Cold Recycling

Wirtgen Cold Recycling

Technology
Cold Recycling
Wirtgen Cold Recycling
Technology
 Wirtgen GmbH
Reinhard-Wirtgen-Strasse 2 53578 Windhagen Germany

Phone: +49 (0) 26 45/131-0

Fax: +49 (0) 26 45/131-242

1st edition 2012

Copyright 2012 by Wirtgen GmbH.

All rights reserved. No part of this publication may
be reproduced, stored in a retrieval system or
transmitted in any form without the prior written
permission of Wirtgen GmbH.
Acknowledgements

This first Edition of the Wirtgen Cold Recycling are evident throughout the manual, particularly in
Technology manual was compiled by a team of those chapters concerned with pavement design
specialists with extensive experience in all aspects and stabilising agents. In addition, engineers from
of pavement rehabilitation, especially those relat- Wirtgen GmbH provided valuable guidance in
ing to the reuse of material from existing road addressing the shortcomings of this publications
pavements. predecessor, the Second Edition of the Wirtgen
Cold Recycling Manual and concerns previously
The team included Engineers from Loudon voiced by customers and field engineers have
International who have assisted Wirtgen and their been addressed in this new publication.
customers in the application of cold recycling
technology for nearly twenty years. In recognition Wirtgen GmbH are grateful to all who contributed
of rapid advances being made in the specialised to this manual and invite feedback from those who
arena of pavement engineering, an invitation to read it. Any comments will be welcomed, regard-
join the team was extended to selected academics less of the nature of such comments.
from some leading universities, especially Stel- These should be sent to [email protected]
lenbosch University. Their invaluable contributions
Preface

For the last twenty years, Wirtgen has been at Experience gained from these projects has allowed
the forefront of the development of cold recycling Wirtgen to actively push the barriers of technol-
technology. During this time, both the technology ogy by investing in research and development
itself and the machines that carry out the recycling and trialling new ideas and concepts (e.g. bitumen
work have evolved from simple beginnings to the stabilisation). These developments have all been of
current status where cold recycling is recognised benefit to the global industry.
worldwide as a normal process for constructing
pavement layers, particularly for rehabilitating This 1st edition of Wirtgen Cold Recycling
distressed pavements. Technology is a compilation of lessons learned
over the past twenty years. It includes everything
Cold recycling technology is currently employed that is needed to understand the technology,
to construct all types of pavements, ranging from what it is, where it can be applied and how to
minor access roads to major multi-lane highways. design pavements that incorporate cold recycled
Where a pavement is in a distressed state, there materials. It is particularly useful for those with little
is always an option to recycle the existing material experience in recycling and wish to learn about
and reap the benefits in terms of lower construc- the technology. However, it is also useful for the
tion costs, improved durability (service life) and, more experienced practitioner since it includes
equally important, a significant reduction in the advances achieved from the most recent research
negative impact that construction has on the efforts, especially those in the exciting field of bitu-
environment. men stabilisation.
In this regard, the technology has evolved from
The cold recycling process has been success- recycling materials that include mixtures of granu-
fully used to rehabilitate and/or upgrade many lar, cemented and asphalt materials to recycling
thousands of kilometres of roads during the material comprised entirely of recycled asphalt
past twenty years. The list of projects on which pavement (RAP) material.
Wirtgen recyclers have been deployed worldwide
is exhaustive and covers all climatic regions on all In publishing this new manual, Wirtgen GmbH wish
continents (except Antarctica where there are no to share their knowledge and understanding of
roads). Pavements in both the Developed and Un- cold recycling, not only with their customers who
developed Worlds are increasingly being recycled have provided many of the experiences, but also
as a solution to the deteriorating condition of their with the global road construction community, in
road networks. the belief that sharing is the path to advancement
and a brighter future for all.
Glossary of abbreviations

AADT Average annual daily traffic (Appendix 2)

AASHTO American association of state highway and transportation officials
ADE Average daily equivalent traffic (Appendix 2)
BSM Bitumen stabilised material (Chapter 4)
BSM-emulsion BSM made with bitumen emulsion (Section 4.1.4)
BSM-foam BSM made with foamed bitumen (Section 4.1.4)
CIR Cold in place recycling (Chapter 6)
CBR California bearing ratio
CTB Cement treated base
DCP Dynamic cone penetrometer (Section 2.5.4)
ELTS Effective long term stiffness (Section 2.6.4)
EMC Equilibrium moisture content
ESAL Equivalent standard axle load (80 kN) (Appendix 2)
FWD Falling weight deflectometer (Section 2.4.1)
GCS Graded crushed stone material
HMA Hot mixed asphalt
HVS Heavy vehicle simulator
ITS Indirect tensile strength (Appendix 1)
LTPP Long term pavement performance
MDD Maximum dry density
OMC Optimum moisture content
Pen Penetration grade (bitumen standard test)
PMS Pavement management system
PI Plasticity index
PN Pavement number (Section 2.6.4)
PWoC Present worth of cost (Appendix 4)
RAP Reclaimed asphalt pavement (asphalt millings)
SN Structural number (Section 2.6.3)
TSR Tensile strength retained (Section 4.3.11)
UCS Unconfined compressive strength (Appendix 1)
UTFC Ultra-thin friction course (asphalt surfacing layer)
WAM Warm mix asphalt
Content

1 Introduction  10

1 Road Pavements  15

1.1 Pavement structures 16

1.2 Pavement components 18
1.2.1 Surfacing  18
1.2.2 Structural layers  19
1.2.3 Subgrade 21
1.3 Primary considerations for the pavement structure 22
1.3.1 Environmental conditions 23
1.3.2 Traffic loading 25
1.4 Pavement distress mechanisms 26
1.4.1 Advanced pavement distress 26
1.5 Pavement maintenance and structural rehabilitation 28
1.6 Rehabilitation options 30
1.6.1 Surface rehabilitation 31
1.6.2 Structural rehabilitation 34

2 Pavement rehabilitation 38

2.1 General 41
2.2 Pavement rehabilitation: investigation and design procedure 42
2.3 STEP 1: Data acquisition/process available information 44
2.3.1 Information on the existing pavement (historical information)45
2.3.2 Design traffic 46
2.4 STEP 2: Preliminary investigations 48
2.4.1 Determination of uniform sections 49
2.4.2 Visual inspection 52
2.4.3 Reassessment of uniform sections 54
2.5 STEP 3: Detailed investigations 55
2.5.1 Excavating test pits 55
2.5.2 Laboratory testing 56
2.5.3 Extracting core specimens 57
 2.5.4 Dynamic cone penetrometer (DCP) probes 58
2.5.5 Analysis of deflection measurements 60
2.5.6 Rut depth measurements 60
2.5.7 Synthesis of all available data 61
2.6 STEP 4: Preliminary pavement rehabilitation design options 62
2.6.1 Pavement design approach 62
2.6.2 Catalogue design methods 63
2.6.3 Structural number method 63
2.6.4 Pavement number method 64
2.6.5 Mechanistic design methods 66
2.6.6 Deflection based methods 67
2.6.7 Summary of pavement design approaches 67
2.7 STEP 5: Laboratory mix design 68
2.8 STEP 6: Finalise pavement design options 70
2.9 STEP 7: Economic analyses 71

3 Cold recycling 73

3.1 General 75
3.2 The cold recycling process 77
3.2.1 In-plant recycling 78
3.2.2 In-place recycling 79
3.3 In-place recycling machines 84
3.4 Cold recycling applications 90
3.4.1 100% RAP recycling 93
3.4.2 Blend of RAP/Granular material 94
3.5 Benefits of cold recycling 97
3.6 Applicability of the cold recycling process 98

4 Stabilising agents 100

4.1 Types of stabilisation agents  103

4.1.1 General 103
4.1.2 Material behaviour 104
Content

4.1.3 Cementitious stabilising agents 105

4.1.4 Bitumen stabilising agents 106
4.1.5 Summary of different stabilising agents 109
4.2 Stabilising with cement 110
4.2.1 General  110
4.2.2 Factors affecting strength 110
4.2.3 Cracking of cement stabilised layers 111
4.2.4 Surface crushing 114
4.2.5 Durability concerns 115
4.2.6 Working with cement 116
4.2.7 Early trafficking 120
4.2.8 Key features of cement stabilised materials 121
4.3 Stabilising with bitumen  123
4.3.1 Overview 123
4.3.2 BSM Distress Mechanisms 126
4.3.3 Primary determinants of BSM performance 127
4.3.4 Material to be stabilised with bitumen 128
4.3.5 Bitumen stabilising agents 136
4.3.6 Active filler 140
4.3.7 Water quality 141
4.3.8 Mix design procedure 142
4.3.9 Classification of BSMs 144
4.3.10 Working with BSMs 146
4.3.11 Mechanical tests 152
4.3.12 Pavement design approaches for BSMs 154
4.4 Summary: Advantages and disadvantages of cement and bitumen stabilising agents 160

5 Recycling Solutions 163

5.1 Guidelines for recycling different pavements 165

5.1.1 Lightly trafficked roads (structural capacity: 0.3 million ESALs) 166
5.1.2 Low volume roads (structural capacity: 1 million ESALs) 168
5.1.3 Secondary rural roads (structural capacity: 3 million ESALs) 170
 5.1.4 Main rural roads (structural capacity: 10 million ESALs) 172
5.1.5 Interurban highways (structural capacity: 30 million ESALs) 174
5.1.6 Major multi-lane highways (structural capacity: 100 million ESALs) 176
5.2 Alternatives for pavement rehabilitation 178
5.2.1 Existing pavement 180
5.2.2 Rehabilitation requirements 181
5.2.3 Rehabilitation options 182
5.2.4 Maintenance requirements 190
5.2.5 Construction & maintenance costs 192
5.2.6 Energy consumption 195
5.2.7 Relevant comments 199

6 Recycling 100% reclaimed asphalt pavement (RAP) material 201

6.1 RAP material 203

6.1.1 Bitumen binder 203
6.1.2 Grading of the RAP material 205
6.2 Uses for cold recycled RAP material 206
6.2.1 Untreated RAP material 206
6.2.2 RAP material treated with cement 207
6.2.3 RAP material treated with bitumen emulsion 207
6.2.4 RAP material treated with foamed bitumen 211

Bibliography214

Appendix 1 Laboratory procedures for stabilised materials (mix designs) 218

Appendix 2 Determining structural capacity from traffic information 307

Appendix 3 Guidelines for compiling specifications for recycling projects 320

Appendix 4 The principles of economic analysis 355

Introduction

The Wirtgen Cold Recycling Manual was first These developments and improvements effectively
published in 1998, in English. Due to advances in outdated some sections of the Second Edition.
recycling technology, it was necessary to rewrite This, coupled with the reference document sta-
the manual after six years and, accordingly, the tus enjoyed by the Manual warranted a thorough
Second Edition was published in 2004. review and update of the contents, a process that
The Second Edition was well received; within a highlighted the need for a total rewrite. In addition,
couple of years it had been translated into a dozen the information that needs to be included in the
different languages and, by the end of 2009, more Manual has grown exponentially and is now too
than 50,000 printed copies had been distributed much for a single publication. Accordingly, the
worldwide with at least the same number down- decision was taken to replace the manual with two
loaded from the www.wirtgen.de website. publications:

As with the First Edition, the Second Edition at- > Wirtgen Cold Recycling Technology that
tracted considerable attention with an increasing focuses on aspects of pavement theory and
number of reports, conference papers and other design that are relevant to cold recycling.
technical publications making reference to the Included is a detailed explanation of cold
manual. It would appear that the Second Edi- recycling and is particularly useful for engineers
tion of the Wirtgen Cold Recycling Manual has involved with material utilization and pavement
successfully built on the reputation established by design.
its predecessor as the primary reference book for
cold recycling technology. > Wirtgen Cold Recycling Application that
covers the practical aspects of applying the
Almost a decade has now passed since the technology. This separate publication describes
Second Edition was published. During this time, the various construction processes that can be
interest in recycling has intensified, reflected by used for cold recycling and is useful for practi-
the increasing number of recycling machines that tioners and field engineers
are sold around the world every year. This has
encouraged the research and development team As with the previous Cold Recycling Manuals,
at Wirtgen headquarters to continue to improve these new publications are focused on recycling
the machines they manufacture, based on feed- cold material for use in flexible pavements. They
back received from their global network of service do not include recycling hot material, nor do they
engineers and customers. Such an increase in consider rigid (concrete) pavements, both being
field activity succeeded in attracting interest from separate specialist fields. In addition, they do not
the academic community with the result that cold include half-warm and warm asphalt. Foamed
recycling technology took a giant step forward bitumen technology is ideally suited to such mixes
thanks to their research efforts. Since 2004, much but adaptations to mix evaluations and pavement
groundbreaking research work has been carried design are required and are not covered in this
out, especially concerning bitumen stabilisation, a manual.
technology that is ideally suited to cold recycling.
This 1st Edition of Wirtgen Cold Recycling been included. These developments have moved
Technology includes the following: the technology forward from that covered in the
Second Edition of the Wirtgen Cold Recycling
Chapter 1 provides an overview of pavements. Manual, triggering the need for a rewrite.
The composition of pavement structures is ex-
plained along with a brief description of the main Chapter 5, entitled Recycling Solutions, uses a
factors influencing the selection of the various catalogue design format to show a series of typi-
materials used to construct the different layers and cal pavement structures suitable for rehabilitation
how they behave (and deteriorate) when subjected by recycling, including both cement and bitumen
to dynamic wheel loads. This leads on to the sub- stabilisation options. Six traffic classes between
ject of pavement rehabilitation and introduces the 300,000 and 100 million equivalent standard axle
concept of cold recycling, both in situ and in plant. loads (ESALs) are included, each with different
subgrade support conditions that would normally
Chapter 2 focuses on pavement rehabilitation and be encountered for such a class. This is followed
describes the engineering input required to formu- by an example of different options that can be
late a suitable design, particularly those aspects used to rehabilitate a specific pavement with a
that are relevant for cold recycling. Pavement structural capacity requirement of 20 million
investigations, material analyses and pavement ESALs. An existing (distressed) pavement
designs are all covered in detail in a seven step structure is used to select four different rehabilita-
procedure culminating with a section on economic tion solutions as well as the maintenance require-
analyses to assist in evaluating the financial merits ments for each over a 20-year service life, together
of different rehabilitation options. with the relevant rehabilitation requirement after
20 years. The whole-of-life cost for each option is
Chapter 3 explains cold recycling and the various then evaluated using different discount rates.
applications that can be considered, both in situ In addition, the energy consumed by all con-
and in plant. The range of Wirtgen recycling struction activities during the service life of each
machines is introduced along with an explanation rehabilitation option is evaluated.
of the type of recycling best suited to each
machine. Also included is a summary of the Chapter 6 is focused on reusing 100% reclaimed
benefits to accrue from adopting a cold recycling asphalt pavement (RAP) material in a cold recycling
approach and the suitability of the process for process. This subject was not covered in any detail
constructing pavement layers, for both new roads in the previous manuals and has been included
and for rehabilitating distressed pavements. to address the increasing interest being shown
worldwide in recycling this specific type of material
Chapter 4 focuses on the stabilising agents that using a cold process. (The in situ recycling pro-
are normally applied in the cold recycling process. cess is known in some countries as cold in place
Mix and pavement design procedures for both recycling (CIR) or partial depth reclamation).
cementitious and bituminous stabilising agents are
explained in detailed. Recent developments in the A list of relevant Bibliography is included immedi-
field of bitumen stabilised materials (BSMs) have ately after Chapter 6.

10/11
The four appendices contain a host of additional countries are facing a steady decline in the stand-
information, all relevant to cold recycling, but to ard of their ageing road network.
include them in the chapters would make the
manual cumbersome. Ever increasing maintenance and rehabilitation
efforts required to retain acceptable levels
Appendix 1 describes the laboratory procedures of service place tremendous pressures on national
for stabilised materials (mix designs). This is fol- budgets. This situation is exacerbated by the glob-
lowed by a schedule of equipment required for al pattern of growing traffic volumes compounded
carrying out the laboratory work. by increasing axle loads and tyre pressures, fac-
tors that all contribute to pavement deterioration.
Appendix 2 describes the methodology used to This downward spiral can only be addressed by
determine the correct pavement design criteria a massive increase in road budgets coupled with
(structural capacity requirement) from traffic data. innovation in the field of pavement engineering.

Appendix 3 includes guidelines for compiling Since few road budgets are increasing in real
appropriate construction specifications for cold terms, focus is being placed on innovation to
recycling projects. achieve more with relatively less expenditure.
Recycling clearly falls into this category and
Appendix 4 provides useful background informa- records show that the number of lane-kilometres
tion for economic analyses. of distressed pavement being rehabilitated using
the cold recycling process is increasing annually.
Pavement rehabilitation is becoming more impor- Simple economics is the main reason for this phe-
tant as the overall condition of the worlds road nomenon as it reflects the cost effectiveness
infrastructure continues to deteriorate and many of the process.
12/13
1 Road Pavements

1.1 Pavement structures 16

1.2 Pavement components 18
1.2.1 Surfacing  18
1.2.2 Structural layers  19
1.2.3 Subgrade 21
1.3 Primary considerations for the pavement structure 22
1.3.1 Environmental conditions 23
1.3.2 Traffic loading 25
1.4 Pavement distress mechanisms 26
1.4.1 Advanced pavement distress 26
1.5 Pavement maintenance and structural rehabilitation 28
1.6 Rehabilitation options 30
1.6.1 Surface rehabilitation 31
1.6.2 Structural rehabilitation 34

14/15
1.1 Pavement structures

Road pavements comprise three basic properties of comfort, safety and environmental
components: consideration (e.g. low noise). The structural layers
distribute the high intensity surface loads gener-
> Surfacing: The riding surface which is usually
ated by traffic over a wider area of subgrade, as
the only part of a road that is visible.
illustrated in the figure below.
> Structural layers: The load spreading layers,
consisting of different materials, often extending The individual layers in a pavement structure vary
to depths in excess of one metre. in composition (material type) and thickness.
Those layers closest to the surface are construct-
> Subgrade: The existing earth upon which the ed using high-strength materials (e.g. hot-mix
road is built. asphalt) to accommodate the higher stresses.
Individual asphalt layers seldom exceed 100 mm
Subgrades are usually of relatively low bearing thickness. As the load is distributed over a wider
capacity and cannot carry traffic loading directly, area in the lower layers, the level of stress reduces
so protective overlying layers are needed. The and can be carried by poorer quality materials
purpose of the surfacing is predominantly func- (e.g. natural gravels or lightly cemented materials).
tional providing an all weather riding surface with Consequently, the materials in the lower layers are

Wheel load

Contact area

Surfacing

Load Structural layers

Transfer

Subgrade

Load Transfer through the pavement structure

generally inexpensive relative to those in the upper sually demolished at the end of their useful
u
layers. The thickness of these individual structural lives. Although this manual is concerned only
layers generally varies between 125 mm and with flexible pavements that are characterised by
250 mm. Section 1.2 below discusses the various bituminous surfaces, unreinforced concrete pave-
pavement components. ments have been successfully recycled in situ.
(Information on this specialised application can be
There are two fundamental types of pavement: obtained from Wirtgen.)
> Rigid pavements with a thick layer of high-
Once constructed, a road is subjected to det-
strength concrete overlying a bound layer; and
rimental forces emanating from two primary
> Flexible pavements constructed from natural sources, the environment and traffic. Both act
materials with the upper layers sometimes continuously to reduce the riding quality and
bound (normally by bitumen and/or lightly structural integrity. These destructive forces are
cemented) to achieve the higher strength discussed in Section 1.3. The sections that follow
requirements. describe pavement deterioration mechanisms
and what can and should be done to retard the
Generally, only flexible pavements can be process (maintenance), and measures to restore
economically recycled in-situ. Rigid pavements serviceability once deterioration has reached an
constructed from high-strength concrete are unacceptable level (structural rehabilitation).

Ideal candidate for in situ recycling

16/17
1.2 Pavement components

Each of a pavements three primary components different purpose, as explained below.

described above, serves a specific, although

1.2.1 Surfacing

The surfacing is the pavements interface with > abrasion of tyres on the surface, especially
the traffic and the environment, its function be- whilst cornering, tends to polish the surface.
ing to protect the pavement structure from both, In time this polishing effect reduces the friction
providing durability and waterproofing. properties (skid resistance) and texture depth
of the surface. Such surfaces become slippery,
Protection from traffic. Traffic affects the surfacing especially when wet, and can be hazardous.
in two ways:
> stresses imparted by wheel loads at the surface Protection from the environment. The surfacing
is continually being subjected to various forms of
are predominantly in the vertical plane, but hori-
attack from the environment.
zontal stresses can become significant, particularly
with the turning and braking actions of traffic and
Thermal effects, oxidation and ultraviolet radiation
on steep gradients. The strength and stiffness
are most aggressive. A surfacing therefore needs
characteristics of the material used in the surfac-
to have the following properties:
ing must be able to withstand all these stresses
without crushing or deforming; and

Upper layers Base course surface

Surface layer

Wearing course
Binder course
Base course
Subbase course
Capping layer and / or
selected subgrade
Embankment
or Fill Subgrade Stratum (top of subgrade)

Lower layers (select subgrade and fill)

Road design in embankment and cut

> elasticity to allow it to repeatedly expand and In addition to skid resistance, the bituminous sur-
contract as the temperature changes; and facing provides flexibility, durability and superior
waterproofing. Hot-mix asphalt (with a bitumen
> durability to absorb the daily bombardment of content of approximately 5% by mass) is generally
ultraviolet radiation, sporadic exposure to water used as a premium surfacing for heavily trafficked
and chemical effects whilst maintaining accept- roads, whilst the more economical chip-seal
able performance. surface treatments are applied where the traffic
volumes are lower.

