and 15 and 60 seconds; with these readings, yield
values of 0% consolidation is at 2092 and 2094 respectively.
The point of 50% consolidation is half-way between the corrected zero and the 100%; t50 is
the time for 50% primary consolidation in minutes.
For each load increment, the coefficient of consolidation (c v) can be calculated using the
following equation, which applies in the case of double drainage conditions (Method 6.6.1
2
requires the units to be in m /year):
2
0.026H m2
cv =
t 90 year
where
2
cv = coefficient of consolidation (m /year)
H = average thickness of specimen for the load increment (mm)
t 50 = time for 50% primary consolidation (minutes)

95 HB 160—2006
Where the coefficient of secondary compression (c α ) is required, determine the specimen

height change (∆H α ) over one complete log cycle from the straight line drawn through the
data points after the completion of readings over sufficient time to cover such a cycle
beyond the t100 point.
The coefficient of secondary compression is calculated from the following equation:
∆H α
cα =
H0
where
cv = coefficient of secondary compression
∆H α = specimen height change in millimetres over one log cycle time
H0 = initial height of specimen (mm)

6.6.8.3 Root time basis
The deformation gauge dial readings are plotted as ordinates on an arithmetic scale; the
square root values of the corresponding times in minutes, at which the deformation gauge
readings were taken, are plotted as abscissae. The form of the graph is shown in Figure 2
of Method 6.6.1. It is usually found that the graph consists of an approximately straight
portion followed by a curve. It is on the assumption that the 90% consolidation is 1.15 times
the abscissae value of an extension of this so-called straight line that the following
construction is used. To obtain the value of the 90% consolidation, a straight line is drawn
through the initial part of the straight portion and is extended back to zero time, which gives
the corrected point of zero consolidation. A second straight line with abscissae 1.15 times
those of the first is drawn; where the second line intersects the curve is the point of 90%
primary consolidation.
From the square root-time basis—
2
0.112H m2
cv =
t 90 year
In some cases, the determination of cv can be quite difficult to determine. Plots of data do
not always exhibit the shapes set out in Figures 1 and 2 of Method 6.6.1. It is recommended
that both the graphs of log time and root time basis be plotted the see which shape best fits
the theoretical shape, and comparisons of values obtained from the two methods be
compared when possible. Although researchers have determined that these graphical
methods are the best means by which cv parameters are to be evaluated, they can clearly
have some shortcomings.
6.6.9 Void ratio or percent change in thickness vs. pressure
It is common to graph the voids ratio or percent change in thickness as an ordinate against
the log pressure (p) as abscissae. An example of this plot is presented in Figure 6.6.9(A).
The most important aspect of this graph is to determine if the loads being applied by the
proposed structure are in the normally consolidated (which some texts refer to as the virgin
curve) or over consolidated (which can be referred to as re-compression) pressure range. In
the case of the example below, considerably more settlement is likely to occur when
pressures on this soil exceeds 330 kPa. Figure 6.6.9(B) shows the shape of pressure
release part of the graph and the shape of a graph when recompression cycles are applied
to a specimen.

HB 160—2006 96
Recompression cur ve
(over consolidated)
from consolidation test
Virgin
(nor mally
e0 consolidated)
compression
VOID RATIO
cur ve
e
∆p
P0 P
PRESSURE (log scale)
FIGURE 6.6.9(A) VOID RATIO VERSUS PRESSURE


97 HB 160—2006
1.00
Virgin cur ve
0.95
α/2
e 0 Extrapolated from test

α
0.90
Assumed effect of unloading

sample when recovered
∆e 0.85-0.75
VOID RATIO e
0.85 Cc = =
log p 2 /p 1 log 10.5/5.1
0.80
Rebound ∆e
cur ve
0.75
∆p
Cv
0.70
0
0.65
0.1 1.0 10.0 tsf
10 100 1000 kPa
log p 1
FIGURE 6.6.9(B) VOID RATIO VERSUS PRESSURE INCLUDING REBOUND CURVE
6.6.10 Preconsolidation load

Various texts set out ways to determine the preconsolidation load. The method shown
above, as set out by Bowles (1992), is a good example of this and is based on the method
devised by Casagrande (1932). This method is for the laboratory specimen only and does
not account for the sampling disturbance effect. Schmertmann in 1955 proposed a method
of determining the field virgin curve, which is set out by Winterkorn and Fang (1975).
The consolidation test was originated by Terzaghi and to a large extent it is carried out in a
way that is compatible with his theory. The test has not been altered much since its
inception, although there have been modifications in detail. The test takes nearly a week to
carry out and can be expensive in labour and equipment.