1.2.2 Structural layers

The pavement structure transfers the load from the lower down and therefore need to be constructed
surface to the subgrade. As previously described, from stronger and stiffer material.
the stresses applied by a wheel at the surface The figure on the previous page shows the dif-
are effectively reduced within the pavement struc- ferent layers that are typically used to construct
ture by spreading them over a wide area of the flexible pavements.
subgrade.
The pavement structure generally consists of The response of a layer to an imposed load de-
several layers of material with different strength pends largely on the material properties (elasticity,
and stiffness characteristics, each layer serving plasticity and viscosity) and the characteristics of
the purpose of distributing the load it receives the load (magnitude, rate of loading, etc.).
at the top over a wider area at the bottom.
The layers in the upper part of the structure Flexible pavements are constructed from three
are subjected to higher stress levels than those types of material:

Moisture Aggregate
> Unbound (granular) materials, which include
crushed stone and gravels, transfer applied
loads through the individual particles, or
skeleton, of their matrix. Inter-particle friction
maintains structural integrity, but under repeated
loading (often associated with an increase
in moisture content), a gradual densification
process occurs as the particles re-orientate and
move closer together. This can occur at any
level in the pavement structure, ultimately result-
ing in deformation at the surface. Such deforma-
Air voids tion is normally manifest as wide radius rutting
in the wheel-paths.
Unbound Granular

18 / 19
> Bound materials, which include cement stabi- Aggregate
lised materials and asphalt, act more like a wide
beam. Applying a vertical load to the surface
of a beam generates horizontal compressive
stresses in the upper half of the beam and
horizontal tensile stress in the lower half, with
maximum horizontal stresses at the top and
bottom. The strain resulting from these stresses,
particularly the tensile strain at the bottom,
ultimately leads to a fatigue type of failure after
many load repetitions. Cracks develop at the Bitumen
bottom of the layer and then propagate vertically complete coating
as the load repetitions continue.
Bound Hot Mix Asphalt
> Non-continuously bound materials, which com-
prise bitumen stabilised materials (BSMs) with
either foamed bitumen or emulsified bitumen
Bitumen spotwelds Aggregate
as binders, behave like granular materials with
retained inter-particle friction but increased cohe-
sion, and stiffness. Permanent deformation is the
main mode of distress of BSMs. Bitumen is non-
continuously dispersed in these materials and
fatigue is therefore not a design consideration.

Deformation occurring in unbound and non-con-

tinuously bound material and fatigue cracking of
bound material are both related to the number of
load repetitions. This allows the functional life of a Moisture
pavement to be determined in terms of the number
of times it can be loaded before it fails, termed Non-continuously bound Bitumen stabilised
the structural capacity of the pavement.

Note:

Bitumen stabilised materials (BSMs) are

non-continuously bound
1.2.3 Subgrade

The natural material supporting a pavement This approach was first adopted in the 1950s
structure can be either in-situ material (cut condi- with the empirical California Bearing Ratio (CBR)
tion) or imported (fill condition). The strength cover design method and has endured into the
characteristics of this material dictate the type 21st century. In general, thick pavement structures
of pavement structure required to spread the are required to protect poor subgrades and such
applied surface load to a magnitude that can be thickening is often achieved by the addition of
supported without the subgrade failing due to selected subgrade, or capping layers.
permanent deformation.
In some cases, subgrades can comprise collaps-
Pavement design methods usually use subgrade ible soils, heaving clays, soft / consolidating clays
strength and stiffness as primary input parameters and dispersive / erosive soils. For such conditions,
and aim at providing a structure of sufficient thick- specialist geotechnical investigations, testing and
ness and strength to protect the subgrade. design is required.

20 / 21
1.3 Primary considerations
for the pavement structure

Roads are built throughout the world in all types The specific environmental conditions and antici-
of climate, from hot dry deserts to high rainfall pated traffic loading are the two primary structural
regions and icy tundra conditions. Yet, regardless design considerations for any pavement and
of the environmental conditions, every road is are discussed separately below. These factors
designed to withstand traffic loading by the same determine the pavement condition and rate of
fundamental mechanism of transferring the high deterioration. Generally, pavement deterioration
intensity forces imparted at the surface by the is measured indirectly by assessing riding quality,
wheel loads to lower levels that the subgrade can but obvious visible features such as rut depth and
accommodate without deforming. surface cracking are also relevant. Each mecha-
nism of distress has its own defining performance
function versus time as shown in the figure below.

Riding Quality

Terminal Condition
Pavement condition

Rutting

Cracking

Time

Evolution of Pavement Distress

1.3.1 Environmental conditions

Environmental conditions are considered separate- The pavement structure. Water is the great-
ly for the surfacing and the underlying structural est enemy of road structures. Water saturation
layers. causes materials to soften and deteriorate, and
also provides inter-particle lubrication when load
The surfacing. In addition to traffic, road surfaces is applied. The bearing capacity of a material in a
are exposed to sunshine, wind, rain, snow and dry state is significantly greater than in a wet state
other natural elements. Of importance are the con- and the more cohesive (or clayey) the material,
sequences of these elements on the engineering the more susceptible it is to moisture. In addition,
properties of the road surface which manifest in: water that is present when frost progresses into a
pavement structure will expand and cause exten-
sive damage when it thaws. Hence the importance
> thermal effects which cause changes in volume
of preventing water ingress into a pavement struc-
as materials expand and contract in response to
ture, especially into the poorer quality materials
changing temperatures. The daily temperature
found in the lower layers.
range of the road surface is important. In desert
areas, the surface of a blacktopped road can
experience a temperature range in excess of
70 C between dawn and noon, whereas road
surfaces inside the Arctic Circle during winter
will be buried under snow and ice and therefore
remain at a relatively constant temperature;
> freezing effects which create the phenomenon
known as frost heave. Repeated freeze/thaw
cycles can cause major damage to road sur-
faces;
> radiation effects which cause road surfaces to
experience a type of sunburn. The ultraviolet
radiation to which the road surface is subjected
causes the bitumen to oxidise and become brit-
tle. This process is known as ageing; and
> moisture effects where rainfall penetrates the
voids of the surfacing and builds up pore pres-
sures under wheel loading, breaking the bond
between the bitumen and the aggregate, leading
to stripping and ravelling of the asphalt.

22/23
Environmental factors are responsible for most Once the surface integrity has been lost due to
cracking that initiates at the surface. The major cracking, the pavement tends to deteriorate at an
contributor to this phenomenon is ultraviolet radia- accelerated rate due to water ingress. The primary
tion from sunlight that causes a continuous slow environmental factors effecting pavements are
hardening of the bitumen. With hardening comes a shown in the figure below.
reduction in elasticity that results in cracking when
the surface contracts as it cools or flexes under
wheel loads.

Radiation

Bitumen softening Stripping / water ingress / Frost heave

/ageing loss of shear strength

Environmental effects
1.3.2 Traffic loading

Roads are constructed to carry traffic. The volume Passenger cars typically have tyre pressures in
and type of traffic that a road is expected to carry the range of 180 to 250 kPa and support less than
dictates the geometric and structural require- 350 kg per tyre, or 7 kN on an axle. This loading
ments. Transportation engineers work with antici- is structurally insignificant when compared to that
pated traffic statistics (in terms of vehicle numbers, imparted by a large truck used for hauling heavy
composition and sizes) in order to determine loads, which range between 80 and 130 kN per
the geometric capacity requirements (alignment, axle (depending on legal limits and mass control)
number of lanes, etc.). Pavement engineers need with tyre pressures ranging between 500 to
anticipated traffic statistics (in terms of vehicle 1,300 kPa. Clearly, the loading of such heavy
numbers, configuration and axle mass) to deter- vehicles will have the greatest influence on the
mine the structural requirements. Accurate predic- strength requirements of a pavement and is there-
tions of future traffic volume and type is therefore fore discussed in Chapter 2, Pavement Rehabilita-
of paramount importance. tion and covered in detail in Appendix 2, Determin-
ing Structural Capacity from Traffic Information.
Important features of the traffic from a pavement
design perspective are those that allow definition
of the magnitude and frequency of surface loads
that the road can anticipate during the expected
life of the pavement. The load that is imparted on Axel load Tyre pressure
a road surface by a tyre is defined by three factors:

> force (in kN) actually carried by the tyre which,

together with Vehicle
speed
> inflation pressure (in kPa) determines the
footprint of the tyre on the road. This footprint
defines the area on the surface that is subjected
to the load, and Contact area

> speed of travel that defines the rate at which the Stress
pavement is loaded and unloaded. distribution

Traffic loading

24 / 25
1.4 Pavement distress mechanisms

Traffic loading is responsible for the development ously bound material, such a loss of voids leads
of ruts and for cracking that initiates within the to an increase in strength (denser materials are
bound layers. Every vehicle using a road causes stronger).
a small measure of deformation in the pavement In asphalt the converse applies. It must, how-
structure. The deformation caused by a light vehi- ever, be appreciated that a reduction in the void
cle is so small that it is insignificant whilst heavily content in asphalt not only causes rutting in the
loaded vehicles cause relatively large deforma- wheel paths, but it also allows the bitumen to
tions. The passage of many vehicles has a cu- start acting as a fluid when warm, creating a
mulative effect that gradually leads to permanent medium for hydraulic pressures to be gener-
deformation and/or fatigue cracking. Overloaded ated from the imposed wheel loads. This causes
axles cause a disproportionate amount of dam- lateral displacement, or shoving along the edges
age to the pavement structure, accelerating such of ruts; and
deterioration. This deterioration is caused by two
> fatigue cracking of bound materials. These initi-
different mechanisms, namely:
ate at the bottom of the layer where the tensile
strain caused by wheel loads is at its maximum.
> permanent deformation caused by densification, These cracks then propagate to the surface.
where stresses from repeated loading cause the Top down cracking can occur in thick asphalt
individual particles within the pavement layer to layers. Permanent deformation of the underly-
move closer together, resulting in a loss of voids ing material exacerbates cracking by effectively
or shearing of particles past each other (local- increasing the tensile strain imposed by wheel
ised shear failure). In granular and non-continu- loads.

1.4.1 Advanced pavement distress

Once a crack penetrates through the protective transmits predominantly vertical wheel loads into
surfacing, water can ingress into the underlying pressures that rapidly erode the structure of a
pavement structure. As previously described, granular material and causes stripping of bitumen
the softening effect of water leads to a reduction from the aggregate in asphalt. Under these condi-
in strength that results in an increased rate of dete- tions the fines fractions of the pavement material
rioration under repeated wheel loads. are expelled upwards through the cracks (known
as pumping) resulting in voids developing within
In addition, water in a saturated material becomes the pavement. Potholing and rapid pavement dete-
a destructive medium when the pavement comes rioration soon follows.
under load. Similar to a hydraulic fluid, the water
 of cracks, making for extremely poor riding
quality.

Another failure condition often seen in desert

environments is block cracking caused by an
extremely low moisture content in dense material.
This phenomenon is known as pore fluid suction
pressure. Due to the low relative humidity regime,
water is lost from the pavement structure due to
evaporation, reducing the moisture content to
levels similar to those achieved when air drying
samples in the laboratory. At such low moisture
contents, the menisci of the tiny water droplets
remaining within the small voids of a compacted
material exert sufficient tensile forces to cause the
material to crack.

This condition is likely to manifest wherever

Typical pavement distress with pumping relative humidity levels are low and the road is
unsealed, allowing the moisture in the pavement
Where temperatures drop below 4C, any free structure to evaporate. It has also been cited as
water in the pavement expands as it freezes, the cause of top-down cracking at high altitudes
creating hydraulic pressures, even in the absence (> 2,000 m). The only effective treatment for this
of imposed wheel loads. Frost heave caused condition is to seal the road so that the equilibrium
by repeated freeze/thaw cycles are the worst moisture content is retained (i.e. the hydro-genesis
scenario for a cracked pavement, resulting in rapid effect). If the material is allowed to dry out, severe
deterioration. and deep cracking will occur, even in compacted
sand. If the material is allowed to wet up, capillary
Under dry desert conditions, cracks in the surfac- forces will reduce and the apparent cohesion will
ing lead to a different type of problem. At night dissipate.
when temperatures are usually relatively low (often
below freezing) the surface contracts, causing the One further cause of surface cracking, particularly
cracks to widen and act as a haven for wind-blown in thin asphalt surfacings, results from the lack of
sand. When temperatures rise during the day, the traffic. The kneading action of traffic keeps bitu-
surface is restricted from expanding by the sand men alive. Oxidation and subsequent hardening
trapped within the crack, resulting in large horizon- causes thermal cracks to initiate at the surface
tal forces that cause localised failure (spalling) at of the bitumen binder. Subjecting the bitumen to
the edge of the crack. stress repetitions causes sufficient strain to close
the cracks as they form, thereby preventing them
These forces can ultimately lead to the surfacing from propagating.
lifting off the pavement structure in the vicinity

26/27
1.5 Pavement maintenance and
structural rehabilitation

Pavement maintenance activities are normally ment. Deformation and fatigue cracking caused
focused on keeping water out of and away from by traffic loading cannot be treated effectively
the pavement structure. This involves maintaining by superficial maintenance activities and require
the surfacing in a waterproof state and ensuring some form of structural rehabilitation.
that drainage measures are effective so that water
cannot pond on the road surface or along the road Pavement deterioration begins at a relatively slow
edge. pace. Pavement indicators can be used to monitor
the rate of deterioration. Road authorities often
Water normally ingresses the upper pavement employ a data-base system, known as a Pave-
structure through cracks in the surfacing, often ment Management System (PMS), to continuously
assisted by water ponding on the surface. Cracks monitor the riding quality of all road pavements
should therefore be sealed as they appear and within their network, thereby drawing attention to
road verges trimmed to promote runoff. If ad- those that most require attention. The figure below
dressed early, ageing effects can be effectively provides a typical PMS plot that illustrates the
treated by the application of a light fog spray of effectiveness of timely maintenance and rehabilita-
dilute bitumen emulsion. More serious condi- tion measures.
tions require a chip seal application where traffic
volumes are low, or a conventional hot-mix asphalt This figure highlights the importance of taking
overlay. timeous action to maintain as high a riding quality
as possible. The rate of deterioration is indicated
Such measures, which are aimed at maintaining by the riding quality. Poorer riding quality encour-
the flexibility and durability of the surfacing, only ages faster rates of deterioration through dynamic
really address deterioration due to the environ- loading. As the riding quality reduces, the scale of

Constructed
riding quality
Resurfacing Structural rehabilitation
Riding quality

Consequences of
not resurfacing
Terminal riding quality

Structural Design
Period

Time / Traffic

Managing pavement maintenance and rehabilitation by monitoring riding quality

remedial measures becomes greater, as does the until it is combined with an upgrading exercise to
cost of such measures. improve the geometrics of the road and add ad-
ditional lanes. Each rehabilitation decision needs
The decision as to which remedial measures to to be taken independently within the context of the
undertake to either improve a pavement or just overall road network. But, to do nothing and allow
maintain it at its current riding quality is often the pavement to deteriorate further is generally the
dictated by budgetary constraints. Short-term worst decision because of the exponential rate of
holding measures can be extremely cost effective. deterioration with time.
Pavement rehabilitation is sometimes postponed

28/29
1.6 Rehabilitation options

There are usually many options available for the re- The answers to these two questions will narrow
habilitation of a distressed road and sometimes it down the rehabilitation options to only those that
is difficult to determine which is the best. However, will be cost-effective within the context of, essen-
the answer to two important questions that must tially, the nature of the problem and the time frame.
be asked at the outset will assist in selecting the By separating the nature of the problem into two
correct one, the one that is most cost-effective categories (surface and structural) from the time
in meeting the road owners needs. The two impor- frame (short-term or long-term), selecting the best
tant questions are: option is simplified.

One other important point that affects the decision

> what is actually wrong with the existing pave-
is the practicality of various rehabilitation methods.
ment? A cursory survey consisting of a visual
Traffic accommodation, weather conditions and
inspection coupled with a few basic tests (eg.
availability of resources can all have a significant
deflection measurements) will normally be
influence on how a project is executed and may
sufficient to be able to understand the distress
preclude certain options.
mechanism. Of importance is to determine
whether the distress is confined to the surfacing
This whole exercise has one sole purpose:
(upper pavement layers) or whether there is a
determining the most cost-effective solution to the
structural problem; and secondly
actual problem within the context of the project
> what does the road authority really want and environment.
what can they afford? Is a 15-year design life
expected, or is a smaller capital outlay envis-
aged that will arrest the current rate of dete-
rioration and hold the pavement together for a
further five years?

Note:

Rehabilitation design must address the root

cause of distress cost effectively
1.6.1 Surface rehabilitation

Surface rehabilitation measures address prob- the road user. Modified binders are often used
lems that are confined to the uppermost part of in the asphalt to improve performance, thereby
the pavement, usually within the top 50 mm to extending the life of the overlay. Active cracks in
100 mm. These problems are normally related to the existing surface will reflect quickly through a
aging of bitumen and cracking that initiates at the new overlay and therefore need to be identified
surface due to thermal forces. and addressed either by applying a stress-re-
lieving bandage or by patching. Repeated over-
The most commonly used methods for dealing lays, however, increase road surface elevations
with this type of problem include: that can cause drainage and access problems.

> Asphalt overlay. Paving a thin (40 50 mm)

hot-mix asphalt overlay on the existing surface.
This is the simplest solution to a surface prob-
lem since the time required to complete the
work is short and there is minimal impact on

Existing pavement

Asphalt overlay

30 / 31
> Mill and replace. This method removes the of- high production capabilities of modern milling
fending cracked layer of asphalt and replaces it machines. The problem is removed with the
with fresh hot-mix asphalt, often with a modified layer of asphalt and pavement levels are main-
binder. The process is relatively fast due to the tained.

Mill off all sphalt Replace asphalt

Mill & Replace

> Recycling a relatively thin (100 mm 150 mm) plant by transporting milled material to a cold
layer of asphalt material from the existing pave- mixing plant KMA 220 or in place by using the
ment. Such recycling can be done either in- 2200 CR or WR 4200.

Mill off 150 mm asphalt Treat stockpiled RAP in KMA with

Place RAP in stockpile 2% foamed bitumen + 1% cement

Pave
Existing 110 mm Pave 40 mm
pavement thick layer HMA surfacing

Recycle upper 150 mm

2200 CR

Existing pavement

Recycle upper 100 mm

32 / 33
1.6.2 Structural rehabilitation

Rehabilitation to address problems within the deformation can lead to problems manifesting in
structure of a pavement is normally treated as a the overlying layers, especially where such layers
long-term solution. When addressing structural are constructed from bound material.
problems, it should be remembered that it is the
structure of the pavement that is distressed, sel- As a rule, structural rehabilitation should aim to
dom the materials within the structure. In addition, maximise the salvage value of the existing pave-
upgrading an existing pavement by strengthening ment. This implies that material that has densified
the structure (e.g. upgrading an existing gravel should not be disturbed. The continuous kneading
road to blacktop standards) may be regarded as a action of traffic takes many years to achieve this
form of rehabilitation. state and the benefits that such high densities
offer should be utilised where possible.
Densification (or compaction) of granular ma-
terials is, in fact, a form of improvement since Various options that are popular for structural
the higher density of a natural material leads to rehabilitation include:
better strength characteristics. However, the
consequences of densification and the resulting
> Total reconstruction. This is often the preferred > Construction of additional layers (either from
option when rehabilitation is combined with an untreated or treated granular material and / or
upgrading exercise that demands significant asphalt) on top of the existing surface. Thick
changes to the alignment of the road. Essen- asphalt overlays are often the easiest solution to
tially, reconstruction implies throw-away-and- a structural problem where the traffic volumes
start-again. Where traffic volumes are high, it are high. However, as described above, an
is often preferable to construct a new facility on increase in surface elevations often gives rise to
a separate alignment, thereby avoiding traffic unforeseen drainage and access problems.
accommodation problems.