HB 160—2006 98
If, as is usual during a site investigation, it is found that the soil over the site varies, then it
might be necessary to carry out a number of tests on each of the main soil types. In
practice, it is more usual to carry out tests to determine the range of results and, which
might occur together with a determination of the average results, which might be expected
for the predominant soil types; therefore, to obtain an average of the conditions, a number
of tests is usually carried out on the different soil types.
6.6.11 Temperature
There is no requirement for temperature control in Method 6.6.1; however, if there are
significant changes in temperature then these will cause some fluctuations in the
measurements. It is best to carry out the test in a reasonably well-controlled environment.
The theory of consolidation is based upon, amongst other things, the assumption that the
coefficient of permeability is a constant; i.e., that the viscosity and unit weight of water are a
constant. These values are a function of the temperature and if there is a change in the
temperature, then there will be a change in these values. Corrections should be made
where the temperature of the test will be different from that in the soil in the ground. For all
practical purposes, the temperature of the soil in situ may be taken to be 10°C. The correct
value of the coefficient of consolidation may be obtained from the graph in Figure 6.6.11. If
no correction is made, then there may be an error in the percentage of the total settlement
that has taken place compared with that calculated from the laboratory results.
(From T.N.W. Akroyd ‘Laboratory Testing in Soil Engineering’, p 138,

William Clowes & Sons Ltd, London & Beccles 1957)
FIGURE 6.6.11 CORRECTION FACTOR FOR COEFFICIENT OF CONSOLIDATION
6.7 RESILIENT MODULUS

Resilient modulus is a highly specialised area of testing, which is not covered in this
Handbook. Details of its use are in the following Austroads publications:
• AP-T29/03
• AP-T30/03

99 HB 160—2006
6.8 CLEGG IMPACT VALUE

6.8.1 Background and application
The test for clegg impact value (CIV) was developed by Dr B. Clegg in the 1970s in the
Department of Civil Engineering at the University of Western Australia. The test was
presented in 1976 at the 8th ARRB conference held in Perth, Western Australia (Clegg
1976).
CIV provides a strength/stiffness parameter or index that may be used to observe changes
in strength/stiffness for a material or ranges of materials with changes in compactive effort,
density and moisture content, either at the time of compaction or over time. It may also be
used for quality control and quality assurance.
The primary applications of the test, in addition to compaction control, are assessment of
strength/stiffness and assessment of variability based on statistical concepts.
The test apparatus is simple to use and portable such that it requires only one person to
operate with minimal training. CIV may be used as an aid in the design and evaluation of
flexible pavements and earthworks, including trench reinstatement, roadworks and
unsealed airstrips.
The test may be used during compaction to check for soft spots as a result of under-rolling
(especially along the edges, adjacent to structures, over trench backfill, around sharp
curves, around gully pits, etc.); from excess moisture; from poor quality gravel; or from
subgrade influence due to too thin a layer (e.g., less than 150 mm of gravel), allowing for
necessary corrective action to take place at the time of construction.
Post-construction, the test may be used as a means to monitor:
• The increase in strength through evaporation of the ‘as completed’ moisture such that
a minimum ‘dry back’ strength is first observed prior to sealing.
• The ongoing soundness, particularly in relation to structural weakness and pavement
‘break up’ occurring as a result of heavy rains, flooding, undetected underground
water resources or water ingress at the shoulders.
Due to the simple and rapid nature of the test method, large areas can be scanned
relatively quickly; for example, it is possible for one person to obtain well over 200 data
points in the field in one day. Variability may be obtained for the evaluation of existing
situations or to check the uniformity during or after compaction.
For assessing variability, Clegg suggested that the mean, standard deviation and the
associated coefficient of variation (100 times the standard deviation divided by the mean,
written as a percentage) be obtained for a number of tests over an area. In terms of a mean
CIV, it is suggested that reporting be to the nearest 1.0; and for the standard deviation, to
the nearest 0.1. A coefficient of variation (cv) of around 4% or better is suggested as
representing an upper bound of job uniformity, while a c v of around 20% is suggested as
representing a highly variable condition, i.e., poor job quality.
The number of tests to perform on an area is usually discretionary and will depend to some
extent on the size and shape of the area and the variability encountered while testing,
though a more sophisticated approach based on statistical concepts may be employed. As
a guide, 5 to 10 random tests per lot is suggested as a starting point, with this number
increased or decreased according to the variability and/or the size and shape of the lot.
6.8.2 Apparatus and theory
The test method for CIV is described in AS 1289.6.9.1 and ASTM D5874. The standard
specification for it in the UK was established by British Gas in the late 1980s/early 1990s
for trench reinstatement testing in conjunction with their method specification.
The apparatus for the Clegg impact test is based on the instrumentation of a 4.5 kg
compaction hammer (or rammer). The instrumentation provides an instantaneous display of
the CIV result with no calculation required by the operator for determining the
strength/stiffness parameter.

HB 160—2006 100
Clegg made the observation that while compacting a sample of soil in a compaction mould,
the tonal quality of the impacts under the blows of the compaction hammer change as the
soil becomes stiffer. The initial idea was that if this tonal quality could be captured and
displayed electronically, a strength/stiffness property would result.
Experiments using an oscilloscope connected to an accelerometer fastened to a
compaction hammer were the most encouraging, and led to devising specially designed
electronics to measure and display the peak deceleration of the hammer on impact. This
then led to the output for CIV to be based on the universal dimensionless absolute value
ratio of the peak magnitude of the hammer deceleration during impact to the acceleration
due to gravity, expressed in the same units. In the past, this has been labelled as g but is
now becoming known as G max to distinguish it from the symbol for gravity and to indicate
that it is the peak deceleration of a free-falling mass upon impact with the surface.
Regarding the selection of the peak deceleration as the basis for CIV, Clegg (1978) stated
that ‘Many observations on a wide range of soils and conditions have shown that the
acceleration versus time relationship may be assumed sinusoidal. By integration of the area
under this curve, velocity-time is obtained and integration again produces penetration-time.
The force may also be obtained from the acceleration and hence a force-penetration
relationship may be established.’ The use of the peak deceleration therefore appeared to—
and has proved to provide a suitable and sufficient basis for a parameter for testing the
strength/stiffness of soils and soil-like materials; and the use of the peak deceleration kept
the test method and electronics simple, especially in the days before the availability of
super-miniature, high-speed micro chips and related technology and software.
The CIV test might be thought of as more of a strength test when testing soft materials or
conditions where there is a high amount of penetration/plastic deformation under the impact
of the hammer, and as more of a stiffness test when there is little or no discernible
deformation with the response being more elastic in nature. Clegg (1980) stated that ‘The
basic principle of the Impact Test is that the peak deceleration of the compaction hammer
when it is brought to rest is directly related to the resistance offered at contact resulting
from the stiffness and shearing resistance of the material. This resistance may be seen as
deriving largely from particle interaction leading towards densification or resulting in shear
deformation.’ For simplicity, the term ‘strength’ alone is often used in place of
‘strength/stiffness’.
While different weights of hammers were experimented with (and have entered production
for various applications such as testing of sand, turf and lightly sealed flexible pavements),
the 4.54 kg (10 lb) hammer (steel with hardened steel striking end) with a 50 mm hammer
diameter and a 457 mm (18 inch) drop height was found to be well suited as a general
purpose arrangement. Its output in terms of ‘tens of G max ’ provides a convenient 0–100
scale that offers sufficient output resolution and sensitivity for testing the full range of soils
and soft-rocks typically encountered in earthworks and flexible pavement construction, be
they subgrade, sub-base or basecourse material. The output parameter of CIV is also
referred to as impact value (IV). As a result of standardisation, the mass is now given as
4.55 ±0.05 kg with the drop-height as 455 ±5 mm.
For the sake of convenience and clarity, it is suggested that ‘tens of Gmax ’ be termed ‘Cmax ’
(with the ‘C’ standing for Clegg) regarding measurement of the peak deceleration of a free-
falling mass upon impact. The terms G max and C max relate purely to the peak deceleration
regardless of the construction or design of the apparatus, the drop-height or the blow count;
whereas the term ‘CIV’ is specifically meant to relate to the output as defined in the
Australian Standards and ASTM Standards for the 4.5 kg Clegg hammer. Both of these
Standards call for the CIV to be selected from the highest blow result of four successive
blows made on the same exact test spot.