Remove existing Reconstruct each layer

distressed layers (Conventional construction equipment)

Total reconstruction

34 / 35
> Deep recycling to the depth in the pavement pavement is marginal and requires strengthen-
at which the problem occurs, thereby creat- ing. Recycling aims for maximum recovery from
ing a new thick homogeneous layer that can the existing pavement. In addition to salvaging
be strengthened by the addition of stabilising the material in the upper layers, the pavement
agents. Additional layers may be added on top structure below the level of recycling remains
of the recycled layer where the pavement is to undisturbed.
be significantly upgraded. Stabilising agents
are usually added to the recycled material,
especially where the material in the existing

Recycle 250 mm deep Pave 40 mm thick HMA surfacing

Add 2.5% foamed bitumen
+ 1% cement

Existing
pavement

Deep recycling
> Combining two recycling methods, in-place thereby achieving additional structural capac-
with in-plant. This option allows for an increased ity. The thickness of the paved layer can be
depth of existing pavement to be treated and selected to suit the final surface level require-
requires that a portion of the upper pavement ments. For example, where existing surface lev-
be initially removed and placed in temporary els are to be maintained after rehabilitation and
stockpile. The underlying material is then recy- a 40 mm thick asphalt surfacing is required, the
cled / stabilised in-place. The material placed thickness of paving for the upper stabilised layer
in temporary stockpile is then treated in-plant is reduced by 40 mm to allow the final surface
and paved on top of the in-place recycled layer, levels to match those prior to rehabilitation.

Mill off 150 mm asphalt Treat stockpiled RAP in KMA with

Place RAP in stockpile 2% foamed bitumen + 1% cement

Recycle 200 mm Pave 40 mm

Existing Pave 150 mm
deep and stabilise HMA surfacing
pavement thick layer
with cement or
foamed bitumen

Two -part recycling

The purpose of considering several pavement re- approach that will allow recycling to be included
habilitation options is to determine the most cost- as one of the options. Economic evaluations of the
effective solution. This manual is aimed at provid- different options will then help identify the optimal
ing sufficient information together with a design solution, as discussed in the next chapter.

36 / 37
2 Pavement rehabilitation

2.1 General 41
2.2 Pavement rehabilitation: investigation and design procedure 42
2.3 STEP 1: Data acquisition/process available information 44
2.3.1 Information on the existing pavement (historical information) 45
2.3.2 Design traffic 46
2.4 STEP 2: Preliminary investigations 48
2.4.1 Determination of uniform sections 49
2.4.2 Visual inspection 52
2.4.3 Reassessment of uniform sections 54
2.5 STEP 3: Detailed investigations 55
2.5.1 Excavating Test Pits 55
2.5.2 Laboratory testing 56
2.5.3 Extracting core specimens 57
2.5.4 Dynamic cone penetrometer (DCP) probes 58
2.5.5 Analysis of deflection measurements 60
2.5.6 Rut depth measurements 60
2.5.7 Synthesis of all available data 61
2.6 STEP 4: Preliminary pavement rehabilitation design options 62
2.6.1 Pavement design approach 62
2.6.2 Catalogue design methods 63
2.6.3 Structural number method 63
2.6.4 Pavement number method 64
2.6.5 Mechanistic design methods 66
2.6.6 Deflection based methods 67
2.6.7 Summary of pavement design approaches 67
2.7 STEP 5: Laboratory mix design 68
2.8 STEP 6: Finalise pavement design options 70
2.9 STEP 7: Economic analyses 71

38/39
As was explained in the previous chapter, pave- more economic sense to rehabilitate the whole
ments deteriorate with time and usage. As the pavement rather than addressed localised areas of
end of the service life approaches, the rate of distress. Although there are no definitive guidelines
deterioration in the riding quality increases due and each road has its own unique features, there
to deformation, cracking, potholing and other appears to be general consensus that once 15%
such distress symptoms. The maintenance ef- of the surface area has been patched, it is cheaper
forts necessary to hold the surface together and to rehabilitate the whole road than to continue ap-
maintain an adequate level of service tend to plying patches on an ad hoc basis.
escalate until the point is reached where it makes

Heavy patching indicates the end of the service life

2.1 General

Pavement rehabilitation is the term used to de- The various steps in a pavement rehabilitation
scribe the work required to reinstate a distressed exercise include gathering of relevant informa-
road and restore the structural integrity of the tion (e.g. traffic data), conducting surveys and
pavement. Where a road is properly designed and tests to identify and determine the composition
constructed and where routine maintenance and and condition of the various layers in the existing
resurfacing interventions are undertaken timeously, pavement structure (e.g. deflection measure-
the need for rehabilitation can be delayed until ments), summarising and interpreting all available
the pavement reaches a terminal condition due to data to allow alternative design options that meet
structural deterioration. However, in practice, such the required design life (structural capacity) to be
maintenance and resurfacing activities are often formulated and, finally, to decide which option is
not carried out, resulting in the need for the pave- most attractive. Although these steps are common
ment to be rehabilitated sooner than was originally to all pavement rehabilitation exercises, the focus
envisaged. In addition, pavement rehabilitation of this chapter is on identifying and understanding
is often included with strengthening and/or geo- the materials in the upper pavement structure and
metric improvements required to accommodate their potential for being recycled.
increased traffic volumes.
A flowchart is included to illustrate the various
This chapter describes the various steps that are steps and methods used in the pavement rehabili-
included in a pavement rehabilitation exercise. tation exercise. Methods that are normally used for
These are explained and guidelines given in order investigating distressed pavements are outlined.
to provide a practical overview of an extremely Different methods for designing pavements are
involved (and at times complex) process that has described, particularly those most suited to pave-
recently become an area of specialisation in the ments that are rehabilitated by recycling.
field of Pavement Engineering. These explanations
and guidelines are certainly not all-inclusive and
reference should be made to the literature included
in the Bibliography (after Chapter 6) should more
detailed information be required.

40/41
2.2 Pavement rehabilitation:
investigation and design procedure

The need for rehabilitation is usually triggered by The flowchart on the facing page is applicable
an unacceptable level of distress that reflects as to all rehabilitation projects and can be adapted
a poor surface condition (e.g. bad riding qual- according to specific needs. The various activities
ity, pothole development, etc.), often highlighted are grouped under seven sequential steps:
and prioritised by using an appropriate Pavement
Management System (PMS). Once a road has Step 1. Acquisition of available information
been identified for rehabilitation, a full pavement Step 2. Preliminary investigations and identifica-
investigation and design procedure needs to be tion of uniform sections
followed to determine the most appropriate reha- Step 3. Detailed investigation of each uniform
bilitation solution. section and synthesis of all data
Step 4. Preliminary pavement design options
The primary objectives for investigating an existing based on estimates of stabilised material
pavement are to determine the composition of the properties
pavement structure, gain an understanding of the Step 5. Laboratory mix designs to determine
behaviour of the materials in the various layers and stabilised material properties
establish the cause of distress that has increased Step 6. Pavement design finalisation
the demand for maintenance measures. Step 7. Economic and other analyses to indicate
the optimal solution

The following sections describe each step in detail.

STEP 1 Define the Roads Authoritys
Data Acquisition
specific requirements

NO
Sufficient data?
e.g. traffic

Acquire additional information YES

Process data

STEP 2
Preliminary investigations /
identify uniform sections

STEP 3
Detailed investigations
Synthesis of all information

STEP 4 Formulate preliminary pavement

design options based on estimates
for stabilised material properties

STEP 5
Consider alternative options available for Laboratory mix designs
changing the material properties:
- different stabilising agent
- import fresh material (dilution)
- recycle deeper, etc.
Were the properties estimated for
NO stabilised material achieved?

YES
STEP 6
Finalise pavement design options

STEP 7
Conduct economic analyses

42 / 43
2.3 STEP 1: Data acquisition/process
available information

At the start of every pavement rehabilitation exer- These requirements provide the design engineer
cise, those responsible for formulating the design with the scope and limits of the project. The inves-
must have an unambiguous understanding of what tigation phase commences by sourcing all infor-
is required in terms of: mation on the existing pavement that is available.
This information falls under two major headings
> Design life. Is a short-term or long-term service
described below:
life required?
> records of historical information; and
> Functional properties of the rehabilitated road
(e.g. specific requirements for riding quality, skid > traffic data to determine the structural capacity
resistance and noise levels) requirements.
> Available budget. The level of funding available
for the rehabilitation works and for the routine
maintenance measures that will be required dur-
ing the service life.
2.3.1 Information on the existing pavement (historical information)

All available information should be sourced and In addition, as much information as possible
analysed in order to place the project in context should be obtained about locally available con-
and provide an early appreciation of what can be struction materials. The type, quality and quantity
expected when starting the field investigations. of materials that can be obtained from both com-
Where available, construction and maintenance mercial sources and previously opened borrowpits
records can provide valuable information on: and quarries should be investigated for possible
use in the rehabilitation works. Also, the location
> details of the pavement that was originally con-
and distance from site of any established asphalt
structed;
plant should be established.
> the thickness of as-built layers;
Furthermore, meteorological records from the
> details of materials used in the construction of closest weather station should be obtained and
the original layers as well as those used in any analysed to determine seasons that are best suited
subsequent rehabilitation or improvement meas- for the type of construction that is envisaged.
ures;
> results of quality control tests conducted during
construction; and
> geological data along the route.

The proximity of locally available materials will influence rehabilitation options

44/45
2.3.2 Design traffic

The volume and type of traffic that a road is

expected to carry during its service life dictates
the structural requirements. Pavement engineers
therefore need traffic statistics (in terms of vehicle
numbers, configuration and axle loads) to estimate
the anticipated structural capacity requirements for
rehabilitation.

Structural capacity requirements

The structural requirements of a pavement are
referred to as the structural capacity that defines
the loading a pavement can withstand before it
Traffic spectrum comprised of many different v
ehicle
deteriorates to the stage that a failure condition
types
is reached. Structural capacity is expressed in
terms of equivalent load repetitions with the load
carried on a single axle expressed in tons (or kN), ity of a pavement is usually defined in terms of
usually in accordance with the maximum legal equivalent axle load repetitions with the axle
axle load that varies between 8 tons and 13 tons, load stated in kN (e.g. 25 million equivalent 110 kN
depending on the country. Different axle loads are axle loads). Using an appropriate damage factor,
related to this maximum to obtain an equivalent the number of equivalent 110 kN loads can be
value. For example, where the maximum legal axle converted to ESALs (e.g. 25 million equivalent
load is 8 tons, a single axle load of 4 tons would 110 kN axle loads would translate to some 90 mil-
be between 0.1 and 0.3 equivalent 80 kN axle lion ESALs).
loads, depending on the function used to relate
a 4 ton load to an equivalent 8 ton load. Structural capacity is often referred to as the de-
(This relationship function is also known as the sign traffic or the bearing capacity of a pave-
damage factor). ment and, provided they are all quoted in terms of
millions of equivalent axle load repetitions, they are
The term equivalent standard axle load (ESAL) the same.
originated with the AASHO Road Tests carried out
in the USA during the late 1950s that marked the
advent of pavement engineering. The standard
axle load referred to in ESALs is 8 tons or 80 kN
and, as explained above, different axle loads are Note:
related to an equivalent number of 80 kN axle
loads with the structural capacity of a pavement Design traffic
expressed in terms of millions of ESALs. Accurate traffic data and growth information
are essential
Where the legal axle load is different from 8 tons
(e.g. 11 tons in Germany), the structural capac-
Pavements are therefore designed to provide for a currently using the road, the average number of
specific structural capacity. Although a design life axles per heavy vehicle and the average mass
is often quoted in years, pavements are actually carried on each axle. This information should be
designed to accommodate the number of load supplemented wherever possible with information
repetitions that are anticipated during that period. obtained from weighbridge stations (including the
Any unforeseen changes in the estimated traffic results of any tyre pressure surveys).
loading will therefore impact on the design life.
This is one of the most fundamental aspects of It must always be borne in mind that the informa-
pavement engineering and is so important that a tion used to calculate structural capacity is based
full explanation is included in Appendix 2, entitled on assumptions concerning traffic growth rates,
Determining Structural Capacity from Traffic damage factors and other data that can only be
Information. estimated. It is therefore important to carry out
sensitivity analyses to understand the conse-
Obtaining additional traffic information quence of varying these estimated parameters.
Where the available traffic data is insufficient,
particularly when designing the rehabilitation of
heavy pavements, additional information must
be obtained. Classified traffic counts should be
undertaken and weigh-in-motion data obtained to
estimate the percentage of heavy vehicles

Data capture

Weight sensor

Weigh-in-motion station to measure and record axle loads

46 / 47
2.4 STEP 2: Preliminary investigations

Before any field surveys or investigations com- Road pavements are seldom uniform over long
mence, it essential that the existing road surface distances. Both the underlying geology and the
is accurately pre-marked with an appropriate materials used in the construction of the individual
referencing system (usually the chainage or km layers will vary along the length of the road. All
distance (e.g. km 121 + 400) is adopted). It is nor- roads are therefore comprised of a series of differ-
mal to paint prominent marks every 20 m on the ent sections of relative uniformity and the length of
centre-line or outer edge of the carriageway and each section will be different. These sections are
to write the chainage value every 100 m. These known as uniform sections and may be as short
marks are then used as the primary reference for as a few hundred metres or as long as several kilo-
all survey and test locations. metres. Uniform sections are identified visually by
changes in distress patterns. Deflection measure-
ments are also useful for identifying differences in
the underlying pavement structure.
2.4.1 Determination of uniform sections

One of the main objectives of undertaking prelimi- impulse (falling weight) or a known wheel load that
nary investigations is to identify uniform sections. simulates a heavy vehicle. The magnitude of the
This is usually achieved by analysing available con- deflection that occurs under a given load, as well
struction records, analysing any deflection data, and as the shape of the deflection bowl produced by
from conducting a comprehensive visual inspection. the load, provides a useful means of assessing the
Similar distress symptoms and/or deflection meas- in-situ properties of the pavement.
urements indicate similar conditions in the underly-
ing pavement structure. This information is used Various methods for measuring pavement deflec-
to identify the boundaries between the different tion have been developed, primarily for use as
uniform sections, and the type of distress (indicating indicators of the structural condition and load car-
the mode of failure). The following sections describe rying capacity of the pavement. Those most widely
how this is achieved in practice. used are the Benkelman Beam and the Falling
Weight Deflectometer (FWD).
Deflection method
When a load is applied to the surface of a road,
the pavement deflects. Deflections may be meas-
ured by applying a load on the pavement, either an

Measuring pavement deflections using a FWD

48/49
The sketch below illustrates typical deflection bowls measured on two different pavements.

wheel / impulse load

Maximum deflection

Softer material Stiffer material

Typical deflection bowls

As primary input for their PMS, road authorities The cumulative-sum value is then plotted at each
normally carry out deflection surveys at 3 to 5 year respective location, normally together with the
intervals on all major roads in their network. Where maximum deflection value plotted on the same
available, such information is invaluable for the graph, as shown in the sketch opposite. A relatively
initial definition of uniform sections using simple constant slope for the cumulative-sum plot indi-
statistical techniques (cumulative-sum analysis) cates sections of similar pavement response, or a
to identify where changes occur. The cumulative- uniform section.
sum values of maximum deflections are calculated
using the formula: Note: The cumulative-sum method is not re-
stricted to maximum deflection. This method is
often used with other deflection indices, such as
the Surface Curvature Index (SCI).
Si = (i mean) + Si-1 (equation 2.1)

Other methods
Where no deflection data is available, uniform sec-
where Si = cumulative-sum value at location i; tions must be identified by other means. As-built
i = maximum deflection at location i; construction information (when available) is often
and used as an initial guideline, supplemented by a
mean = mean of maximum deflection for detailed visual assessment, as discussed below.
the entire section. However, when the required structural capacity of
Si-1 = cumulative-sum value at location the pavement exceeds 10 million ESALs, it is al-
before location i ways advisable to undertake a FWD survey at the
 Maximum Cumulative
Distance (m)
deflection (d) sum (S)
30,060 64.80 -41.14
30,080 76.70 -70.38
30,100 86.60 -89.71
30,120 94.00 -101.65 Uniform sections
30,140 79.10 -128.49 300 250
1 2 3
30,160 72.70 -161.73
30,180 71.30 -196.36
30,200 79.50 -222.80 200
30,220 82.40 -246.34 200
30,240 71.70 -280.58
100

Maximum deflection (10-2 mm)

30,260 76.80 -309.71
30,280 78.90 -336.75
Cumulative sum

30,300 110.40 -332.29 150

0
30,320 98.70 -339.53
30,340 86.70 -358.76
30,360 97.40 -367.30 -100
30,380 139.60 -333.64 100
30,400 134.70 -304.88
30,420 164.00 -246.81 -200
30,440 129.50 -223.25
30,460 142.50 -186.69 50
30,480 152.30 -140.33 -300
30,500 150.10 -96.16
30,520 163.50 -38.60
30,540 198.90 54.36 -400 0
30,560 119.60 68.02 30,000 30,100 30,200 30,300 30,400 30,500 30,600 30,700 30,800
30,580 208.60 170.69
Distance (m)
30,600 132.80 197.55
30,620 72.10 163.71 = Maximum deflection (d)
30,640 63.20 120.98 = Cumulative sum (S)
30,660 61.10 76.14
30,680 29.80 0.00
Mean (D) 105.94

Identification of uniform sections

outset. In addition to identifying uniform sections,

Note:
the information derived from such a survey is
invaluable for the statistical assessment of various Cumulative-sum methods can be used to
in-situ pavement properties (see Section 2.5.5). identify uniform sections based on deflection
measurements or other relevant information
(e.g. subgrade CBR value) collected along
the length of the road.

50 / 51
2.4.2 Visual inspection

Visual inspections are undertaken by walking the be recognized during the inspection are normally
length of the road and recording all relevant fea- classified into the categories shown below.
tures that can be observed or detected. Detailed
notes are taken of all distress that is evident at the
surface over the full width of pavement as well as Note:
other observations concerning drainage, geologi-
cal changes and geometric features, (e.g. steep Visual inspection data provide valuable clues
grades, sharp curves, cuttings and high embank- regarding the cause of distress
ments). The mode and type of distress that can

Mode of distress Type of distress Description

Ravelling, stone loss
Environmental damage Thermal cracking
Surface damage
Traffic damage Rutting (asphalt)
Stripping, bleeding, polishing
Rutting in wheel paths
Lateral shoving
Permanent deformation
Longitudinal in wheel paths
Structural damage Cracking
Crocodile
Advanced damage
Other (transverse, etc.)
Potholes, patches, etc.
Erosion, washouts, etc.
Drainage
Functional condition Edge break
Riding quality
Undulations, corrugations, etc.

The different modes and types of pavement dis- Whilst carrying out the visual inspection, digital
tress are recorded for each occurrence in terms of photographs of the road surface are normally
location, severity and frequency. Visual inspections taken at regular intervals ( 250 m in both direc-
provide valuable clues regarding the causes of tions) and to record specific features (e.g. localised
pavement distress since failure patterns tend to failure). In addition, video clips are powerful tools
be highlighted when all the data is summarised on for recording traffic-related problem areas that
one sheet. This feature is illustrated in the example should be addressed under the rehabilitation pro-
shown opposite. ject (e.g. dangerous traffic movements).
 To Hellngone To Good City
Visual inspection Cross access River
National Road 1 Section 12
Good City to Hellngone culvert culvert culvert culvert culvert culvert
km 30 - 36
km 30 31 32 33 34 35

Condition
Riding quality
measurement

rating
Instrument

Deflection
Sound
Rutting

Cracking
Warning
Disintegration (Surface)
assessment
Visual

Deformation

Smoothing (bleeding) Severe

Patching

Comments Poor shape Dangerous Erosion

(no camber) intersection downstream

52 / 53
The primary difference between surface and struc-
tural distress is shown graphically in the following
scetches.

Distress confined to the surfacing Distress due to structural inadequacy

2.4.3 Reassessment of uniform sections

The uniform sections initially determined from definition of the boundaries between individual
deflection analyses should then be reassessed us- uniform sections and facilitates the identification of
ing information from the visual inspection, together sections with a similar pavement structure.
with all other available information (e.g. construc-
tion records). This process allows a more accurate
2.5 STEP 3: Detailed investigations

For each uniform section, a detailed investiga- and to determine in-situ subgrade support. The
tion is required to evaluate the existing pavement tests and surveys normally employed in a detailed
structure (components and mode of distress) investigation are described below.

2.5.1 Excavating test pits

Test pits are without doubt the most important The following important information can be
source of information concerning an existing accurately determined from test pits:
pavement structure. In addition to gaining a visual
appreciation of the different layers and materials in > individual pavement layer thickness;
the pavement structure, test pits provide the op- > moisture content of the in situ material in
portunity to determine the in situ condition of the each layer;
various materials and to take bulk samples from > in situ density of the material in each layer; and
each layer for laboratory tests (for classifying the > condition of the material in the various layers
materials and for stabilisation mix designs). (e.g. the degree of cracking, cementation or
carbonation of any cement-stabilised layer).

0 mm

85 mm Asphalt

Cement stabilised
crushed stone material

420 mm
Test pit
Crushed stone layer
555 mm

Cement stabilised crushed stone material

755 mm

900 mm Crushed stone layer

Dark Red Brown moist sandy silt + intact

1.200 mm weathered granite aggregate

Example of testpit excavation and profile (Depth from surface)

54 / 55
A minimum of two test pits are normally excavated Test pits must be carefully excavated so that
for each uniform section, one where distress is evi- each individual layer of different material type can
dent, and the other where there is no distress. Test be separated and removed independently. Each
pits are generally located in the outer wheel path type of material encountered is carefully removed
of the traffic lane and are sometimes positioned (usually by hand) and heaped separately next to
to straddle the shoulder and traffic lane. Test pits the excavation for later sampling. As the digging
are usually 1.0 m in length (across the wheel path), progresses, density and other in situ tests can
0.75 m wide (along the wheel path) and at least be carried out on each successive layer as it is
1.0 m deep. Additional shallow slots (0.5 m wide) exposed.
are often excavated across the full width of a traf-
fic lane as a means of investigating the depth to Once the excavation is complete, the pavement
which deformation extends and to determine the profile is carefully measured and recorded, as
existance of any pavement widening, as well as shown on the previous page. Bulk samples from
the location of the boundary between the original each different layer are retained for laboratory test-
and widened pavements. ing before the test pit is backfilled.