101 HB 160—2006
Clegg (1978) stated that ‘The readings at successive blows almost invariably show increase
with blows ... Early in the development of the device the blows were continued until more or
less constant or where the increase per blow was steady and down to one division. It has
since been observed the fourth blow may be used without detrimental effect. This
considerably simplifies the procedure since it only requires a re-set at the beginning, drop
four times and take a reading which will be the maximum of the first four blows’ (where it is
that a ‘peak/hold’ logic circuit, as produced in Australia for the Clegg impact soil tester
since its advent, is used in the readout).
Clegg (1980) goes on to say: ‘The fact that the Impact Value increases with successive
blows appears to be inconvenient but in some respects it is essential to the technique.
Should the surface be so strong that there is virtually elastic rebound from the first blow it is
more difficult to observe stiffness differences. The desirable mechanism appears to be that
a wad of compacted material is projected into the body of the material. The recommended
practice is to use the 4th blow reading. The first one or two blows flatten and compact. Too
many blows pulverise and loosen the immediate surface or may continue to densify the
material. Early in the use of the device rubber pads were used on the ground surface but
later abandoned as an unnecessary encumbrance and a modest number of successive
blows are used instead.’
Whenever the hammer construction and dimensions are altered, and/or if the drop height or
selection of the value from other than the designated blow count is altered, the output, i.e.,
the peak deceleration on impact is, generally speaking, altered. Therefore, when testing for
CIV, the construction and design of the hammer and electronics, along with the drop-height
and method of testing, should meet the specifications, tolerances and procedures as found
in the existing Standards for CIV or else be clearly reported and understood regarding any
variance from the method or apparatus given in the existing Standards. Where ‘CIV’ is
specified or expected, the apparatus and method should be in accordance with
Method 6.9.1 or ASTM D5874.
The peak force, in Newtons, as related to the output in terms of Cmax of the free-falling
hammer is according to the following equation:
N peak = hammer mass (kg) × G max (Cmax × 10) × g (9.80665 m/sec/sec)
For example, an output of 30 C max is equivalent to 13.39 kNpeak where the hammer mass is
4.55 kg.
The energy per unit area may be represented by the formula—
( )
0.5 m v i2 / (A )
where
m = mass, in kilograms
vi = impact velocity, in metres per second
A = surface area, in millimetres squared
The velocity at impact can either be measured or else calculated using the equation
½
vi = (2gh) . With the hammer diameter for the standard Clegg hammer being 50 mm, the
mass as 4.55 kg, the drop-height as 0.455 m and the velocity at impact as 2.99 m/sec as
2
derived from the above formula, the energy per unit area is equal to 10.34 kN/m .
In part, to support the theory that CIV relates to soil stiffness, research for comparing CIV to
the well-recognised CBR test was initiated by Clegg. In 1986, he presented a revised
equation for a correlation between CBR and CIV (for non-soaked, no surcharge conditions
and using the 4.5 kg Clegg impact soil tester from a 0.45 m drop-height with the highest
value blow of four successive blows used for the CIV). The revised equation of 1986 is
2
similar to the initial correlation he reported in 1980, which was CBR(%) = 0.07 ((IV) ). The
revised equation confirmed the previous general relationship but with a slight correction at
2
the lower end of the scale. The revised equation is CBR(%) = (0.24 (CIV) +1) with
2
‘r’ = 0.957 (highly significant), or alternatively CBR(%) = 0.06 ((CIV) ) + 0.52 (CIV) +1. For
example, the CIV for a field CBR of 100% using the second equation above correlates to a
CIV of 38.

HB 160—2006 102
Regarding the relationship between a CBR value and CIV, Clegg (1986) stated: ‘These
revised equations appear appropriate for general use. However, since CBR is particularly
subject to high variability, even within one organisation, one soil type, etc., correlations
from individual sources may vary from the general equation. To avoid a change of
standards it is appropriate therefore that each organisation consider establishing its own
relationship for specific materials and conditions, particularly where there is strong reliance
on CBR for design purposes. It should be noted that these data are for both samples at
essentially the same density and moisture content and prepared in the same manner.
Attempts have been made to correlate unsoaked CIVs with soaked CBRs but with little
success. Also the question of the effect of surcharge may need to be considered.’

103 HB 160—2006
SEC T ION 7 S O IL R EAC T I VIT Y T EST S
7.1 GENERAL
The term ‘soil reactivity’ as used herein refers to the pronounced swelling-and-shrinking
characteristics of some cohesive soils. It is defined as the rate of strain of the soil sample
with change in soil suction.
AS 2870 defines a reactive site as a site consisting of a clay soil that swells on wetting and
shrinks on drying by an amount that can damage buildings on light strip footings or
unstiffened slabs.
The estimation of the movement of the surface of a reactive site caused by moisture
changes is discussed in AS 2870 Supplement 1 (Clause C2.2.3). The estimation is based
on a soil suction model, which requires the change in soil suction profile and an instability
index of the soil. AS 2870 recommends three tests in the AS 1289 series which can be used
in the estimation of the instability index of a reactive soil, viz.:
• Method 7.1.1, Determination of the shrinkage index of a soil—Shrink swell index
• Method 7.1.2, Determination of the shrinkage index of a soil—Loaded shrinkage index
• Method 7.1.3, Determination of the shrinkage index of a soil—Core shrinkage index

The background, applicability and reliability of each of the three tests are described and
discussed in detail by Cameron (1989). Guidance on the estimation of instability index from
shrinkage index is given in Appendix F of AS 2870. Cameron and Yttrup (1992), Thorne
(1985), Mitchell and Avalle (1985) and McInnes (1999) discuss some engineering properties
and behaviour of reactive soils as found in Melbourne, Sydney, Adelaide and south west
Western Australia respectively.
7.2 METHOD 7.1.1—SHRINK SWELL

7.2.1 General
Method 7.1.1 is considered to be the most reliable test and does not require the
determination of suction on a specimen of the soil. Another advantage is that a
determination can be made from either wet or dry samples.
7.2.2 Procedure
The shrink-swell test requires two core samples, preferably of approximately 50 mm
diameter. The first (the shrinkage sample) is allowed to shrink to laboratory conditions to
avoid severe cracking, then to oven-dry conditions and the ultimate shrinkage strain (ε sho) is
recorded. The second (swell sample) is placed in a consolidometer under a minimum
pressure of 25 kPa. After allowing initial settling in, the displacement gauge is zeroed and
distilled water is added. At completion of primary swelling, the final strain (ε sw ) is recorded.
Note that since this is swelling under both load and lateral restraint, unrestrained swell is
estimated by dividing the recorded swell by two. This factor of two to allow for restraint was
originally estimated by the committee but has subsequently been confirmed by tests such
as those by Cameron (1989) and Fityus (1998) (see Figure 7.2.2).
7.2.3 Result
Derivation of the shrink-swell index requires an assumption of the effective range of suction
over which swelling and shrinking is significant and a value of 1.8 pF has been adopted. It
has been proposed by McKeen (1992) that further drying of soils past 5.5 pF will not result
in any significant shrinkage, i.e., 5.5 pF corresponds to the shrinkage limit. Like most
generalisations, it is not applicable to all soils, as is evident from Cameron’s study. If 5.5 pF
is a reasonable average value, then in effect the wet limit is assumed to be 3.7 pF.

HB 160—2006 104
Swell on saturationD H
(distilled water added) e
25 kPa
D H
H
Swell
Natural
water
content
Water content w
Shrink
Shrink on dr yD L
Shrink-swell
testing
apparatus D L
FIGURE 7.2.2 SHRINK-SWELL TESTING APPARATUS

The shrink-swell index may be derived from—

I ss = (ε sw / 2 + ε sho ) / 1.8% / pF
Note that the minimum value of εsw is zero.

7.2.4 Accuracy
Obviously there are approximations involved in the derivation, but overall it seems to be the
most reliable test. It results in a coefficient of variation in predicting movements of about
25% compared with 40% for most other methods. Since the same assumptions in the test
are also made in the calculation of ground movement, to some extent any errors should
cancel out. Subtle effects such as hysteresis, load effects and non-linearity are ignored but
the overall accuracy is acceptable for the design of residential footing systems.
7.2.5 Notation
The general term for shrinkage index ignoring load is Ips while the value of I ps for a specific
test may be given a more specific subscript such as I ss for shrink-swell.
7.3 METHODS 7.1.2 AND 7.1.3—CORE SHRINKAGE INDEX

7.3.1 General
Methods 7.1.2 and 7.1.3 require the determination of suction on a specimen of the soil.
7.3.2 Procedure
The shrinkage index is determined by allowing a core sample to shrink under air-drying
conditions (see Figures 7.3.2(A) and 7.3.2(B)). The sample is measured and weighed
periodically allowing the shrinkage strain (ε in percent) to be plotted against the water
content (w) with the slope of the linear section being found as (∆ε/∆w).

105 HB 160—2006
FIGURE 7.3.2(A) CORE SHRINK TEST SPECIMEN
7.3.3 Water content/suction

To convert water content to suction the slope of the water content versus suction curve is
needed. It can be found by firstly determining the initial soil suction and water content of the
sample and then equilibrating two samples at known humidity. At mass equilibrium, the soil
water content is determined and the end suction assumed from chemistry or measured if
less than 5pFo. Since two suctions and two moisture contents are then known, a water
content-suction relationship or moisture characteristic can be found, i.e., (∆w/∆u).
Equilibration at a specific humidity can be achieved using large glass sealed containers
(desiccators) containing saturated salts that are known from chemistry to produce the
required humidity, and hence final suction at the temperature of the controlled room.
Alternatively, acid or salt solutions at defined molar concentrations can be used.
Equilibration times are long and depend on the size of the sample. This lengthy process can
be circumvented by establishing a characteristic suction at zero moisture content for a
given soil type. Mitchell and Avalle (1984) proposed a value of 6.8 pF based on limited
testing of Adelaide soils, but Cameron (1989) showed that this value could not be adopted
confidently for all soils.
Thus Ics, the core shrinkage index is given by—
I cs = (∆ε / ∆w ) ⋅ (∆w / ∆u )% / pF
7.3.4 Restraint and load effects

The restraint and load effects test does not include any allowance for lateral restraint or
load and is, therefore, an estimate of I ps . Although simple to perform, it has not proved as
accurate as the shrink-swell method. It is also dependent on sampling at a suitable
moisture content, i.e., not too dry.