Note:

Representative samples must be used for

laboratory tests

2.5.2 Laboratory testing

Bulk samples from test pits are tested in the Laboratory tests normally carried out on these
laboratory to determine the quality of the mate- samples include: sieve analysis, Atterberg limits
rial in each of the different layers, as well as the and California Bearing Ratio (CBR). The results
underlying subgrade. The testing programme must are primarily used for material classification, i.e. to
also include samples of aggregates that may be provide an indication of relevant parameters (such
required for blending with any in situ recycled as elastic modulus) for use in analysing the exist-
material. Representative samples of such blend ing pavement structure. The results are also used
materials are to be obtained from the same borrow to indicate the materials suitability for stabilisation
pits and quarries that will be used as the respec- and which stabilising agent(s) would be appropri-
tive sources for construction. ate.
2.5.3 Extracting core specimens

Extracting core specimens by rotary drilling

Note:
through layers of bound material is relatively
quick and less destructive than excavating test The recoverable length of a core specimen is
pits or inspection holes. Provided full recovery limited by the height of the core barrel used.
is achieved, core specimens can then be meas-
ured to accurately determine the thickness of the When determining the thickness of bound
layer(s) of bound material (e.g. asphalt and cement layers from recovered cores, ensure that full
stabilised material). Where required, core speci- recovery is achieved (i.e. that the specimen
mens recovered from asphalt layers may be tested did not break during extraction and/or that
for volumetric composition/engineering proper- drilling continued to the bottom of the bound
ties and the unconfined compressive strength layer).
determined by testing specimens recovered from
Larger diameter core barrels (150 mm
cemented layers.
diameter) are preferred, especially where the
bound material includes aggregate larger
than 19 mm.
Unbound materials cannot be recovered and
sampled by coring.

Extracting core specimens from layers of bound material

56/57
2.5.4 Dynamic cone penetrometer (DCP) probes

The DCP is a simple tool consisting of a steel rod

with a hardened steel conical tip that is driven
into the road pavement using a drop hammer of Handle
standard mass falling a constant distance. The Upper Stop
penetration rate, measured in mm / blow, provides Hammer (8 kg)
an indication of the in situ bearing strength of the Cone dimensions
material in the different pavement layers and a

3 mm
change in penetration rate indicates the boundary
between layers. DCP probes are normally driven
to a depth of 800 mm, or deeper into heavier Zero mark
pavement structures. The penetration rates can 575 mm
Approx. 1,935 mm

then be plotted and used to indicate the thickness Cone angle 60

of the various layers and the properties of the in
situ material in each layer. 20 mm

Anvil -
DCP penetration rates correlate well with the Cali- where rods screw together
fornia Bearing Ratio (CBR) in relatively fine materi-
als and only reasonably well with coarser granular Upper clip -
materials (at the in situ density and moisture reference point for scale
content). Correlations of penetration rate with the Steel rods, 16 mm
Unconfined Compressive Strength (UCS) of lightly
cemented materials have also been developed. Measuring rod with
In addition, the DCP penetration rate provides a adjustable scale
rough but useful guide for the elastic modulus of Lower clip
in situ pavement materials.

Since the coefficient of variation is often rela- (not according to scale)

tively high, numerous DCP probes are normally
DCP dimensions
required to achieve statistical reliability. Measure-
ments should therefore be analysed statistically to
obtain the relevant percentile value (normally the Difficulties are sometimes experienced when
20th percentile is used for minor roads and the 5th attempting to drive a DCP probe through coarse
percentile for major highways). or bound material. A refusal condition is usually
defined by a penetration of < 1 mm for each of
Individual DCP penetration measurements are two successive sets of 5 hammer drops (i.e. <
normally taken once for every five blows. These 2 mm measured for 10 drops of the hammer). If
measurements are then analysed using computer this condition is encountered when attempting
software to indicate in-situ CBR, UCS, Resilient to drive the probe through the upper portion of a
Modulus and layer thickness, as shown in the pavement, a 25 mm diameter hole should first be
example analysis opposite. drilled through the layers of coarse and / or bound
 DCP Curve Layer strength diagram Elastic modulus

0 0 0

100 100 100

Pavement depth d (mm)

Pavement depth d (mm)

Pavement depth d (mm)

200 200 200

300 300 300

400 400 400

500 500 500

600 600 600

700 700 700

800 800 800

0 50 100 150 200 1.000 100 10 1 10 100 1.000
Number of blows CBR (%) Elastic moduli (MPa)

material, thereby starting to probe at a lower depth be removed from the bottom of the core hole as
in the pavement where the cone can penetrate. soon as the core has been extracted. It must be
appreciated that the water used to cool the core
Alternatively, if cores are extracted from bound barrel whilst drilling will influence the penetration
layers in the upper pavement, DCP probes can be measured for the first 50 mm to 100 mm of the
started from the bottom of the core holes. Where DCP probe.
this procedure is followed, surplus water should

58 / 59
2.5.5 Analysis of deflection measurements

As was introduced in Section 2.4.1, deflection Actual layer thicknesses from field measurements
measurements can be analysed to provide valu- and in situ layer moduli values indicated by DCP
able in situ information about the materials in the probes should be used as guidelines for each layer
pavement structure. In addition to assisting with (especially the subgrade) when back analysing the
the delineation of uniform sections, the relevant deflection bowls. The results of these analyses
deflection measurements within each uniform provide estimates of the in situ stiffness values
section can be analysed statistically, at the ap- for the various pavement layers. As discussed in
propriate level of confidence, and the deflection Section 2.6 below, this information is required for
bowl back-analysed (e.g. only the 95th percentile pavement modelling.
deflection bowl for a specific uniform section is
analysed.)

2.5.6 Rut depth measurements

Rut depths are generally measured manually using

a straightedge positioned transversely across the
wheel paths in each traffic lane. The maximum rut
depth is recorded. Rut depths can also be meas-
ured using sophisticated mobile road surveillance
equipment employing laser measuring techniques
(e.g. ARAN automatic road analyser).

The width of rut in the pavement surface (also re-

ferred to as the radius of rut) indicates the source
of deformation within the pavement structure.
Narrow ruts are generally caused by instability in
layers of asphalt whilst wider ruts indicate perma-
nent deformation in the underlying layers. The cor- Measuring the rut depth in the wheel path
relation of rut depth and deflection (measured at
the exactly same point) also assists in determining
whether the upper or lower layers in the pavement Note:
structure have deformed.
The depth of rut measured is influenced by
the length of straight edge used.
2.5.7 Synthesis of all available data

The detailed investigation phase culminates with When read in conjunction with the visual assess-
the compilation of a summary sheet that includes ment summary (Section 2.4.2), the relevant mode
all data specific to each uniform section, as shown of distress (failure) and problem areas within the
in the example below. existing pavement structure are easily identified.
This allows the design engineer to focus on alter-
The information shown in the summary sheet is native rehabilitation measures to address the iden-
typical for a comprehensive investigation. Since tified weakness and problem areas, as described
this summary sheet (or strip map) contains the in the next section.
primary input required for pavement design, all
details concerning the different pavement layers
and material characteristics must be included.
A separate summary sheet (or strip map) is com-
piled for each uniform section.

121+400 121+500 121+600 121+700

Uniform section

mm 0.8
FWD 0.6 0.60 0.64
Maximum 0.4 0.57 0.49
deflection 0.2
0
DCP 200
Indicated 400 450 410 440 510 490 460
stiffness 600 260 220 235 270 240 255
(MPa) mm 800 120 100 110 125 120 125
20
Cores 40
HMA 60
thickness 80
mm 100 84 90 93 92
100 CBR PI 90 Asphalt
200
GCS > 100 NP 190 GCS base 280
300
Test pits 400
Gravel 65 SP 310 Gravel
and 500
test results 600 Sand 25 NP s / base
700 690
800 S / grde 15 10 150 Sand s / g 840
mm 900

60 / 61
2.6 STEP 4: Preliminary pavement rehabilitation
design options

Once the investigations are complete and the tify the three most appropriate options, thereby
summary sheets described above have been com- reducing the amount of analytical work required
piled, each uniform section can be considered in for designing the different pavements. These three
isolation and relevant rehabilitation design options alternative solutions must have similar structural
formulated. As with all diagnostic procedures, the capacities and, where a solution calls for additional
key to finding the best solution is to identify all layers and/or a thick asphalt overlay, the implica-
possibilities at the outset. However, a sense of rea- tions of raising road levels must be included in the
sonableness must prevail in identifying alternatives analyses.
since some will obviously be inappropriate (usually
due to excessive cost and/or constructability Pavement design for rehabilitation is different from
implications) and can be disregarded. designing new pavements, as discussed in the
following sections.
When all alternatives have been identified, a sub-
jective selection process is then followed to iden-

2.6.1 Pavement design approach

Over the past 60 years, many pavement design > Analytical methods. These methods all include
methods have been developed, ranging from an analytical process that is followed by inter-
relatively simple empirical methods to the more pretation (empirical element) to translate the
complex modelling approaches that require so- results of the analyses to structural capacity
phisticated computer software. (known as transfer functions):
mechanistic analyses. These methods are
The various pavement design methods can be based on stress and strain analysis using
summarised under two primary headings: multi-layer linear-elastic, elasto-plastic or finite
element models; and
> Empirical methods. These include: methods using deflection measurements
the CBR cover design method, based on the (deflection bowl analyses).
strength of the underlying subgrade;
catalogue design approach, based on typical As a general rule, more heavily trafficked pave-
pavement structures for specific applications; ments (> 10 million EASLs) should always be
the DCP design method that uses data from designed using an analytical approach. An empiri-
DCP surveys to indicate shortcomings in exist- cal method may suffice for lighter pavements but,
ing pavements; where there is any doubt that a design may not
the Structural Number method that assigns be appropriate for the anticipated traffic loads, it
coefficients to various material types; and should be checked using an analytical method.
the Pavement Number method that uses
intelligent structural numbers.
2.6.2 Catalogue design methods

Catalogue design methods are prescriptive in the (as it cannot include all options) and not easily
types and quality of materials required for a suit- transferable (as it is often developed for local ma-
able pavement structure. The catalogue provides terials and climatic conditions). The support condi-
a list of pavement types appropriate for different tions used in catalogue designs also need to be
support conditions and structural capacities. Al- analysed on the basis for which the design options
though this design approach is usually developed were developed. Catalogue designs are therefore
using analytical procedures, it is both restrictive of little use for pavement rehabilitation.

2.6.3 Structural number method

Based on experience, structural coefficients have The Structural Number (SN) approach is simple
been developed for certain pavement materials for as it uses known materials with a track record of
use in structural design. The AASHTO 1993 pave- performance in given climatic conditions. Caution
ment design method uses a structural number that should be exercised when using this method
is determined by summating the product of these where climatic conditions are severe or where
structural coefficients and the respective layer local materials are significantly different. Further-
thickness. If the total exceeds a certain minimum more, since there is no inherent control system for
number for the specific subgrade condition and maintaining pavement balance in terms of relative
structural capacity requirement, the pavement stiffness of overlying layers, this design method is
structure is considered to be adequate. not recommended for pavements with a struc-
tural capacity requirement in excess of 10 million
ESALs.

Pavement Structural Layer thickness

coefficient SN layer
Layers Material (per in) mm inches

Asphalt 0.4 100 4 1.6

GCS
0.14 200 8 1.12
CBR > 80

Natural
gravel 0.12 300 12 1.44
CBR > 45

Subgrade
SN = 4.16
> 15

Structural number calculation example

62 / 63
2.6.4 Pavement number method

This method is similar to the structural number As with structural numbers, the pavement number
method, but uses the effective long term stiff- (PN) method is simple to use. The main difference
ness (ELTS) values for different pavement materi- between the two methods is the procedure fol-
als in place of structural coefficients. lowed in determining appropriate ELTS values as
This design method (described in TG2 (2009)) is opposed to selecting a structural coefficient. This
based on the analyses of long term pavement procedure includes a comprehensive classifica-
performance (LTPP) exercises combined with tion system for the materials in the various layers
laboratory research and heavy vehicle simulator and takes into account the climate, the location
(HVS) trials. of the layer within the pavement structure and the
amount of cover over the subgrade. It utilises the
The pavement number method summates the modular ratio rule to ensure pavement balance.
product of the ELTS value and the respective layer
thickness. The number obtained is then used in a Different material classes and their respective
frontier curve (derived from the LTPP exercise) modular ratio / maximum allowable stiffness value
to indicate the structural capacity of the pavement are summarised in the table below (see also TG2
structure. (2009)).

Principal strength Maximum stiffness

Material type Class Modular ratio
characteristic (MPa)*

HMA AC Marshall 5 2500

BSM ITSDRY > 225 kPa
Bitumen stabilised 3 600
Class 1 ITSWET > 100 kPa
CTB
Cement stabilised 1.5 < UCS < 3 MPa 4 550
Class C3
G1 2.0 700
Crushed stone G2 CBR > 100 1.9 500
G3 1.8 400
G4 CBR > 80 1.8 375
G5 CBR > 45 1.8 320
Natural Gravels
G6 CBR > 25 1.8 180
G7 CBR > 15 1.7 140
G8 CBR > 10 1.6 100
Soils G9 CBR > 7 1.4 90
G10 CBR > 3 1.2 70
* These maximum stiffness values are relevant only to the empirical Pavement Number model.
They are not intended as input for mechanistic models (see Section 4.3.12)
 Pavement structure Layer Maximum
Modular ELTS
thickness stiffness PN layer
Layers Material Class ratio
mm MPa MPa
Asphalt AC 100 5 2,500 2,340 23.4

GCS
G1 200 2 700 468 9.36
CBR > 80

Natural
gravel G5 300 1.8 320 234 7.02
CBR > 45

Subgrade
Subgrade
G7 cover 130 SN = 39.78
> 15
600 mm

Pavement number calculation example

Pavement Numbers may be used with confidence set is expanded and updated, the frontier curve
to design pavements with a structural capacity up will be extended to include pavements with struc-
to 30 million ESALs, the limit of the frontier curve. tural capacities in excess of 30 million ESALs.
This is an artificial limit since it is dictated by the
maximum traffic volume carried on pavements that
were included in the LTPP data set. As this data

64 / 65
2.6.5 Mechanistic design methods

The mechanistic design process uses struc- ratio). The loading condition is defined (axle mass,
tural mechanics and material models to analyse tyre / wheel configuration and tyre pressure) and
stresses, strains and deflections that develop the response of each layer is then calculated.
within a pavement structure when a load is ap-
plied. The relevant stress, strain or deflection value The input parameters for the materials in each
is then related to structural capacity (number of layer are obtained during the detailed investigation
load repetitions to failure) by means of a transfer stage. Material classification from laboratory tests
function, an empirical relationship derived from provides an indication of stiffness and relevant
research and / or pavement performance data. Poisson Ratio values for the specific materials.
In addition, analyses of DCP probes and deflection
Multi-layer linear elastic theory is generally used bowls are invaluable for the in situ properties of
for analysing the pavement model, mainly because pavement materials.
it is relatively simple to use and easy to set up.
(Other analytical tools that can be used for model- The mechanistic design method has distinct
ling include finite element analyses incorporating advantages for rehabilitation design as it enables
non-linear-elastic, elasto-plastic and stress- non-standard materials in existing pavement layers
dependent constitutive models.) The pavement is to be modelled effectively. In addition, the method
modelled by defining the layer thickness together accommodates all rehabilitation options, especially
with the relevant properties of the material in each thick stabilised layers that are a feature of recycled
layer (in terms of elastic modulus and Poissons pavements.

Applied load: 80 kN per axle carried on dual tyres, tyre pressure 750 KPa
Pavement Layer Resilient modulus range
thickness Poissons (moisture dependent) Failure criterion
Layers Material ratio (evaluation parameter)
mm MPa
Asphalt 100 0.4 2,500 5,000 Tensile strain

GCS
200 0.35 300 800 Deviator stress ratio
CBR > 80

Natural
gravel 300 0.35 150 400 Deviator stress ratio
CBR > 45

Subgrade
Infinite 0.35 70 140 Vertical strain
> 15

Input parameters for a multi-layer linear elastic model

2.6.6 Deflection based methods

Deflection based rehabilitation design methods levels in the new material from exceeding the elas-
are primarily used to determine the thickness of tic limit, thereby achieving the required structural
asphalt overlays. Deflection measurements are capacity.
analysed to determine the effective stiffness of the
existing pavement structure and this is used as the Alternatively, deflections can be used to gain an
primary input in a mechanistic exercise to indicate indication of the strain in the upper part of the
the thickness of asphalt required to accommodate pavement which will indicate the thickness of
applied loading conditions, given the effects of asphalt overlay required.
reflective cracking from existing bound layers.
An iterative approach is adopted to determine the
thickness of asphalt overlay that prevents stress

2.6.7 Summary of pavement design approaches

Investigation and design phases should be integrat- ing pavement, identifying weaknesses, and for
ed with the primary objective being to understand designing pavements that satisfy the rehabilitation
the behaviour of the existing pavement. The second requirements. Empirical design methods may be
objective is to determine the most cost-effective used for lower levels of design traffic or as a first
pavement design that satisfies expectations attempt at rehabilitation design.
regarding design life and functional properties, that
minimises maintenance interventions and one that
produces a pavement that can be rehabilitated at
the end of the design life at minimal cost.

Mechanistic design methods are favoured as a

means of checking the adequacy of the exist-

66/67
2.7 STEP 5: Laboratory mix design

Laboratory mix designs play the role of verifying Step 1: Initial selection of stabilising agents, taking
the suitability of selected materials for treatment into account:
with additives. Such additives can include stabilis-
ing agents, chemicals, aggregates and natural > suitability of a specific stabilising agent in re
materials. Mix designs are a fundamental part of lation to the type and quality of material to be
the pavement investigation and design proce- treated. The initial selection of the most appro-
dure and serve the purpose of establishing the priate stabilising agent is based on the results
most effective method of treating the materials to of laboratory tests carried out on the untreated
improve their engineering properties. (Chapter 4, material;
Stabilising Agents includes detailed information
> required engineering properties of the stabilised
on stabilising agents.)
material;
Representative samples taken from the material > availability, in terms of being able to procure
to be treated are subjected to mix design testing. sufficient daily volume requirements, as well as
These samples should be prepared to simulate as consistency of the quality of stabilising agent
closely as possible the material that will be pro- that can be supplied; and
duced on site during the actual treatment process.
Where bound material from an existing pavement > the relative cost of different stabilising agents.
(e.g. asphalt) is to be recycled in situ, samples
should be obtained using a small milling machine Based on the above, a decision is made to
to simulate the grading of the material that will be proceed with the full mix design using the most
produced by a recycler. appropriate stabilising agent. Alternatively, several
mix design options can be investigated simultane-
Mix design procedures that can be used for stabi- ously to determine which is the most appropriate.
lising with cement, bitumen emulsion and foamed
bitumen are described in Chapter 4 and included
in Appendix 1. These procedures essentially incor-
porate five basic steps:
Step 2: An optimisation exercise is carried out by In order to determine the optimum stabiliser con-
preparing several identical portions of the sample tent, the results of these tests are plotted against
and mixing each with different amount of stabi- the stabiliser content for each of the mixes. The
lising agent. Simultaneously, sufficient water is stabiliser content that best meets the desired
added to bring the mixture to its optimum moisture properties is regarded as the optimum stabiliser
content (for compaction). Typically at least four content (also termed the optimum application rate
mixes are prepared, each one mixed with a differ- of stabilising agent).
ent stabiliser content.

Step 3: Specimens are manufactured using stand-

ard compaction effort.
Note:
Step 4: The specimens are cured, ideally to simu-
late field conditions. Mix design results must be integrated with
pavement designs
Step 5: After curing, the specimens are subjected
to various tests to assess their engineering proper-
ties, as well as their susceptibility to moisture.

68/69
2.8 STEP 6: Finalise pavement design options

The initial pavement design options described in material properties, blending with an imported
above in Section 2.6 were, of necessity, based good-quality aggregate (e.g. crushed stone) may
on assumed values for the engineering properties be considered. However, such a change invari-
of the material in the new stabilised layer(s). The ably necessitates repeating the mix design.
actual properties are determined by carrying out
the mix designs described above in Section 2.7. If > Use a different stabilising agent or blend of al-
the actual properties are significantly different from ternative stabilising agents. This option requires
those assumed, the initial pavement design and a new mix design.
material utilisation needs to be reviewed. Where
the assumed values were not achieved in the mix > Import and stabilise an additional layer on top of
designs, the following options may be considered: the existing pavement structure. This will create
a deeper pavement, thereby reducing the levels
> Increase the layer thickness. On recycling of stress and strain in the existing portion of the
projects, this increase will result in a deeper cut pavement.
that may incorporate different material (usually
of poorer quality) from the underlying pavement. The pavement design is then finalised by input-
Where the change in material is significant, the ting the mix design values for stabilised material
mix design must be repeated to determine the into the pavement design model and refining the
correct properties. layer thickness of stabilised material to meet the
structural capacity requirements.
Alternatively, the recycled material can be blend-
ed with new imported aggregate to effectively
increase the layer thickness. Where it is obvious
that recycling deeper will not address the shortfall

Layer thickness insufficient for Option 1: Increase layer Option 2: Increase layer thickness by
structural capacity requirements thickness by recycling deeper importing fresh aggregate and main-
into the existing pavement taining the same recycling depth

Sufficient depth of cover Insufficient depth of cover

2.9 STEP 7: Economic analyses

Economic analyses are considered an effective discounted to account for the time value of money.
tool for selecting the most appropriate rehabilita- In such an exercise, it is of paramount importance
tion option. Alternative pavement rehabilitation to adopt the correct discount rate so that the pre-
designs cannot be compared on the basis of sent worth of future maintenance and rehabilitation
construction costs alone. In addition to the cost costs are realistic.
of maintenance that will be required during the ser-
vice life of the road (dependent on the pavement It is often difficult to estimate comparative con-
type, structure, materials, etc), the salvage value struction and future maintenance costs. Local
(cost of rehabilitation at the end of the service life) knowledge of materials and the environment, as
also needs to be included in the economic analy- well as data on pavement performance functions
sis. Therefore, all costs incurred during the service (normally obtained from a Pavement Management
life of the pavement (the whole-of-life costs) System) will assist in determining realistic future
should be integrated into relevant calculations. maintenance measures and their timing.

The economic analysis method normally used The process of evaluation is explained by means
to compare alternative pavement options is the of an example in section 5.2.
Present Worth of Cost (PWoC) method. This Appendix 4 contains a comprehensive background
method is based on estimates of all costs that will to economic evaluations and includes some of
be incurred during the life of the pavement (the the techniques that are applicable to the eco-
whole-of-life costs that include initial construction nomic analyses of different pavement structures
costs, plus the cost of all routine maintenance, (e.g. PWoC, Benefit/Cost Ratios, Internal Rate of
plus rehabilitation at the end of the service life), Return and Net Present Value).

70/71
3 Cold recycling

3.1 General 75
3.2 The cold recycling process 77
3.2.1 In-plant recycling 78
3.2.2 In-place recycling 79
3.3 In-place recycling machines 84
3.4 Cold recycling applications 90
3.4.1 100% RAP recycling 93
3.4.2 Blend of RAP/Granular material 94
3.5 Benefits of cold recycling 97
3.6 Applicability of the cold recycling process 98

72/73
3.1 General

Cold recycling is the term used for recovering and This chapter describes the various applications
re-using material from an existing pavement, with- of cold recycling and introduces the equipment
out the addition of heat. Unlike hot recycling that is required to carry out the work, specifically the
confined to heating and reusing asphalt material, range of Wirtgen machines. Also covered are the
the field of cold recycling enjoys a wide range of benefits that accrue from adopting cold recycling
applications, from relatively thin layers compris- and, in addition, the major factors that influence
ing mainly asphalt material (also known as cold the suitability of using such a process on a specific
in-place recycling) to thick layers that include two project.
or more different pavement materials (also termed
full depth reclamation (FDR)).

74/75
In situ recyclers e.g. WR 2500 S can work to With these capabilities, recyclers have introduced
depths in excess of 300 mm, break down and uti- pavement engineers to a whole new range of
lise 100% of the in situ material, maintain a preset possibilities, the most important being the ability
horizon within the existing pavement and produce to construct thick monolithic layers of stabilised
a quality of mixed material that is comparable with material. From a structural perspective, a single
off-site mixing. Conflict with public traffic is limited 300 mm thick layer of stabilised material has a far
to supply vehicles accessing the work zone. Weak higher load-carrying capability than two separate
areas in the underlying pavement structure are not 150 mm thick layers constructed one on top of the
disturbed and exposure to inclement weather is other. This concept is well known in the building
significantly reduced by processing the material in industry where laminated timber beams are used
a single pass and compacting the treated product as structural members.
immediately behind the recycler. In addition,
production rates as high as 10,000 m per shift are
possible and the risk of failures occurring due to
poor quality work is drastically reduced.

A thick layer of bound material is similar to a laminated timber beam

3.2 The cold recycling process

Cold recycling can be achieved either in-plant by > the material in the existing pavement is highly
hauling material recovered from an existing road variable and requires a process of selection;
to a central depot where it is fed through a mix- and/or
ing unit, or in-place using a recycling machine. > the material in the existing pavement is so hard
Plant processing is generally the more expensive that it cannot be properly pulverised in situ.
option in terms of cost per cubic metre of material Such materials are removed from the road and
processed, primarily due to haulage costs that are pre-treated before being reused as a pavement
absent from in-place recycling. However, in-plant material (e.g. crushing lumps of aged asphalt or
processing becomes attractive when: concrete).
> additional pavement layers are required. In-
Where it can be adopted, in-place recycling will al-
plant processing is usually preferred where
ways be the preferred recycling method due solely
previously-stockpiled material recovered from
to the economic advantages that are offered. In
existing pavements can be recycled and used
light of pavement deterioration worldwide, reha-
to construct a new pavement layer. Material
bilitation of existing pavements far exceeds the
treated with a bitumen stabilising agent can
demand for new roads and in-place recycling has
also be placed in stockpile for later use. This is
been universally accepted as the preferred method
becoming increasingly popular for dealing with
for addressing the enormous backlog in pavement
stockpiles of reclaimed asphalt pavement (RAP)
rehabilitation. In-place recycling therefore justifies
material;
> different materials are to be blended in accurate the focus of this manual with in-plant treatment
receiving less attention, but always remaining an
proportions; and/or
option. Each process is discussed separately in
the following sections.

Note:

Cold recycling existing pavements is the

universally preferred method of structural
rehabilitation

76/77
3.2.1 In-plant recycling

In-plant recycling allows materials from an existing proportions when mixing in-plant. Input materi-
pavement to be selected and pre-treated, thereby als can be selected, pre-treated (e.g. by crush-
increasing the level of confidence that can be ing and screening), stockpiled and tested prior
achieved in the final product. The main benefits to to mixing. The proportioning of various input
accrue from in-plant compared to in-place treat- materials can then by changed as and when
ment are: required to obtain the required mix.
> Control of input materials. Whereas in-place > Quality of mixing. An adjustments can be made
recycling allows no control over the type of to the mixer to vary the time the material is
material that is recovered from an existing pave- retained within the mixing chamber, thereby
ment, a required end-product can be obtained improving the quality of the mix.
by blending different aggregates in accurate

Typical in-plant operation with stockpiles of input material

> Stockpiling capabilities. Particularly with bitu- Placing the mixed material on the road as a new
men stabilised materials, the mixed product can pavement layer can then be undertaken by paver,
be placed in stockpile and used when required, grader or, where required, by hand using labour-
thereby removing the inter-dependency of the intensive methods.
mixing and placing processes. Cognisance
must, however, be taken of the strict time
restrictions applicable to stockpiling mixes that
include cement.

3.2.2 In-place recycling

Recycling machines have evolved over the years point-attack tools. Different tools are available to
from modified milling machines and basic soil suit different cutting conditions (e.g. hardness,
stabilisers to the specialised recyclers of today. abrasiveness, etc.). The drum normally rotates
Since they are specifically designed to recycle upwards and, as the machine advances, the in situ
thick pavement layers in a single pass, modern pavement material is pulverised by the tools and
recyclers tend to be large powerful machines, lifted into the mixing chamber that encloses the
mounted either on tracks or high flotation pneu- drum (the drum housing). Recyclers are equipped
matic tyres. During the past 20 years, Wirtgen has with at least one pumping system for adding fluid
invested heavily in developing this technology and (e.g. water) to the recovered material. The rate
their current range of high-performance machines of delivery of the fluid is metered accurately by
pays testament to the advances that have been means of a micro-processor that regulates the flow
achieved. A brief overview of the different types of in accordance with the volume of material in the
recyclers manufactured by Wirtgen and their typi- mixing chamber. The fluid is injected into the mix-
cal applications is included in Section 3.3. ing chamber through a series of nozzles spaced
equidistant on a spraybar that spans the full width
The heart of all recycling machines is the cut- of the chamber as illustrated in the diagram on the
ting drum that is fitted with a large number of next page.

78/79
 Display /
Keyboard

Open and closed-loop

control for metering

Processor
Printer

Control Control Control Measured flow rates

module 1 module 2 module 3 Pump control
Injection rail control

Measured flow rates

Pump control

Working
direction

Measured speed of advance

Measured flow rates

Pump control

Micro-processor control for the injection systems on the Wirtgen WR 2500 S

Recycling is undertaken by coupling bulk supply The combination of tankers coupled to the recycler
tankers to the recycler. The recycler pushes or is configured in accordance with the particular
pulls the tankers supplying the additives required recycling application and the type of stabilising
in the mix (e.g. bitumen emulsion). agent that is applied.
Sequency of machine in use

As the recycler advances, the mixed material falls

The simplest combination consists of a recycler back into the void created by the cutting drum and
coupled to a single tanker containing water. As the is struck off by a sturdy door fitted to the rear of
machine advances, the in situ pavement material the drum housing. As shown in the above sketch,
is recovered and mixed with water drawn from the a roller follows behind the recycler to compact the
tanker. The micro-processor ensures that the re- material before a grader is used to cut final levels.
quired amount of water is injected into the mixing Not shown in the sketch is the final compaction
chamber through the spraybar that is mounted on and finishing process that uses both vibratory and
the leading face of the drum housing (illustrated in pneumatic-tyred rollers working together with a
the sketch on the next page). water tanker.
The rotating drum mixes the water with the
recovered material to achieve a uniform consist- Powdered stabilising agents (e.g. cement or hy-
ency. The rate of water addition is controlled to drated lime) are normally spread on the surface of
achieve the moisture content that will allow a high the existing road ahead of the recycling operation.
level of density to be achieved when the material is As the recycler advances, the powder is lifted and
compacted. mixed together with the recovered material and
water, all in a single operation.

Note:

The micro-processor is of paramount impor-

tance since it controls the rate of application
of water and bitumen

80 / 81
Alternatively, the powder can be mixed with water Where a bitumen stabilising agent is added (either
to form a slurry suspension that is then injected bitumen emulsion or foamed bitumen), a second
into the mixing chamber. Where this method of application system is mounted on the recycler
application is adopted, a special mixing unit is complete with separate spraybar that is attached
coupled to the recycled. This slurry mixing unit on top of the drum housing, as illustrated in the
manufactures the slurry by combining the precise sketch below.
amounts of both cement and water required to
treat the volume of material being recycled. The
slurry is then pumped across to the recycler by
means of a flexible hose and injected through the
spraybar.

Dual application of water and foamed bitumen

This application requires a bulk tanker contain- mixing unit is always placed immediately ahead of
ing the bitumen to be coupled to the recycler. In the recycler and the bitumen tanker becomes the
addition, active filler (cement or hydrated lime) is lead vehicle, as illustrated in the sketch below.
normally added with a bitumen stabilising agent
(as explained in Chapter 4). Where the active filler The in-place recycling process described and illus-
is spread as a powder on the road surface ahead trated above focused on tyre-mounted recyclers.
of the recycling process, the bitumen tanker is Although the process is similar for track-mounted
coupled directly to the recycler and a water tanker machines, there are some fundamental differences
pushed in front as the lead vehicle. However, and these are described in the following section.
where the active filler is added as slurry, the slurry

Recycler coupled to a slurry mixer and bitumen tanker

82 / 83
3.3 In-place recycling machines

This section includes a general overview of the As shown in the illustration on the facing page,
three different types of in-place recycling machines the point attack tools are positioned in a chevron
manufactured by Wirtgen and their respective ca- pattern and mounted on stanchions to promote
pabilities. Individual machines are not described, mixing.
nor are any specifications included. These can all
be found in the companion publication Wirtgen Such a tool configuration promotes mixing in the
Cold Recycling Application that is concerned vertical plane but not the horizontal. This means
with the application of recyclers and the various that the recovered material is not thrown sideways
construction processes. (laterally) or lengthways (longitudinally) from its
original location within the pavement (the maxi-
The different types of in-place recycling machines mum movement measured from tests is
are: 200 mm). This means materials in the recycled
horizon will be returned to the pavement, after
> Tyre-mounted recyclers. recycling, in approximately the same location,
These machines are designed primarily for reflecting any original differences.
recycling. Wirtgen has several models of differ-
ent mass and capability in their product range; a
mid-size machine is shown in the picture.

The Wirtgen WR 2400 tyre-mounted recycler

The cutting drum on a tyre-mounted recycler The rear wheels of tyre-mounted recyclers run on the
treated material

Layers of dense unbound material are easily Tyre mounted recyclers are also used extensively
broken apart, returning to their original uncom- to pulverise thick layers of bound material in
pacted state. The degree to which layers of bound existing pavements (larger machines are used for
material (e.g. asphalt) are pulverised is influenced thicker layers). Increasing use is also being made
mainly by the advance speed of the recycler, but of these machines to pre-treat thick layers (up to
also by the rotation speed of the drum. The faster 500 mm thick) in the lower portion of the pave-
the advance speed and the slower the speed of ment. Such pre-treatment includes breaking down
rotation, the coarser the product. Large lumps of clods of material or soft rock and the addition of
material that are not pulverised in the process tend water to achieve a consistent moisture content
to be thrown to the bottom of the layer. that facilitates compaction.

The rear wheels of tyred mounted recyclers are The cutting drum of tyred mounted recyclers must
located inside the extremities of the cutting drum first penetrate through to the underside of the
and therefore run on the outer edges of the recy- bound material (e.g. asphalt) before the machine
cled material. As shown in the picture, the material can advance and pulverise the asphalt, as shown
in the wheel paths is compacted whilst the mate- in the following sketches.
rial between the wheel paths remains in a loose
(fluffed) state.

84/85
Tyre-mounted recyclers must attack bound material (asphalt) from underneath the layer

> Track mounted recyclers. achieved by using such a screed, thereby eliminat-
The cutting drum of these recyclers is the same ing the need for a grader to cut final levels.
as that used for milling asphalt with the point
As with the tyre-mounted recyclers, micro-proces-
attack tools mounted in a helical pattern to wind-
sor controlled pumping systems are incorporated
row the material to the centre of the drum.
with spraybars attached to the outside of the drum
housing to inject fluid additives into the material in
Instead of being lifted onto a belt and removed
the milling chamber.
(as happens when milling), the recovered mate-
rial exits through a door in the rear of the milling
Unlike tyre-mounted recyclers, the windrowing
chamber and passes between the rear tracks to be
action of the tool pattern promotes blending of the
spread across the width of cut by a variable screed
material recovered by each half-width of the drum.
fitted with individual left and right feed augers. The
required surface levels and shape can often be
The Wirtgen 2200 CR track-mounted recycler The standard milling drum on the Wirtgen 2200 CR

Milling machines are designed and built to provide These machines are ideal for recycling 100%
stability when cutting into layers of hard asphalt. asphalt material. A flagship of the Wirtgen Cold
Both the drum housing and the milling drum are Recycler product range is the 2200 CR equipped
attached to the frame of the machine and the with a 3.8 m wide drum to allow the full width of a
depth of cut is varied by lifting or lowering the traffic lane to be recycled in a single pass. In addi-
whole machine. This means that: tion, the rotation of the drum is changed from up
cutting (anti clockwise) to down cutting (clockwise)
> the drum does not have to penetrate to the to promote fragmentation when recycling thin
underside of thick bound layers to be able to asphalt layers only.
recycle the material. It is therefore possible to
recycle only the upper portion of thick bound
layers using these machines; and
> the volume of the milling chamber is constant,
regardless of the depth of cut. The amount of
material that can be mixed is therefore limited
and this restricts the depth of cut that can be
recycled to a maximum of 250 mm (or less when
applying a cohesive stabilising agent).

86/87
Paving screed fitted to the rear of the Wirtgen 2200 CR

> Machines fitted with an on-board twin shaft > the width of recycling can be varied whilst
pugmill mixer. working, making it easy to recycle tapering
Wirtgen manufactures one model of recycler sections often associated with highway inter-
that falls into this category: the WR 4200 shown changes and toll plazas;
in the picture on the facing page.
> the material in the recycling horizon is milled
This track-mounted machine is capable of and lifted into a twin-shaft pugmill mixer
recycling to a maximum depth of 200 mm and mounted on the machine. Two pumping
includes: systems allow for the accurate addition of
stabilising agents and water and the mixing
> an adjustable working width ranging between quality achieved is similar to that of conven-
2.8 m and 4.2 m. This allows the entire width tional stationary mixing plants; and
of one traffic lane to be recycled in a single
> lifting the recovered material off the road
pass. It also permits the location of longi-
and mixing on board the recycler achieves a
tudinal joints (between adjacent cuts) to be
uniformity of mix across the full cut width. In
selected to fall outside the trafficked wheel
other words, the material recovered from the
paths;
width of cut is fully cross-blended.
The Wirtgen WR 4200 with twin shaft pugmill mixer

The treated material is discharged from the pugmill material true to the required profile. This screed
mixer onto the road as a windrow and spread by is equipped with both tampers and vibration for
auger. A variable width paving screed is attached pre-compaction.
to the rear of the machine for placing the recycled

88/89
3.4 Cold recycling applications

The cold recycling process has numerous possible > The outcome required (i.e. service life
applications for maintaining and rehabilitating road expectations).
pavements. However, each application will be
project specific with three primary factors dictating The following sketches show three different pave-
the method of recycling that is appropriate: ment distress conditions with some of the different
options that can be applied for addressing the
> The type of pavement distress that needs relevant distress.
to be addressed;
> The quality of material in the recycling horizon;
and
 Depth from surface Extent of pavement problem: depth 100 mm - 150 mm

0 mm
100 mm

300 mm

600 mm
Alternative rehabilitation options
Cold recycling options Conventional options

Surface recycling - Asphalt overlay

(100% RAP) - Mill and replace

Rehabilitation options for upper pavement / surfacing distress

Depth from surface Extent of pavement problem: Maximum depth 300 mm

0 mm
100 mm

300 mm

Alternative rehabilitation options

600 mm
Cold recycling options Conventional options

In-place recycling with various - Asphalt overlay after extensive

stabilising agents, with / without patching
blend material / additional layers - Construction of additional
layer(s)

Rehabilitation options for structural distress in the upper pavement layers

90 / 91
 Depth from surface Extent of pavement problem: > 300 mm

0 mm
100 mm

300 mm

600 mm

Alternative rehabilitation options

Cold recycling options Conventional options

Two-part recycling with various - Reconstruction

stabilising agents, with / without - Construction of additional
blend material / additional layers layer(s)

Rehabilitation options for deep-seated structural distress

The materials encountered within the recycling These two material classes are discussed in the
horizon can be classified in to two primary material following sections together with the different types
types: of treatment that may be considered for improving
the engineering properties of the recycled mate-
> 100% reclaimed asphalt pavement (RAP) mate- rial, where necessary, in order to meet service life
rial where the depth of recycling encounters only expectations of the rehabilitated pavement.
asphalt; and
> Blend of RAP / Granular material where the recy-
cling depth includes layers of different materials
used to construct the upper portion of the pave-
ment. These include RAP, bituminous surfacing
materials, crushed stone and natural gravels, as
well as materials that were previously stabilised
(mainly crushed stone and natural gravel).
3.4.1 100% RAP recycling

Recycling only RAP material requires consideration Depending on the structural requirements, the
of the following factors: recycled RAP material may be treated with an
additive, or returned to the pavement as a granular
> nature and composition of the existing asphalt material (see Chapter 6).
(i.e. type of mix, grading of aggregate, binder
content, ageing of bitumen, etc.);
> type and cause of distress (i.e. permanent defor-
mation (wheel path rutting / shoving) or cracking
(thermal or fatigue mode));
> extent of distress (i.e. isolated or extensive); and
> purpose of recycling (i.e. holding action or
restoring structural integrity).