HB 160—2006 106
ε
L = 100 mm
Water content
ε
w
50 mm
Shrinkage limit
L
Water content w
Suction (pF)
w
u
Water content w
FIGURE 7.3.2(B) CORE SHRINKAGE INDEX TESTING

7.3.5 Loaded shrinkage index

An alternative to the core shrink test is the loaded shrinkage index, which uses a much
smaller sample, held under a minimum 25 kPa vertical load (see Figure 7.3.5). The sample
is dried to a defined level in a desiccator. Initial and final suctions are measured, the
equilibrium suction being about 4.5 pF, which approaches the maximum dry field limit. The
test requires considerable time to perform (6 weeks minimum) and seems to have little
advantage other than the control of drying and the restraint of the sample while being
tested.
FIGURE 7.3.5 LOADED SHRINKAGE INDEX TESTING SHOWING SAMPLE

UNDER LOAD

107 HB 160—2006
APPENDIX A
APPROXIMATE TIME REQUIREMENTS FOR TESTS
Clause number of AS test method Actual test Lead time

this Handbook (AS 1289) (hours) (working days)
2.2 2.1.1 0.5 1

2.2 2.1.2 — —
2.2 2.1.4 — —
2.2 2.1.5 — —
2.2 2.1.6 — —
2.4 2.2.1 2 2
2.3 2.3.1 2 5
3.3.3 3.1.1 2 3
3.3.3 3.1.2 2 3
3.4 3.2.1 2 3–5
3.5 3.3.1 2 3–5
— 3.3.2 2 3–5
3.6 3.4.1 2 3–5
3.7 3.5.1 2 2
3.7 3.5.2 2 2
— 3.6.1 2 3
— 3.6.2 5 3
3.8.3 3.6.3 4 3
3.9 3.7.1 1 1
3.10.2 3.8.1 1.5 1–3
3.10.3 3.8.2 4 3
3.10.4 3.8.3 1 2
3.3.2 3.9.1 2 3
3.3.2 3.9.2 — —
4.2 4.1.1 — —
4.3 4.2.1 — —
4.4.3 4.3.1 — —
4.5.2 4.4.1 1.5 1–3
5.4.4 5.1.1 1 3
5.4.4 5.2.1 2 3
5.5.1 5.3.1 2 1
5.5.1 5.3.2 2 1
5.5.1 5.3.5 3 2

HB 160—2006 108
Clause number of AS test method Actual test Lead time

this Handbook (AS 1289) (hours) (working days)
5.6.2 5.4.1 3 3
5.6.2 5.4.2 12 5
5.4.5 5.5.1 2 3
5.6.2 5.6.1 2 3
5.6.2 5.7.1 3 3
5.5.3 5.8.1 0.5 1
— 5.8.4 48 7
6.1.2 6.1.1 2 7
6.1.3 6.1.2 2 7
6.1.3 6.1.3 1 1
6.2 6.2.1 0.5 1
— 6.2.2 3 2–10
6.3 6.3.1 1–3 1

6.4 6.3.2 0.5 1
6.4 6.3.3 0.5 1
— 6.4.1 0.5 1
— 6.4.2 3 7
6.6.5 6.6.1 — —
— 6.7.1 3 5
— 6.7.2 3 7
— 6.7.3 3 10
— 6.8.1 0.008 —
6.8 6.9.1 0.5 1
— 7.1.1 0.5 7
7.3 7.1.2 2 3
7.3 7.1.3 2 7

109 HB 160—2006
APPENDIX B
BIBLIOGRAPHY
AASHTO T251—Soil investigation and sampling by hollow-stem auger borings
Akroyd, T.N.W., Laboratory testing in soil engineering, William Clowes & Sons Ltd, London
& Beccles 1957
Austroads (1998a). Guide to stabilisation in roadworks. Austroads, Sydney
Austroads (2002). Mix design for pavement materials stabilised in situ (Flow charts) AP-T16
Sydney
Austroads, (2004). Pavement design—A guide to the structural design of road pavements,
Section 5.5.2. Austroads Inc. Sydney, Australia
Austroads AP-T29/03 Development of Performance-Based Specifications for Unbound
Granular Materials, Part A: Issues and Recommendations
Austroads AP-T30/03 Development of Performance-based Specifications for Unbound
Granular Materials, Part B: Use of RLT Test to Predict Performance
Austroads Pavement Materials, Part 1—Search, National Association of Australian State