Working direction

Recycling 100% RAP material

92 / 93
3.4.2 Blend of RAP / Granular material

Structural distress is usually addressed by recy- > Reprocessing. Both surfaced and unsurfaced
cling the existing pavement to depths in excess roads constructed from natural materials are of-
of 200 mm. The recovered material will normally ten rehabilitated or upgraded to blacktop stand-
include the surfacing and base layers and typically ards by recycling the existing base material.
comprises bituminous surfacing material (e.g. RAP The primary objective of this type of recycling is
or aged chip seal material) and granular material to recover material from the existing pavement
from the underlying base (e.g. crushed stone or that already meets the strength requirements.
previously stabilised gravel). Distress in such pave- Water is added whilst recycling to achieve the
ments is usually manifest in the form of severe optimum moisture content for compaction. The
cracking in the surfacing, deformed granular layers mixed material is then placed and compacted to
and potholing. The structural capacity demands the correct layer thickness, surface shape and
and the type of traffic carried on the road will density.
largely dictate whether the materials recovered
from the existing pavement are sufficiently strong > Mechanical modification. Surface distress or
or if they need to be enhanced by stabilisation, as structural deterioration is sometimes caused
discussed below. by a mechanical deficiency in the existing base
material (e.g. poor grading or plasticity). Such
Without stabilising agents deficiencies can sometimes be addressed at
Rehabilitation by recycling the upper portion of minimal cost by blending the existing base
an existing pavement does not always demand material with a suitable imported material (e.g.
the addition of stabilising agents to improve the graded crushed stone) that is spread as a layer
engineering properties of the recycled material. on top of the existing road surface prior to
The two types of treatment discussed below are recycling. Water is added during the recycling
often considered where the road carries relatively process to achieve the optimum moisture
low volumes of traffic. content for compaction. The mixed material is
then placed and compacted to the correct layer
thickness, surface shape and density.

Mechanical modification achieved by spreading new material before recycling

It should be noted that blending a clayey mate- Where the structural capacity of an existing pave-
rial with sand using this technique is not always ment needs to be increased to meet additional
successful, especially where the moisture content traffic demands, the depth of recycling is gener-
of the clay is above the optimum. Under such ally increased to achieve the thickness of new
conditions, it is best to first pulverise the in situ stabilised layer that will provide such additional
clayey material with a recycler and allow the fluffed capacity. However, this is only possible where the
material to dry back. The loose material must then existing pavement includes layers of good quality
be pre-shaped and compacted before importing material that are sufficiently thick to accommodate
the sand and using the recycler to mix the material such an increase in the depth of recycling.
by making a second pass. If these conditions do not exist, an additional layer
of material may be imported and spread on top of
With stabilising agents the existing road surface prior to recycling, thereby
Stabilising agents are used to improve the en- achieving the required increase in thickness.
gineering properties of a material and different
stabilising agents used to enhance different prop- Unsurfaced roads are often upgraded to blacktop
erties. For example, stabilising with hydrated lime standards by recycling the existing gravel wearing
reduces the moisture susceptibility of a material course with an appropriate stabilising agent. The
by modifying the clay fraction, whereas a bitumen upgrading of such roads to blacktop standards is
stabilising agent increases the flexural strength of normally undertaken for the following reasons:
a material. These are discussed in the following
chapter. > Economics. High maintenance costs normally
associated with increased traffic volumes;

> Environmental concerns. Annual gravel loss

of between 25 mm and 50 mm is common for
unsurfaced roads, requiring continual regravel-
ling with material imported from borrowpits. In
addition, dust generated by unsurfaced roads
has been shown to be the cause of health prob-
lems for local residents and can cause serious
damage to agricultural produce.

> Strategic decisions. Safety concerns in wet

conditions and/or political priorities.

94/95
Since the structural capacity requirement for these Where the existing base was originally constructed
roads is usually low (< 300,000 ESALs) the thick- from cement stabilised material, distress is often a
ness of existing gravel wearing course material consequence of shrinkage cracking that mani-
normally recycled is between 100 mm and fests as block cracks at the surface. The distance
150 mm when stabilising with bitumen. Stabilising between individual cracks reduces with time and
with either cement or hydrated lime requires the the level of distress increases (normally associated
depth of recycling to be increased to between with water ingress and pumping) until the stage is
175 mm and 200 mm to achieve a similar struc- reached when rehabilitation is required. Such pre-
tural life. The new base layer is normally surfaced viously stabilised materials can usually be recycled
with a relatively light chip seal with a polymer and restabilised.
modified binder applied on a cemented base.
3.5 Benefits of cold recycling

Some of the more important benefits that can be > Subgrade disturbance is minimised. Distur-
realised by adopting cold recycling for pavement bance of the underlying pavement structure is
rehabilitation include: minimal since recycling is typically a single-pass
operation and the wheels of the recycler are not
> Environmental benefits. Full use is made of the in contact with lower layers since they run on
material in the existing pavement and the vol- top of the recycled material.
ume of new material that needs to be imported
from quarries is minimised. As a result, haulage > Shorter construction time. Recyclers are
is drastically reduced, as is the damage caused capable of high production rates that signifi-
by heavy vehicles travelling on existing roads cantly reduce construction times compared to
in the vicinity of the project. The overall energy most alternative rehabilitation methods. Shorter
consumed by recycling is significantly less com- construction times reduce project costs. Other
pared to all other rehabilitation options. benefits accrue to the road user since traffic is
disrupted for shorter time periods.
> Quality of the recycled layer. Consistent, high
quality mixing of the in-situ materials with water > Safety. One of the most important benefits of
and stabilising agents is achieved using modern this process is the relatively high levels of traffic
recyclers. Micro-processor controlled pumping safety that can be achieved. The full recycling
systems ensure the accurate addition of fluids operation can be accommodated within the
(water and stabilising agents). The tool pattern width of one traffic lane. On roads with two
and the method of mounting on the cutting lanes, recycling is usually undertaken in half-
drum are specifically designed to promote mix- widths with one-way traffic accommodated on
ing and achieve a homogeneous product. the opposite half during working hours. The full
road width is usually opened to traffic outside
> Structural integrity. Modern recyclers are capa- working hours, including the completed recycled
ble of producing thick layers of bound material lane.
that are homogeneous and do not contain weak
interfaces between thinner pavement layers (the > Cost effectiveness. The above benefits all
lamination effect). combine to make cold recycling a most attrac-
tive process for pavement rehabilitation in terms
of cost effectiveness.

96/97
3.6 Applicability of the cold recycling process

Each road rehabilitation project is different in terms > Availability of materials. The feasibility of vari-
of the structure of the existing pavement, the qual- ous recycling options is significantly influenced
ity of materials in the various pavement layers and by the availability of construction materials,
the service life requirements. The most cost-effec- especially stabilising agents. These must be
tive method of rehabilitation will always be project- procurable in sufficient quantities that are of a
specific. It is therefore important to determine the consistent and acceptable quality. Recyclers
most appropriate solution for each project and consume large quantities of water and stabilis-
that solution may not necessarily be one based on ing agents and it is necessary at the outset to
recycling. The following important factors need to determine whether the required volumes can be
be considered in evaluating the appropriateness of reliably sourced and delivered.
recycling for a specific project:
The next chapter focuses on stabilising agents, es-
> Type of project. The most effective solution for pecially those that are generally used with recycled
a particular country or region is influenced by materials. In the past, cement has been used more
the local environment, whether the project is than all other stabilising agents combined but, as
concerned with a highly-trafficked urban street a consequence of the rigidity introduced by such
where only night work will be permitted, or a low treatment, the current trend is towards bitumen
volume unsurfaced rural road in urgent need of stabilisation and the durability offered by improved
upgrading. Very different solutions and stand- flexibility.
ards of service are required in these two extreme
cases. It is important to take cognisance of the
local standards for road construction, as well as
the perceptions of the local population regarding
the levels of service that they regard as accept-
able.

> Physical environment. Local topography must

be considered when determining the most ap-
propriate method of rehabilitation. In particu-
lar, steep gradients may dictate the type of
construction that is practically possible. Climate
plays a vital part in the choice; solutions that
satisfy the requirements for pavements in a dry
region will certainly not be suitable for high rain-
fall areas and the effect of temperature extremes
will also influence the suitability of different op-
tions.
98/99
4 Stabilising agents

4.1 Types of stabilisation agents  103

4.1.1 General 103
4.1.2 Material behaviour 104
4.1.3 Cementitious stabilising agents 105
4.1.4 Bitumen stabilising agents 106
4.1.5 Summary of different stabilising agents 109
4.2 Stabilising with cement 110
4.2.1 General  110
4.2.2 Factors affecting strength 110
4.2.3 Cracking of cement stabilised layers 111
4.2.4 Surface crushing 114
4.2.5 Durability concerns 115
4.2.6 Working with cement 116
4.2.7 Early trafficking 120
4.2.8 Key features of cement stabilised materials 121
4.3 Stabilising with bitumen  123
4.3.1 Overview 123
4.3.2 BSM Distress Mechanisms 126
4.3.3 Primary determinants of BSM performance 127
4.3.4 Material to be stabilised with bitumen 128
4.3.5 Bitumen stabilising agents 136
4.3.6 Active filler 140
4.3.7 Water quality 141
4.3.8 Mix design procedure 142
4.3.9 Classification of BSMs 144
4.3.10 Working with BSMs 146
4.3.11 Mechanical tests 152
4.3.12 Pavement design approaches for BSMs 154
4.4 Summary: Advantages and disadvantages of cement
and bitumen stabilising agents 160

100/101
More than two thousand years ago, the pioneering aged innovation and the formulation of alternative
road construction technology of the Babylonians solutions, such as the development of stabilisation
and Mesopotamians was developed further by the techniques that utilise locally available resources.
Romans. In addition to their advanced segmented The required layer strength and stiffness can often
block paving systems (cobble stones), the Romans be achieved using a local marginal material
also used a form of lime treatment to improve with the addition of small amounts of stabilising
pavement strength for heavily-loaded transport agents at a relatively low cost. These techniques
wagons. Today, many different kinds of stabilis- are as applicable to recycling as they are to new
ing agents are used worldwide to overcome the construction. By adding a stabilising agent, the
inherent limitations of natural materials. In addition material recovered from an existing pavement can
to increasing the strength and stiffness charac- be improved, thereby eliminating the need to im-
teristics of a material, stabilising agents improve port new material to achieve the required strength
durability and resistance to the effects of water in the rehabilitated pavement structure.
and the environment.
The purpose of stabilising agents, their behaviour
Good quality road construction material is and, more importantly, the primary factors influ-
becoming increasingly scarce in many parts of encing their selection or exclusion must be clearly
the world and is often simply not available. The understood. This chapter endeavours to address
economic and environmental impact of importing these issues and remove any misconceptions.
and transporting a suitable material has encour-
4.1 Types of stabilisation agents

4.1.1 General

Currently, a wide range of stabilising agents is in > Material characteristics: Some stabilising
use globally, including: agents are more effective than others on certain
> Wetting agents i.e. surfactants (surface active material types. For example, lime should be
agents) e.g. sulphonated oils used in preference to cement for treating soils
> Hygroscopic salts e.g. calcium chloride with high plasticity (PI >10);
> Natural and synthetic polymers > Durability: The desired effects of stabilisation
> Modified waxes should remain effective for the service period;
> Petroleum resins and
> Bitumen > Policy: Some road authorities have rigid policies
> Cementitious stabilisers e.g. cement, lime, fly concerning the use of certain stabilising agents,
ash etc often influenced by past experiences.

All stabilising agents aim to achieve the same ob- The approach that is adopted towards stabilis-
jective of binding the individual aggregate particles ing agents differs between countries and road
together to increase strength and stiffness and/or authorities. Where these differences are dictated
make the material more water-resistant and dura- by policy they are often derived empirically rather
ble. Some agents are more effective than others than through sound technical evaluation. Technol-
when used with specific materials, some have ogy knows no borders; strength characteristics
clear cost advantages, but all have a place in the measured anywhere in the world are comparable,
market and most are best applied using modern provided the materials are similar and the testing
recycling machines. criteria common. There is therefore no real reason
for ruling out a specific stabilising agent that meets
New proprietary products are continually being all relevant technical requirements.
developed and it is important for the industry that
they be given a fair trial. Innovation should always Being at the frontier of technology can be a risky
be promoted since no single stabilising agent can and lonely experience. Engineers are inherently
claim to be the best for all applications. Engineers conservative and tried-and-tested practices are
should maintain an open-minded approach when most often preferred to experimentation with new
faced with making the decision as to which agent products. Cementitious stabilising agents and, to
to use on a specific project. Such decisions are a lesser extent, their bituminous relatives, have
invariably influenced, in order of importance, by: been well researched. They are used exten-
sively and standard test methods are available
> Price: The unit cost of stabilising will always be for determining optimum mix designs and quality
the primary concern; assurance requirements. In addition, both cement
> Availability: Specific stabilising agents may and bitumen see wide usage in the construction
not be available in some parts of the world. For industry and are generally available worldwide. It
example, bitumen emulsion is not manufactured is therefore not surprising that they are the most
in some countries; popular stabilising agents and, accordingly, are the
focus of this chapter.

102/103
4.1.2 Material behaviour

Unbound (granular) materials in flexible pavements It is important to understand that hot mix asphalt
exhibit a stress-dependent type of behaviour. This (HMA) and bitumen stabilised materials (BSMs)
means that. when confined within a compacted are fundamentally different. As was explained in
pavement layer, the effective stiffness character- Section 1.2.2, although bitumen is the common
istics increase as the loading state is increased. binding agent, the manner in which the bitumen is
When materials are repeatedly loaded to stress dispersed among at the aggregate particles in very
levels that are a significant proportion of their different. HMA is a continuously bound material
ultimate strength, shear deformation occurs. This whilst a BSM is a non-continuously bound material.
shear deformation accumulates, resulting in per-
manent deformation (rutting). Cement stabilised materials are also very different
from bitumen stabilised materials. Adding ce-
Adding a stabilising agent binds the material par- ment to a material promotes rigidity whilst adding
ticles together, changing the behaviour under load bitumen tends to promote flexibility. In addition
such that a layer of bound material tends to act to being continuously bound, cement stabilised
more like a slab with very different stress patterns. materials are prone to shrinkage that manifests as
The fundamental difference in stress distribu- block cracking in the layer, which is exacerbated
tion for unbound and bound materials (the cone by repeated loading. As shown in the below figure,
supporting the load and generation of bending tensile stresses develop in the lower portion of
stresses respectively), is shown in the figure below. layers constructed from bound material as the

Bound material = bending Unbound material = stress distribution cone

pavement deflects under load. Repeated load- BSMs are less stiff than cemented materials but
ing (normally millions of repetitions) causes the have improved shear properties. The main mecha-
material to suffer fatigue failure, or bottom-up nism of failure of BSMs is permanent deformation
cracking. The type of binding agent used is one under loading. BSMs with typical bitumen contents
of the primary determinants of the number of of less than 3% do not experience fatigue cracking
load repetitions a layer can withstand before such because they are non-continuously bound. These
cracking develops. This is applicable to cement concepts are all discussed in the following sec-
stabilised materials and HMA. tions.

4.1.3 Cementitious stabilising agents

Lime, cement and blends of these products with can be detrimental to the performance of the
fly ash, ground blast furnace slag and other such layer. Cementitious stabilising agents produce
pozzolanic materials, are the most commonly used semi-brittle materials. An increase in the strength
stabilising agents. Apart from the early Roman of a stabilised layer results in an increase in the
experiments with lime as an agent, cement has brittleness, with a consequential reduction in flex-
been in use for the longest period of time; the first ibility. Higher strength levels in the cemented layer
recorded application as a formal stabilising agent attract more stresses from the wheel loading as a
being in the USA in 1917. consequence. This invariably leads to accelerated
crack proliferation under repeated heavy traffic
The primary function of these agents is to increase loading, thereby reducing structural performance.
bearing strength. This is achieved either by signifi- It is therefore important that the performance
cantly increasing tensile and compressive strength requirements of the stabilised layer are unambigu-
of the material or by reducing plasticity. Cement is ously understood and that a proper mix design is
the stabiliser that provides most enhancement of conducted on representative samples to determine
strength. Free lime released during the hydration the correct stabiliser application rate.
process reacts with any clay particles that may be
present, thereby reducing plasticity. Lime, how-
ever, is comprised predominantly of free lime and
is therefore the preferred stabilising agent for more
plastic materials (PI>10%). The use of cement and
Note:
cement blends should be limited to the treatment
of materials with a Plasticity Index of less than 10. Cement stabilisation increases strength and
The strength achieved is governed by the amount stiffness but introduces shrinkage cracking
of stabilising agent added and the type of material
being treated. It must, however, be recognised that
adding more cement to obtain higher strengths

104/105
4.1.4 Bitumen stabilising agents

Due largely to technological advances (design The following describes the two bitumen stabilis-
procedures and construction methods) and the ing agents and explains the key differences
benefits that accrue (economic and environmen-
tal), the use of bitumen as a stabilising agent has
become increasingly popular. Although there are Bitumen Emulsion
many forms of bitumen, only two are used as
stabilising agents: bitumen emulsion and foamed This binder comprises bitumen emulsified in
bitumen, both produced from relatively soft road- water. The bitumen is dispersed in the water
grade bitumen (e.g. 80 Pen). Both may be used to in the form of an oil-in-water type bitumen
treat a wide range of pavement materials, dispers- emulsion. The bitumen is held in suspension
ing the bitumen in a non-continuous manner, the by an emulsifying agent, which determines the
hallmark of a bitumen stabilised material (BSM). charge of the bitumen emulsion. Cationic bitu-
Treating a material with cut-back bitumen is not a men emulsion has a positive charge; anionic
stabilisation process since the bitumen disperses bitumen emulsion has a negative charge.
in a continuous manner, as in asphalt. Bitumen emulsion is manufactured in a spe-
cialised plant and has a shelf life of several
Materials stabilised with bitumen do not suffer months in barrels, provided it is stored ap-
from the shrinkage cracking phenomenon associ- propriately.
ated with cement stabilisation. A layer constructed
from a BSM is relatively flexible compared a layer
of the same material treated with cement. BSMs Acid or
may be trafficked immediately after construction Surfactants
Caustic Soda
due to the substantial increase in cohesion that
is realised when the material is compacted. This
cohesion reduces the tendency of the material to Water Bitumen
ravel under the action of traffic. Bitumen stabi-
lisation improves the strength of a material and
reduces the detrimental effects of water.
Mill

Bitumen
droplets
suspended
5 microns
in water

Manufacture of Bitumen Emulsion

When an emulsion is mixed with aggregate,
the charged bitumen droplets are attracted
to the oppositely charged aggregate parti-
cles, focusing on the smaller fractions due to Hot
their surface area and charge concentration bitumen
features. The moisture and type of aggregate
play an important role in dispersing the bitu- Water Air
men emulsion and breaking (separating the
bitumen from the water) during mixing. Since Expansion
the bitumen emulsion acts as a lubricating Bitumen bubbles chamber
agent, the break should occur only after the encapsulating
steam Foamed
material has been compacted. The treated bitumen
material will have a speckled appearance
due to the concentration of bitumen on the
finer particles (as was illustrated in Chapter 1)
resulting in localised non-continuous bonds Foamed Bitumen Production in Expansion Chamber
(spot welding).
Foamed bitumen is produced at the mixing
Material stabilised with bitumen emulsion is
chamber and incorporated into the aggregate
termed BSM-emulsion.
while still in its unstable foamed state. The
greater the volume of the foam, the better the
distribution of the bitumen in the aggregate.
Foamed Bitumen During mixing, the bitumen bubbles burst,
This binder is produced by injecting water into producing tiny bitumen splinters that disperse
hot bitumen, resulting in spontaneous foam- throughout the aggregate by adhering to the
ing. The physical properties of the bitumen are finer particles (fine sand and smaller) to form a
temporarily altered when the injected water, mastic (as was illustrated in Section 1.2.2). The
on contact with the hot bitumen, is explosively moisture content of the material prior to mixing
transformed into vapour, which is trapped in plays an important role in dispersing the bitu-
thousands of tiny bitumen bubbles. The foam- men. On compaction, the bitumen particles in
ing process occurs in an expansion chamber the mastic are physically pressed against and
(a relatively small thick-walled steel tube, adhere to the larger aggregate particles result-
approximately 50 mm in depth and diameter) ing in localised non-continuous bonds (spot
into which bitumen and water (plus air on welding).
some systems) is injected at high pressure. Material stabilised with foamed bitumen is
Foamed bitumen bubbles collapse in less than termed BSM-foam.
a minute.

106 / 107
Current practice prefers to treat BSM-emulsion Some materials treated with a bituminous stabilis-
and BSM-foam equally in terms of their perfor- ing agent have poor retained strength properties
mance properties. The main features of the behav- (ie. they lose strength when immersed in water).
iour of BSMs are: This can be addressed by the addition of an ac-
tive filler such as hydrated lime or cement. Small
> A significant increase in cohesion in relation amounts of active filler (1% by mass) can signifi-
to that of the parent granular material, with no cantly increase retained strength without affecting
significant reduction in friction angle, the flexibility of the layer. Active filler acts as a
> Acquiring of flexural strength due to the visco- dispersion catalyst with foamed bitumen and pro-
elastic properties of the bitumen, motes breaking when used with bitumen emulsion.
> Improved moisture resistance and durability in It is therefore common practice to use cement
relation to the parent granular material. or hydrated lime in conjunction with bituminous
stabilising agents.
4.1.5 Summary of different stabilising agents

Conceptually, the effect of adding cement and / or

bitumen to a pavement material and the properties
of the different products are illustrated in the figure
below.

Stiff, brittle
behaviour Bitumen

Cement
Low Intermediate High

Strongly
cemented High
Increased resistance to permanent deformation

material

Lightly
cemented Intermediate
material
(reduced flexibility)

Bitumen stabilised Low

materials
(BSMs) Asphalt
Unbound concrete
material: High (BSM-foam and
quality crushed BSM-emulsion) None
stone and
aggregate

Moderate quality
natural gravel Time and temperature
dependent, visco-elastic
Low quality behaviour
natural gravel

Increased moisture resistance, flexibility

Stress dependent
behaviour

Pavement material behaviour

108 / 109
4.2 Stabilising with cement

4.2.1 General

Cement is the most commonly used stabilising Cement stabilisation, however, requires a proper
agent; its use worldwide far exceeds all other design approach. The primary function of cement
stabilising agents combined. The main reasons for addition is strength gain and the Unconfined
this are cost and availability; cement is manu- Compressive Strength (UCS) has achieved global
factured in most countries throughout the world acceptance as the principal design criterion. How-
and is relatively inexpensive. Another reason is its ever, several factors other than UCS need to be
proven track record as a construction material. considered, such as the rate of strength gain, the
There is a plethora of standards, test methods and Indirect Tensile Strength (ITS), cracking potential
specifications available and cement stabilised lay- and durability issues. These are addressed in the
ers have provided excellent service on thousands following sections.
of kilometres of roads.

4.2.2 Factors affecting strength

The compressive and tensile strength achieved in mum density as well as obtaining the anticipated
a cement stabilised material is largely determined strengths from the compacted material.
by the amount of cement that is added, the mate-
rial type, the density of the compacted material This is particularly important where ambient
and extent of curing. Strength generally increases temperatures exceed 40 C and where the material
in a linear relationship with cement content, but at is prone to rapid strength gain (e.g. amorphous
different rates for different materials and cement silica reaction). Under such conditions, an alterna-
type. Density plays a major role in determining tive stabilising agent to ordinary Portland cement
the ultimate strength whilst ambient temperature should be investigated, such as blends of slag-
directly affects the rate of strength gain; the higher ment and/or lime, with a slower rate of strength
the ambient temperature, the faster the rate of gain gain. It should also be noted that the finer the
of strength. cement powder, the faster the rate of cementation.

Crystalline bonds start forming between particles

as soon as cement comes into contact with water
in the mixing process. Some of these bonds are
destroyed when the material is compacted, there- Note:
by reducing the strength that can be achieved. In
addition, such bonding has the effect of reducing The strength of cemented material generally
the maximum density achievable. It is therefore increases linearly with cement content
important to expedite the placing and compaction
operations and complete them as soon as pos-
sible after recycling, in order to achieve maxi-
4.2.3 Cracking of cement stabilised layers

All cement-treated materials, including concrete, Tensile stresses develop within a cement-treated
are prone to cracking. The rate of gain of compres- material as a result of shrinkage and / or traffic and,
sive and tensile strength in cement stabilised ma- if these exceed the tensile strength at that time,
terial is a function of time, as shown in the sketch. cracks occur.

Concrete
Strength

Cement stabilised

7 28 365
Time (days, log scale)

Strength / time relationship for cemented material

Such cracks can be controlled and are not hydrates in the presence of water and is therefore
necessarily detrimental. However, it is important not traffic induced. The second is caused by the
to recognise that cement treated material tends repeated loading of traffic over a period of time.
to crack for two very different reasons. The first Crack initiation and subsequent propagation are
is caused by shrinkage that is a function of the entirely different processes, warranting that they
chemical reaction that takes place when cement be considered separately.

110 / 111
Shrinkage cracks. Cracks are inevitable when a cement hydrates and the material dries. Limiting
material is treated with cement. As cement hy- the moisture content (or lowering the water :
drates, complex finger-like calcium silicate crys- cement ratio) at the time of compaction to less
tals form, bonding the material particles together. than 75% of saturation moisture content can
In addition to heat generation, numerous other significantly reduce the degree of cracking;
changes take place during this chemical reaction.
As the bonds develop, the material experiences a > The rate of drying. When cement treated mate-
volume change and shrinks, causing cracks that rial shrinks, internal stresses are induced within
are commonly referred to as shrinkage cracks. the material. The degree of cracking is largely
These shrinkage cracks are unavoidable and are determined by the rate of strength development
one of the features of working with cement. relative to the rate of shrinkage stress develop-
ment. If the material dries quickly then shrinkage
The intensity (crack spacing) and magnitude stresses will inevitably be greater than strength
(crack width), known collectively as the degree of development and the crack pattern will be
cracking, is largely influenced by: intense (2 m x 2 m) with narrow cracks (typically
hair-line). Slow drying will see a less intense
> Cement content. The shrinkage that occurs pattern develop (6 m x 4 m) with wider cracks.
during hydration is a function of the amount of Proper curing of the completed layer will prevent
cement present. Increasing the cement content the surface from drying out, thereby reducing
therefore increases the degree of cracking and both the intensity and magnitude of cracking;
is one of the underlying reasons for minimis- and
ing the addition of cement to achieve only the
design requirements. However, as discussed > Inter-layer bonding. A rough interface with good
below, strength and durability requirements have bonding between the cemented layer and the
to be balanced and it is therefore not always underlying layer will result in a high intensity
possible to keep cement addition low; of hair-line cracks, as described above. This
scenario of full-friction between layers is most
> Type of material being stabilised. Some materi- common. However, in exceptional cases, poor
als tend to shrink more than others when treated inter-layer bonding can occur, resulting in a less
with cement. In addition, some plastic materials intense pattern with wider cracks.
tend to be active, exhibiting significant volume
changes between moist and dry states. Where A feature of shrinkage cracks is that they are wider
the PI of the material is in excess of 10, the at the top than at the bottom (drying initiates at the
addition of lime, or a combination of lime and surface) and the vertical face is irregular, allowing
cement should be used to reduce plasticity, ide- for the effective transfer of traffic loads across the
ally to a non-plastic state; crack.

> Compaction moisture content and cement :

water ratio. The degree of cracking is a function
of the amount of moisture that is lost as the
Cracks caused by traffic. These cracks occur deteriorates further under repeated traffic loads.
as a consequence of repeated tensile stresses This reduces the effective modulus that, in turn, in-
induced by traffic loads in the cement stabilised creases deflection under load, thereby promoting a
layer. Crack initiation occurs at the bottom of the continuous process of degradation until ultimately
layer where tensile stresses are maximum, causing the material approaches its pre-stabilised granular
maximum strain. Being semi-brittle with relatively state. It is important to note that the rate of degra-
poor flexural properties, cement treated layers are dation accelerates once cracks reach the surface
extremely sensitive to overloading (the proverbial and allow water to penetrate the pavement more
pane of glass on a feather bed). freely. The wetter the region, the higher the risk of
this moisture accelerated distress.
Even without overloads, damage from continuous
trafficking accumulates in the cemented layer, ulti-
mately resulting in fatigue cracking. Once initiated,
the cracks take some time to propagate through
to the surface of the layer. In its post-cracked
state, the layer is still able to support traffic loads,
a state that can be modelled by reducing the ef-
fective modulus of the cement treated layer. Crack
intensity and magnitude increase as the layer

Cracks caused by traffic

112 / 113
4.2.4 Surface crushing

Crushing occurs in the upper portion of a cement- Design procedures need to address these condi-
stabilised base layer when traffic induced stresses tions. Increasing the compressive strength and
exceed the compressive strength of the material. thickening both the stabilised and surfacing layers
This failure mechanism was initially identified in will largely solve the problem. However, over-
lightly-cemented base layers with relative thin loading remains the major cause of premature
surface treatments on roads carrying heavy coal pavement failure where axle control is ineffectual.
traffic in South Africa. The conclusions drawn from In addition, increasing tyre pressures over the
research into this mechanism showed that the last decade has exacerbated the potential for this
potential for crushing failure depends on: failure mechanism.

Furthermore, cement-stabilised base layers are

> the compressive strength of the stabilised
vulnerable to crushing failure if subjected to
material in the upper portion
heavy traffic loading before sufficient strength has
of the layer;
developed. This applies especially to traffic that is
> the thickness and type of surfacing; and accommodated on newly constructed cemented
layers rather than deviations. The cement normally
> the tyre pressure as well as the applied axle used for stabilisation and the amount applied
load. results in a relatively slow rate of strength gain
after construction, taking some 7 days to achieve
50% of the ultimate strength level (and 90% after
28 days). Thus, when the target UCS for a base
layer is 2 MPa, less than 0.5 MPa would have been
achieved after 3 days, making the surface particu-
larly vulnerable to slow-moving heavy vehicles with
high tyre pressures.

Asphalt

Crushing Cemented base

Fatigue Cemented
subbase
Fatigue
Subgrade

Mechanisms of failure for cemented layers

4.2.5 Durability concerns

The durability of natural material is mainly related rapidly. Agricultural lime is useful to increase pH
to weathering and degradation of the individual but does not assist in reducing PI. The change
particles under the influence of climatic condi- in molecular structure during carbonation is also
tions and repeated traffic loads. Such degradation associated with a volume change (either decrease
is a slow process and material properties usually or increase, depending on the quality of the ag-
remain reasonably constant over the life span of gregates and type of chemical stabilisation). Where
the road, especially if higher-quality materials are the forces emanating from such volume change
used. However, when poorer quality material is exceed the strength of the cement treated mate-
stabilised with cement, additional aspects con- rial, destruction occurs.
cerning durability need to be considered. Under
certain conditions their properties can change This phenomenon is well known in the concrete
over short periods of time due to carbonation and industry, but is seldom a cause for concern since
climatic influences. the tensile strength of concrete is always far in
excess of the stresses induced by carbonation. In
The CBR (California Bearing Ratio) test is widely addition, the aggregate used in the manufacture of
used as an indicator of bearing strength for natural concrete is generally crushed stone with excellent
materials but is not appropriate for higher strength durability properties. This is not the case where
cemented materials. The UCS test was therefore relatively poor-quality materials are stabilised with
adopted and relevant limits are used worldwide a low application rate of cement.
(e.g. maximum 4 MPa, minimum 2 MPa). How-
ever, research has shown that UCS alone is not a Several durability-type tests can be carried out in
reliable indicator of durability; a stabilised material the laboratory to determine the potential for car-
meeting the specified UCS requirements can bonation, such as the wet/dry brushing test, de-
deteriorate and disintegrate over a short period termination of the Initial Consumption of Lime (ICL)
of time. Additional tests are therefore required to or cement (ICC) and the ITS value. As a general
ensure that a cement stabilised material is suf- guideline for minimising the risk of carbonation,
ficiently durable, particularly against the potentially sufficient cement must be added to achieve both
destructive effects of carbonation. a minimum ITS value of 250 kPa (regardless of
whether the related UCS exceeds the prescribed
Carbonation is the name given to a complex limit) as well as satisfying the ICC demand.
chemical reaction that occurs between a cement-
ed material and carbon dioxide in the presence
of water or water vapour (humidity). In laymans
terms, this reaction produces calcium carbonate Note:
from the free calcium ions. The calcium hydroxide
molecules (or free lime, Ca(OH)2, that are always The initial consumption of stabiliser (ICS)
present in a cement treated material) are converted needs to be satisfied to avoid premature
into agricultural lime CaCO3 through the process carbonation (i.e. durability problems).
of carbonation. This allows plasticity (measured
by the PI) to return to the material and increase

114/115
4.2.6 Working with cement

As described above, one of the main concerns Cement type. Finely-ground cement with rapid-
with cement treated material is the inevitable hardening properties should never be used as a
shrinkage cracking that occurs. However, the stabilising agent.
degree of cracking and the overall quality of stabi-
lised layers are largely contingent on the following Uniformity of application. Two different methods
key factors: are popular for the application of cement as a
stabilising agent on recycling projects. The first
Mix design. It is of primary importance that a spreads dry cement powder on top of the existing
proper mix design is carried out on truly repre- road surface prior to recycling, the second injects
sentative samples of the material to be treated a cement slurry into the mixing chamber whilst
with cement. (An example of such a laboratory recycling:
procedure is included in Appendix 1.) Different ma-
terials require different application rates of cement
to achieve targets for strength and durability.

Cement quality. Cement has a definite shelf life

and, as a rule of thumb, should not be used more
than three months after the date of manufacture.
Determining the age of cement is difficult, particu-
larly when imported in bulk. If there is any doubt
about the age, or any other aspects of quality,
samples should be tested to check the strength
parameters.
> Bulk spreaders. This is the most widely used Any form of dry cement spreading is affected by
method of application. Several different systems the weather, particularly wind and rain. Being a
are used for discharging the cement onto the fine powder (smaller than 0.075 mm), cement is
road surface at the required spread rate (belt susceptible to wind erosion and readily becomes
conveyors, auger feeders, pneumatic blowers) airborne when fanned by a breeze, whether natural
and each has its own particular merits, and or caused by passing trucks, thereby affecting the
demerits. The canvas patch test is normally rate of application non-uniformly. If rain makes
conducted to verify the application rate. All bulk contact with spread cement, it triggers the hydra-
spreaders have their limits and care must be tion process. When this occurs, the spread cement
exercised when trying to apply very low applica- must either be mixed immediately or discarded;
tion rates (< 2%). (Streumaster spreaders can
provide application rates of < 2% cement or
lime with acceptable accuracy.)

Streumaster spreading

116/117
Applying cement with the Wirtgen WR 2500 SK (dustless)

> Recycler equipped with an integrated spread- > Slurry injection. The Wirtgen WM 1000 is
ing device. The WR 2500 SK is a stretched specifically designed to premix cement with the
version of the standard WR 2500 S recycler with amount of water required to achieve the opti-
a 4 m hopper integrated immediately behind mum compaction moisture content. The slurry
the operators cab. Cement or lime is drawn suspension thus formed needs to be sufficiently
from this hopper by means of a cellular wheel liquid to be pumped to the recycling machine
sluice and spread uniformly on the road surface and injected into the mixing chamber through
immediately in front of the mixing chamber. This a spraybar. The water : cement ratio is usually
dustless system is accurate for spread rates in the region of 1:1, but most recycling applica-
below 2% and up to 6% and addresses all the tions call for more water than cement to achieve
weather related concerns of spreading cement the optimum compaction moisture content.
on the road ahead of the recycler.
Applying cement as a slurry using the Wirtgen WM 1000 (dustfree)

Slurry injection is the most efficient means of dis- In addition, low application rates (1%) that are
persing the cement throughout the recycled mate- usually specified when stabilising with a bitumen
rial. This method of application is recommended stabilising agent are best applied by means of ce-
where high application rates are specified (>4%) ment slurry injection to ensure uniformity of
for deep recycling (>200 mm) when bulk spreading application throughout the recycled material.
becomes unmanageable due to the sheer volume
of cement required per square metre. (If the thick- Furthermore, the truly dustfree application that
ness of the layer of cement spread on the road is achieved using this method offers significant
exceeds 25 mm, extreme care must be taken to environmental benefits, both in terms of improved
maintain consistency. health and safety for the workforce and in the
reduction of wind-blown pollution.

118/119
Uniformity of mixing. Sufficient tests have been traffic should be kept off the layer. Alternatively, a
conducted to prove that the mixing capabilities of temporary seal can be applied as a curing mem-
large recyclers are similar to off-site plant mixers, brane. As a general rule, cement treated materials
provided the machine is operated at an advance should always be covered as soon as possible to
speed conducive to the specific site conditions minimise the detrimental effects of rapid drying
(normally between 8 m/min and 12 m/min). It is and carbonation.
therefore not necessary to apply a traditional fac-
tor to the specified application rate of cement as Temperature. Where ambient temperatures are
an allowance for site losses and inefficiencies. above 35 C, short sections of road should be
treated and finished off as quickly as possible to
Water addition. Cement treated material should prevent compacting against the inevitable rapid
be worked as dry as possible, both to minimise gain of strength.
shrinkage cracking and to prevent heaving during Due to the expansion that occurs when water
compaction. Where the addition of water is re- cools below 4 C, no cement stabilisation work
quired, it should always be injected into the mixing should be undertaken if freezing conditions are
chamber and such addition must be judiciously forecast.
controlled to obtain a moisture content that never
exceeds the optimum moisture content of the Control tests. The quality of the completed layer
material. is often judged on the strength (UCS, ITS) of
specimens manufactured from samples collected
Curing. Once complete, the surface of a cement behind the recycler. Where this is done, it is impor-
stabilised layer must be prevented from drying out tant to regularly monitor the time lapse between
for a period of at least seven days. As described sampling in the field and compacting the speci-
above, shrinkage cracks will develop at the surface mens in the laboratory. These tests must simulate
if the rate of drying exceeds the rate of gain of field conditions. Any significant delay could result
strength. Drying out can be prevented by frequent- in poor strengths due to the cement hydrating and
ly spraying the surface with water from a tanker gaining strength that is subsequently destroyed by
fitted with a full-width spraybar. Other construction compaction.

4.2.7 Early trafficking

Outside normal working hours the full road-width Allowing the surface to dry out can lead to ravel-
is sometimes opened to traffic. Concerns regard- ling and loss of strength in the upper portion of the
ing the early trafficking of cement stabilised mate- layer, ultimately causing potholes to develop.
rial are often expressed. As discussed in Section The surface should therefore be kept constantly
4.2.4, such concerns are certainly justified where damp by frequent light watering.
heavy axle loads can be expected and where the
required curing procedures are not followed.
4.2.8 Key features of cement stabilised materials

The three most important features of a cement The table below shows typical cement application
stabilised material are: rates (expressed as a percent of the dry density
of the recycled material compacted to the target
Strength. Both the compressive and tensile density) for two UCS categories: lightly-cemented
strength, measured by UCS and ITS tests respec- (less than 4 MPa) and cemented (up to 10 MPa).
tively, are important parameters for evaluating
cement-stabilised material. Caution: when working with coarse material. Due to
the increased probability of stone columns devel-
> The UCS value. The UCS test is normally used oping within the body of the specimen, a false UCS
for evaluating cemented materials. The UCS measurement that approximates the stone strength
value is usually determined from prepared speci- rather than that of the stabilised mix can result.
mens that have cured for 7 days at a temperature Unexpectedly high measurements should therefore
of 22C and a humidity of over 95%. Some test be investigated by visually inspecting the specimen
methods allow the curing to be accelerated. to determine the extent of aggregate crushing that
occurred during the test. Repeat tests may be nec-
essary to achieve a statistically reliable result.

Typical cement application rates (percent by mass)

Target UCS value
Material type
< 4 MPa Up to 10 MPa
RAP/crushed stone (50/50 blend) 2.0 to 3.0 3.5 to 5.0
Graded crushed stone 2.0 to 2.5 3.0 to 4.5
Natural gravel (PI < 10, CBR >30) 2.5 to 4.0 4.0 to 6.0

> The ITS value. The ITS test shows greater for cement treatment, measured from the time that
sensitivity to stabiliser content than the UCS test the cement first comes into contact with the mate-
and is also becoming increasingly important as a rial and moisture, to the time when compaction is
measure of long-term durability. As described in complete. This may be generous where there is a
Section 4.2.5, recent research has shown that a potential for rapid strength gain (see Section 4.2.2).
minimum ITS value of 250 kPa is required to resist
the destructive forces generated by carbonation. It is strongly recommended that the allowable delay
betwen mixing and compaction is checked by ana-
Processing time. Mixing, placing, compacting and lysing the rate of strength gain versus delay time for
finishing should be carried out in the shortest possi- the material to be stabilised under simulated field
ble time. A time limit of 4 hours is normally specified conditions (especially temperature).

120/121
 UCS: Unconfined compressive Strength (kPa)

1 800

1 750

1 700

1 600

1 550

1 500
0 2 4 6
Delay between mixing and compaction (hours)

The example in the figure shows that a time of the density at the top of the layer may be higher
2.5 hours is allowable to achieve the required than at the bottom. Where specified, it is normal
strength. It is important to minimise this time. also to include a maximum deviation of 2% for the
With proper planning, this time period can be density measured in the lowest one-third thickness
reduced to less than one hour using modern of the layer. Hence, if the average density specified
recycling and compaction equipment. is 100%, then the density at the bottom of the layer
must be more than 98%.
Density. Compaction should always aim to achieve
the maximum density possible under the conditions
prevailing on site (the so-called refusal density). Note:
A minimum density is usually specified as a per-
centage of the modified AASHTO density, normally The available working time in the laboratory
between 97% and 100% for cement-treated bases. must simulate field conditions
A density gradient is sometimes permitted by
specifying an average density. This means that
4.3 Stabilising with bitumen

Great strides have been made in recent years with of these advancements, incorporating the latest
regard to the research of bitumen stabilised ma- approaches to mix and pavement design of BSMs.
terials (BSMs) and the understanding of their key This section summarises the important features of
performance parameters. The Technical Guideline these developments.
TG2 Second Edition (2009) captures the essence

4.3.1 Overview

Bitumen is a versitile binder that is used in pave- This chapter is focused on the use of bitumen
ment layers in several different forms. However, emulsion and foamed bitumen which are the only
since bitumen is a highly viscous liquid and there- two viable bitumen stabilising agents. As de-
fore unworkable at ambient temperatures, the vis- scribed in Chapter 6, although bitumen emulsion
cosity must first be reduced to achieve workability. can also be used as a rejuvenating agent for 100%
In general, there are three ways of doing this: RAP mixes, this section focuses on stabilisation
only.
> applying heat (increasing the temperature of
bitumen and aggregate);
> emulsifying in water to form bitumen emulsion;
or
> creating foamed bitumen in a temporary state
of low viscosity.