Road Authorities, 1982
AustStab (2002). Lime stabilisation practice AustStab Technical Note, Australian
Stabilisation Industry Association, Artarmon, NSW
[www.auststab.com.au/technotes/TNote1.pdf]
Bishop, A.W. and Henkel, D.J. (1962). The measurement of soil properties in the triaxial
test. Edward Arnold, London, Second Edition
Bjerrum, L. (1973). Problems of Soil Mechanics and Construction on Soft Soils (Collapsible,
Expansive and Others). Proceedings of the Eighth International Mechanics and Foundation
Engineering, Moscow, Vol. 3, pp111-159
Bowles J.E. (1992), Engineering Properties of Soils and Their Measurement, McGraw-Hill
Cameron, D.A. (1989). Tests for reactivity and prediction of ground movement. Australian
Civil Engineering Transactions, I.E. Aust., Vol. CE31, No. 3, December 1989, pp121–132
Cameron, D.A. and Yttrup, P.J. (1992). Footings for light structures on the reactive soils of
Melbourne, in Engineering Geology of Melbourne, Peck W.A, Neilson, J.L, Olds, R.J and
Seddon, K.D. (ed.), Balkema, 1992, pp534–354
Casagrande, A. (1932). Research on the Atterburg Limits of Soil. Public Roads 13: 121-130
Clegg (1978), (1980).
Crouch, R.J. et al (2000). Soils and their use in earthworks in ‘Soils—Their properties and
management’ edited by P.E.V. Charman & B.W. Murphy, Oxford Univ Press 2nd edition
Earth Manual A Water Resources Publication. 2nd Edition. 1985 and reprint 1990—United
States Department of the Interior Bureau of Reclamation
Emerson, W.W. (1967). A classification of soil aggregates based on their coherence in
water Aust Journal of Soil Research 5, 47-57
Fityus, S.G. (1998). Calibration of the blunt tipped dynamic penetrometer for silica sands.
Proceedings of the International Conference on Site Characterisation (ISC ‘98), Atlanta,
USA, eds. P.K. Robertson, P.W. Mayne, 927–932
Fredlund.G. & Rahardjo, H. (1993). Soil mechanics for unsaturated soils, John Wiley &
Sons, Inc.

HB 160—2006 110
Glick, G.L. and Clegg, B. (1965). Use of a penetrometer for site investigation and
compaction control at Perth WA, Civil Engineering Transactions. Institution of Engineers,
Australia
Handbook of Corrosion Engineering, McGraw-Hill
Head, K.H. (1992). Manual of Soil laboratory testing, Pentech Press, London, Second
Edition (3 Vols)
Hilf, J.W. (1966). A rapid method of construction control for embankments of cohesive soils.
US Department of the Interior, Water Resources Technical Publications, Engineering
Monograph No. 26, 1966
Liao, S.S. and Whitman, R.V. (1986). Overburden Correction Factors for Sands. JGED
Vol. 112, No. 3, March, pp373–377
Loveday, J. editor (1974). Aggregate stability in ‘Methods for analysis of irrigated soils’
Tech Communication No. 54 of Commonwealth Bureau of Soils, United Kingdom
McInnes, D. (1999). Some aspects of clay soils in South-West Western Australia. Aspects
of Site Classification for Low Rise Building Developments in South-West of WA, Australian
Geomechanics Society, Bunbury, WA, June
McKeen, R. G. (1992). A Model for Predicting Expansive Soil Behavior. Proc. 7th Int. Conf.
on Expansive Soils, Dallas, Texas, pp1–6

Mitchell, P.W. and Avalle, D.L. (1984) A technique to predict expansive soil movements.
Proceedings 5th International Conference on Expansive Soils, Adelaide, May
NATA Technical Note 32 Use of nuclear gauges for testing soils and asphalt
Rendulic, L. (1936) Relation between void ratio and effective principal stresses for a
remoulded silty clay, 1st International Conference on Soil Mechanics and Foundation
Engineering Vol. III, 48-53
RTA (NSW) Test method, T 1 69
Scala, A.J. (1956). Simple methods of flexible pavements design using cone penetrometers.
Proceedings of the Second Aust-New Zealand Conf. Soil Mechanics and Foundation
Engineering, Christchurch, NZ, January 1956, Wellington, New Zealand Institution of
Engineers, p73
Sherrard, J.L. et al (1976). Pinhole test for identifying dispersive soils. J Geotech Eng Div,
ASCE, Paper 11846, Jan
Terzaghi, K. (1943). Theoretical soil mechanics, John Wiley & Sons, New York
Thorne, C.P. (1985). Residual and tertiary clays, properties and engineering behaviour, in
Engineering Geology of the Sydney Region, Pells, P.J.N. (ed.), Balkema, 1985, pp49-63
VicRoads (2000). Determination of the maximum allowable working time for a cementitious
binder. VicRoads Test Method RC 330.02, Kew
Walkley, A. (1946). A critical examination of a rapid method for determining organic carbon
in soils—Effect of variations in digestion conditions and of inorganic soil constituents. Soil
Sciences 63:251–263
Walkley, A. and Black, I.A. (1934). An examination of the Degtjareff Method for determining
soil organic matter and a proposed modification of the chromic acid titration method. Soil
Sciences 37:29–38
Winterkorn, H.F. and Fang, H.Y. (1975). Foundation engineering Handbook, van Nostrand
Reinhold

111 HB 160—2006
STANDARDS
AS 1141.2—Methods for sampling and testing aggregates—Basic test equipment
AS 1141.3.1—Methods for sampling and testing aggregates—Sampling—Aggregates
AS 1141.3.3—Methods for sampling and testing aggregates—Sampling—Preparation of
stabilized pavement materials
AS 1141.51—Methods for sampling and testing aggregates—Unconfined compressive
strength of compacted materials
AS 1289—Methods of testing soils for engineering purposes (all parts)
AS 1726—Geotechnical site investigations
ASTM D2487—Standard Classification of Soils for Engineering Purposes (Unified Soil
Classification System)
ASTM D5874—Standard test method for the determination of the impact value (IV) of a soil
ISO/TS 17892-4:2004—Geotechnical investigation and testing—Laboratory testing of soil,
Part 4: Determination of particle size distribution

HB 160—2006
112
NOTES
113
NOTES
HB 160—2006
HB 160—2006
114
NOTES
ISBN 0 7337 7494 6 3TANDARDS$EVELOPMENT 3ALESAND$ISTRIBUTION
Standards Australia SAI Global
GPO Box 476 Phone: 13 12 42
Sydney NSW 2001 Fax: 1300 65 49 49
Phone: 02 8206 6000 Email: sales@sai-global.com
Fax: 02 8206 6001
Email: mail@standards.org.au
Internet: www.standards.org.au
This page has been left intentionally blank.

HB160 2006

Uploaded by

Copyright:

Available Formats

HB160 2006

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

HB160 2006

Uploaded by

Copyright:

Available Formats

Accessed by SWINBURNE UNIVERSITY OF TECHNOLOGY on 07 Oct 2015 (Document currency not guaranteed when printed)

First published as HB 160—2006.

THE NEED FOR TESTING

• AS 5100—Bridge design (all parts)

SECTION 1 SOIL SAMPLING AND PREPARATION

SECTION 2 SOIL MOISTURE CONTENT

2.4 TOTAL SOIL SUCTION (AS 1289.2.2.1) ................................................................... 17

SECTION 3 SOIL CLASSIFICATION TESTS

SECTION 4 SOIL CHEMICAL TESTS

SECTION 5 SOIL COMPACTION AND DENSITY TESTS

SECTION 6 SOIL STRENGTH AND CONSOLIDATION TESTS

SECTION 7 SOIL REACTIVITY TESTS

S E CT ION 1 SO IL SAM PLI N G AN D

able to perform and supervise sampling.

1.2 TYPES OF SAMPLES

www.standards.com.au  Standards Australia

Method 6.4.1—Soil strength and consolidation tests—Determination of the

1.3 SAMPLING TECHNIQUES

 Standards Australia www.standards.com.au

1.4 FIELD TESTS PERFORMED WHILST SAMPLING

www.standards.com.au  Standards Australia

(b) Dry strength (crushing characteristics):

1.5 IDENTIFICATION AND EXAMINATION OF SAMPLES

 Standards Australia www.standards.com.au

1.6 SELECTION OF SITES

www.standards.com.au  Standards Australia

1.7 SIZE OF SAMPLES

Soil identification and natural Cohesive soils and sands 1

Comprehensive examination of Cohesive soils and sands 25–50

 Standards Australia www.standards.com.au

1.8 PREPARATION OF DISTURBED SAMPLES (AS 1289.1.1)

FIGURE 1.8.1(A) ROTARY CONE SPLITTER (EXTERIOR)

www.standards.com.au  Standards Australia

FIGURE 1.8.1(B) ROTARY SPLITTER (INTERNAL)

FIGURE 1.8.1(C) SAMPLE DIVIDER

 Standards Australia www.standards.com.au

FIGURE 1.8.1(D) SOIL GRATER

1.8.2 Preliminary preparation

www.standards.com.au  Standards Australia

1.8.3 Drying of bulk samples

drastically reduced to the correct amount of representative sample required, or where a

 Standards Australia www.standards.com.au

(b) Dampen material that tends to segregate.

www.standards.com.au  Standards Australia

 Standards Australia www.standards.com.au

1.8.5.6 Preparation by mortar and rubber-covered pestle

www.standards.com.au  Standards Australia

1.8.7.2 Wet preparation

 Standards Australia www.standards.com.au

levels of compaction of the specimen. Whilst it is recognised that various regions of

www.standards.com.au  Standards Australia

SEC T ION 2 S O IL M O IST U R E C O NT ENT

2.2 MOISTURE CONTENT TESTS (AS 1289.2.1.1 to AS 1289.2.1.6)

moisture condition needs to be known.

 Standards Australia www.standards.com.au

2.2.6 Soils that absorb atmospheric water

standard oven-drying method, AS 1289.2.1.1.

2.3 CORRELATION OF TEST RESULTS (AS 1289.2.3.1)

2.4 TOTAL SOIL SUCTION (AS 1289.2.2.1)