122/123
The following table compares the treatment BSM-emulsion (material stabilised with bitumen
process for hot mix asphalt with those for the emulsion) and BSM-foam (material stabilised with
two means of stabilising a material with bitumen: foamed bitumen).

Comparison between different types of bitumen treatment

Stabilisation process Hot-mix Asphalt
Factor
BSM-emulsion BSM-foam (HMA)

- Crushed rock - Crushed rock

Aggregate types - Natural gravel - Natural gravel Crushed rock
applicable - RAP, stabilised - RAP, stabilised 0% to 50% RAP
(Cold mix Chapter 6) - Marginal (sands)

Bitumen mixing 160C to 180C

20C to 70C 140C to 180C
temperature (before foaming)
Aggregate temperature Ambient Ambient Hot only
during mixing (> 10C) (> 15C) (140C to 200C)
Moisture content during OMC plus 1% minus Fluffpoint
Dry
mixing emulsion addition 70% to 90% of OMC
Coating of finer particles
Coating of finest parti-
(and some coarse
cles only. Coating of all aggregate
Type of coating of ag- particles).
Increased cohesion particles with controlled
gregate Increased cohesion
from the bitumen/fines film thickness
from the bitumen/fines
mortar
mortar
Construction and com-
Ambient (> 5C) Ambient (> 10C) 140C to 160C
paction temperature
Air Voids 10% to 15% 10% to 15% 3% to 7%
Rate of initial
Slow (moisture loss) Medium (moisture loss) Fast (cooling)
strength gain
Modification of No. (Modifiers are gen-
Yes Yes
bitumen erally anti-foamants)
- Emulsion type
- Foaming properties - Penetration
Important bitumen (anionic, cationic)
Expansion ratio - Softening point
parameters - Residual bitumen
Half-life - Viscosity
- Breaking time
The type of material that is produced when treating The formulation of a BSM requires the considera-
with either bitumen emulsion or foamed bitumen tion of volumetric/compaction characteristics as
is similar. This has allowed a common approach to well as the engineering/durability properties. As
be developed for both types of treatment. There with all stabilised materials, the performance of the
are, however, some nuances, primarily concerned treated product is largely dictated by the quality
with the moisture/fluid content; these are high- of the parent material and its appropriateness for
lighted in the following sections. treatment with the selected stabilising agent. The
mix design procedure therefore aims to determine
Although BSMs may appear to be challenging the potential of the material in terms of structural
materials due to the different number and types of performance (resistance to permanent deforma-
possible ingredients, they are stabilised materi- tion), and durability (resistance to moisture and
als, not HMA derivatives. Similar to designing for deterioration). At the same time, economic con-
cement stabilisation, each component in the mix siderations remain paramount in the selection of
needs to be optimised to formulate a composite BSMs. Since bitumen contributes significantly to
product for a specific purpose or application. the cost of a BSM, the need for effective optimisa-
These components include the material (aggre- tion of the amount of bitumen added to the mix is
gate) to be treated, water, bitumen and active filler, of primary importance.
each with its own variability, availability and cost.
To produce a BSM with the necessary quality and
consistency to fulfil the intended function, sound
procedures need to be followed that assist in iden-
tifying optimal formulation, blending and produc-
tion. This process is the mix design procedure.

124/125
4.3.2 BSM Distress Mechanisms

Project-specific conditions (e.g. available materi- > Moisture Susceptibility. The partially coated
als, stabilising agents, climate, traffic, supporting nature of the aggregate in a BSM makes mois-
layers, construction techniques, etc.) all play a role ture susceptibility an important consideration.
in the performance of a material and its mode of Moisture susceptibility is the damage caused by
distress. BSMs need to be appropriately selected exposure of a BSM to high moisture contents
for the particular design conditions. By chang- and the pore-pressures induced by wheel loads.
ing the proportions of parent material, blending This results in loss of adhesion between the
aggregates, bitumen and active filler, is possible to bitumen and the aggregate. Moisture resistance
create a BSM mix that meets specific behavioural is enhanced by:
characteristics. Increased bitumen content, limited by stability
concerns and cost implications;
There are two fundamental failure mechanisms of Addition of active filler (maximum 1%);
BSMs that need to be considered: Higher field densities through improved
compaction; and
> Permanent Deformation. This is the accumula- stabilising a material with a smooth continuous
tion of shear deformation (plastic strain) as a grading
result of repeated loading and is dependent on
the materials shear properties and densification
achieved. Resistance to permanent deformation
(rutting) is enhanced by:
Improved material (aggregate) strength,
angularity, shape, hardness and roughness;
Increased maximum particle size;
Improved compaction (field density);
Reduced moisture content (curing);
Addition of a limited amount of bitumen, usu-
ally less than 3.0%. Higher bitumen contents
encourage instability due to the lubricating
Note:
effect of excess bitumen and a consequential
lowering of the angle of internal friction; and BSM behaviour is similar to unbound granu-
Addition of active filler, limited to a maximum lar materials but with greatly improved cohe-
of 1%. Higher application rates of active filler sion and reduced moisture sensitivity
introduce brittleness which encourages shrink-
age and traffic associated cracking
4.3.3 Primary determinants of BSM performance

The performance of a pavement layer constructed Furthermore, additional layers placed on top of
from BSM is determined primarily by: the new BSM layer provide protection from the
environment and high traffic-induced stresses.
> The quality and consistency of the parent mate- Ultimately, as was explained in Chapter 2, pave-
rial (together with any blending material) that ments are designed to accommodate specific
was stabilised; levels of traffic loading. Although a layer of BSM is
> The quality and appropriateness of the bitumen only one of several components that make up the
stabilising agent and active filler applied; overall pavement structure, each feature influenc-
> The dosage of bitumen stabilising agent/active ing performance of the BSM layer needs to be
filler and the mixing effectiveness; understood to ensure that an appropriate design is
> The density achieved in the field by applying ultimately achieved.
appropriate compaction effort; and
> The thickness and uniformity of the constructed
layer.

In addition, key features of the existing pavement

remaining beneath the recycling horizon have a
significant effect on the layer of BSM, both in the
as-built characteristics (especially density) and in
the overall performance of the pavement:

> The quality and uniformity of the underlying

supporting layers (pavement composition);
> The prevailing climatic conditions (hot/cold,
wet/dry); and
> The effectiveness of drainage provisions that will
influence the equilibrium moisture content of all
pavement materials, including the BSM.

126/127
4.3.4 Material to be stabilised with bitumen

The quality and composition of material recycled In addition, the maximum particle size and
from an existing pavement can vary considerably. amount of coarse material plays an important
Such variations are due to: role in achieving sufficient density in the field. The
maximum particle size should be limited to 1/3 of
> The structure of the existing pavement (materi-
the layer thickness (i.e. 50 mm for a 150 mm thick
als in the various layers and their thickness);
layer) and, as a general guideline, the amount of
> Construction variability (material quality and
material retained on the 50 mm sieve should not
thickness);
exceed 10 %.
> Depth of recycling (additional layers may be en-
countered in the recycling horizon as the depth
increases);
> Age of the pavement (particularly for previously- > RAP material. Some projects encounter a high
proportion of RAP material in the recycling
treated materials and materials prone to weath-
horizon (> 90% of the recovered material). In
ering);
> Degree of patching and repair of the existing such cases, the influence of the aged bitumen
in the RAP needs to be carefully considered,
pavement; and
> Thickness and nature of old surfacing materials especially in warm climates where heavy traffic
loading is expected. . In particular, the following
(e.g. asphalt and/or seals).
aspects need attention:
Climatic region. In hot climates, the shear
The recycled material should be well-graded and properties of the mix should be determined
comply with the criteria provided below. from triaxial tests carried out at representative
temperatures;
Actual axle loads. Where a BSM layer is
On some projects, the required grading may be
intended for use in an area where there is
achieved by incorporating a portion of the underly-
limited axle mass control, higher stresses from
ing layer into the composite recycled material.
overloads may cause accelerated deformation.
However, caution must be taken when including an
This needs to be considered when analysing
underlying layer comprised of cohesive material. It
the shear properties.
is preferable to rather incorporate imported mate-
The composition of the RAP material. In hot
rial (normally a crushed product) by pre-spreading
regions with limited axle mass control, the RAP
as a layer on the surface before recycling.
should always be modified by blending with
15% to 25% crusher dust. This will provide an
angular sand skeleton to improve the shear
resistance of the mix.
The use of 100% RAP material in BSMs is covered > Grading. Sieve analyses carried out on repre-
in detail in Chapter 6. sentative samples of material to be stabilised
with bitumen provide a good indication of
Two key characteristics of the parent (pre-stabi- the suitability for such treatment. The follow-
lised) material are used as indicators to determine ing graph and table show the recommended
whether or not bitumen stabilisation is likely to be grading envelops for material treated with either
effective: the grading curve and plasticity index foamed bitumen or bitumen emulsion.
(PI). Although a CBR value can be estimated for
the material using these two characteristics, it
is advisable to carry out the 4-day soaked CBR
test when the quality of the material is considered
marginal for the intended application.

100
90
80
Percentage Passing

70
60
50
40
30
20
10
0
0,1 1 10
Sieve Size (mm)

Target Gradings Less suitable (gravel) Typical RAP

Target grading curves for bitumen stabilisation

128/129
 Recommended grading envelops for bitumen stabilisation

Treatment with Foamed Bitumen Treatment with Bitumen Emulsion

Percentage passing each sieve size (%) Percentage passing each sieve size (%)
Sieve size
(mm) Recommended Recommended
gradings Less Typical gradings Less Typical
suitable RAP suitable RAP
Coarse Fine (Gravel) grading Coarse Fine (Gravel) grading

50 100 100 100 100 100 100 100 100

37.5 87 100 100 85 87 100 100 85

26.5 76 100 100 72 76 100 100 72

19 65 100 100 60 65 100 100 60

13.2 55 90 100 50 55 90 100 50

9.5 48 80 100 42 48 80 100 42

6.7 41 70 100 35 41 70 100 35

4.75 35 62 88 28 35 62 88 28

2.36 25 47 68 18 25 47 68 18

1.18 18 36 53 11 18 36 53 10

0.6 13 28 42 7 12 27 42 6

0.425 11 25 38 5 10 24 38 4

0.3 9 22 34 4 8 21 34 3

0.15 6 17 27 2 3 16 27 1

0.075 4 12 20 1 2 10 20 0

Note. Sieve analyses are always carried out using the washed-fines test method.
The nature of bitumen dispersion is different for As explained in Chapter 6, particles of all sizes in a
foamed bitumen and bitumen emulsion and is the RAP material were previously coated with bitumen
reason for the minor differences between the two and this old bitumen appears to offer the bitumen
recommended grading envelops (confined to the splinters a home. The minimum dust requirement
fractions smaller than 2.36 mm). may therefore be reduced as the amount of RAP
in the mix increases. (Recycled chip seal material
Foamed bitumen. Unless the sample is comprised has the same effect on the mix as RAP.)
mainly of RAP material, foamed bitumen treatment
relies on the dust particles (<0.075 mm) to provide Bitumen emulsion. Mixing a material with bitu-
a home for the bitumen splinters produced when men emulsion is essentially a wetting process.
the bitumen bubbles burst. Hence the minimum However, the charge on the individual emulsi-
requirement for 4% of the material to pass through fied bitumen droplets will tend to see a selective
the 0.075 mm sieve. If there are insufficient dust attraction to those particles with higher opposite
particles to disperse the added bitumen, individual charge concentrations (the finer fractions).
bitumen splinters will tend to adhere to each
other, forming bitumen-rich blobs known as
BSM-emulsion
stringers. These stringers are effectively wasted
bitumen and are actually detrimental to the mix; Bitumen emulsion coats some of the coarse
they have a negative effect on the angle of internal particles, not only the fines.
friction without having a corresponding positive A minimum fines content of 2% is sufficient.
influence on the cohesion of the material.

BSM-foam
The dispersed bitumen splinters in BSM-
foam do not coat the larger particles. The
mastic (fines, bitumen and water) spot weld
the coarser aggregate fractions together in
BSM-foam. 4% fines are normally required to
achieve a satisfactory mix (unless 100% RAP
material is treated see Chapter 6).

130/131
Blending. The grading curve for the material that Since bitumen stabilised materials are reliant on
is to be bitumen stabilised should always be plot- the sand skeleton for their performance, the por-
ted on a graph that includes the target gradings tion of the plot between 0.075 mm and 2.0 mm
(taken from the above table). The plot will highlight is of primary importance. All available sieve sizes
any glaring deficiencies and suggest the need in this range for both asphalt and soils testing
for blending as well as the type/size of blending are used in the sieve analysis to obtain as much
material. Material with a grading curve that falls information on this lower portion of the grading
outside the recommended grading envelope (tar- curve and the curve should always be plotted
get range) can normally be successfully stabilised to determine whether the curve is continuous or
without blending, provided the following condi- bulging (i.e. gap graded). Blending is advised
tions are satisfied: where there is a marked bulge in the 0.075 mm
to 2.0 mm portion of the curve, as shown in the
> Finer material. The plasticity index (PI) is below example opposite.
10.
> Coarser material. The material is comprised In this example, blending the material with 15%
mainly of RAP or recycled chip seal material. minus 5 mm crusher dust largely eliminates the
bulge in the curve below 2 mm. This will improve
It must be appreciated that materials falling out- the overall strength of the material (sand skeleton)
side the target grading envelope have limitations and allow a higher level of density to be achieved
and should therefore only be utilised where alter- which will reduce moisture susceptibility as well as
natives are severely limited. The finer the material, adding to the strength achieved in the field.
the higher the bitumen demand whilst coarser
materials are prone to segregation and can be
extremely difficult to work. Blending often offers a
technical solution whilst reducing the overall cost.

Note:

Under no circumstance should active filler

be used as a blending material. Only inert
materials (e.g. crusher dust) are to be used
for blending. The maximum amount of active
filler that is added to a BSM is 1.0%.

A deficiency in fines should never be cor-

rected by increasing the active filler to > 1%
 100
90
80
Percentage Passing

70
60
50
40
30
20
10
0
0,1 1 10
Sieve Size (mm)

Target Gradings Sample < 5 mm Crusher dust 15% dust blend

Example: Blending to correct poor grading

Alternatively, when selecting the various compo- n

nent materials in a blending exercise (e.g. when
using a KMA 220 for in-plant mixing), the follow-
ing equation can be useful for determining the
P= d
D [ ] x 100
required blend proportions. This allows the best
particle packing and minimum voids (after com-
paction) to be achieved by calculating the amount where d = selected sieve size (mm)
required for each individual particle size. P = p ercentage by mass passing a
sieve of size d (%)
D = maximum particle size (mm)
n = grading coefficient dependent
on particle packing characteris-
tics (a value of 0.45 is recom-
mended)

132/133
> Plasticity. Plasticity in a material is attributed to always needs to be viewed in light of the PI value;
the presence of cohesive clay particles in the if the PI is in excess of 10, the fines are likely to be
fines fraction. Following standard laboratory lumped and therefore unable to act as a dispers-
test procedures, the Plasticity Index (PI) of a ing agent for the bitumen.
material is determined on the fraction passing
the 0.425 mm sieve and is a primary indicator of As a general rule, if test results show the PI of the
the materials moisture susceptibility. The higher material is in excess of 10 then pre-treatment with
the PI value, the greater is the amount of clay in hydrated lime should be carried out. Such treat-
the material. Due to the shape and size of the ment modifies the material by separating the clay
individual particles, clay has an ability to retain particles, thereby eliminating plasticity.
relatively high levels of moisture. In such a moist
state, clay is highly cohesive, implying that the
particles bind together in lumps. Recycling a Durability. Durability is primarily concerned with
high PI material will not necessarily break these the properties of the pre-stabilised material. Where
lumps apart and they will remain in the new a poorer quality marginal material is to be stabi-
layer as localised weak spots that retain their lised with bitumen, it is advisable to check the un-
moisture susceptible nature. treated materials susceptibility to moisture change
and weathering. Since the dispersion of bitumen
In addition, foamed bitumen stabilisation relies in a BSM is selective, most of the coarse fractions
on the fines fraction to disperse the bitumen. If will not be coated with bitumen and will remain un-
the fines include a significant amount of clay, they protected in the mix. Determining susceptibility to
will not be available to do so because they will be moisture change and rapid weathering is therefore
locked together in lumps. The gradings obtained advisable using one of the many different methods
from standard tests may show that there are for testing a materials durability.
plenty of fines present in a material but the grading
Material temperature. At the time of mixing with coated whereas low temperatures can result in
a bitumen stabilising agent, the temperature of little or no dispersion of the bitumen. Tempera-
the material must be sufficiently high as it plays ture measurements of the material are therefore
an important role in determining the quality of mix essential before laboratory or field production
achieved. commences.

> BSM-emulsion Mixing should not be attempted with material

Typically, materials with a temperature of 10C temperatures less than 10C. Where the tem-
or higher can be treated with bitumen emulsion perature of the material ranges between 10 and
without compromising the bitumen distribution 15C, mixes should only be produced with
in the mix. superior quality foamed bitumen (especially the
half-life). Where such conditions are expected,
> BSM-foam the quality of the mix should be checked in the
The temperature of the material has a significant laboratory at the anticipated mixing temperature
influence on the degree of dispersion and the before commencing construction.
properties of the mix. Higher material tempera-
tures increase the size of the particle that can be

134/135
4.3.5 Bitumen stabilising agents

Penetration grade bitumen is used to produce

both the foamed bitumen and bitumen emulsion to traffic. During the mix design phase, and on
that is used as a stabilising agent to manufacture site before full-scale application begins, the
BSMs. The types of bitumen and specific bitumen breaking rate should be tested with repre-
requirements are outlined below. sentative samples of material, active filler and
water, at realistic temperatures.

Compatibility of bitumen emulsion and the

parent material. The selection of the bitumen
emulsion type is influenced by the type of
material that is to be treated. Certain material
Bitumen Emulsion types are not suitable for treatment with ani-
Base bitumens with penetration values onic bitumen emulsions. These are the acidic
between 50 and 100 are generally selected rocks with silica contents above 65% and
for bitumen emulsion production, although alkali contents below 35% and include quartz-
softer and harder bitumen has been success- ite, granite, rhyolite, sandstone, syenite and
fully used. The selection of the correct grade felsites. Treatment of such materials requires a
or category of bitumen emulsion for each cationic bitumen emulsion, as outlined in the
application is essential, as outlined in the table table opposite.
below.
Manufacturers normally recommend that undi-
Slow set stable grade cationic bitumen emul- luted bitumen emulsion is heated to between
sions are almost exclusively used worldwide 50 and 60 C to prevent premature breaking
for BSMs as they generally work well with due to the increase in pressure and shearing
dense graded materials, regardless of the action while pumping and injecting through the
parent rock, as well as with material with high spraybar on the recycler.
fines contents. These bitumen emulsions
have long workability times to ensure good
dispersion and are formulated for mix stability.
(Anionic stable grade emulsion are used in a
few warm and dry climates.)

Breaking rate. There have been many recent

developments in bitumen emulsion technol-
ogy to improve stability without prolonging
the break time. These emulsions are typically
slower setting than the standard products, and
should be used on projects where the treated
layer can be cured for a period before opening
 Classes of Bitumen Emulsion
Bitumen Emulsion Type Anionic Cationic
Emulsifier type Fatty acid or resin acid Amine
Bitumen emulsion charge Negative Positive
pH High (alkali) Low (acid)
Grades Stable mix (slow set) for recycling/stabilising

Bitumen emulsion type/aggregate type compatibility

Trends
Emulsion Type Aggregate (Rock) Type
Breaking rate Adhesion
Anionic Acidic Slow Poor
Anionic Alkaline Medium Good
Cationic Acidic Fast Excellent
Cationic Alkaline Fast Good

136/137
Foamed Bitumen The water application rate and bitumen
Bitumen grades with penetration values temperature are the most important factors
between 60 and 200 are generally selected for influencing foam quality. A higher bitumen
BSM-foam, although harder bitumen has been temperature usually creates better foam.
successfully used in the past without compro- A sensitivity analysis in the laboratory is
mising the quality of the mix (harder bitumen recommended to identify a target bitumen
is generally avoided due to poor quality foam, temperature for foaming. (As with HMA pro-
leading to poorer dispersion of the bitumen in duction, temperature limits should be imple-
the mix). mented to prevent damage to the bitumen.)

The penetration value alone does not qualify The variability of the foam characteristics
bitumen for use in a BSM-foam. The foaming measured in a laboratory, both in terms of
characteristics of each bitumen type needs repeatability and reproducibility, are significant.
to be tested. Two properties form the basis To obtain an acceptable level of statistical
of a bitumens suitability for use, namely the reliability, at least three tests are recom-
Expansion Ratio (ER) and Half-life (1/2): mended for each set of conditions. In addition,
The expansion ratio is a measure of the potential variability in the bitumen composition
viscosity of the foam and determines how from the same source necessitates checking
well the bitumen will disperse in the mix. It is the foam characteristics of each tanker load of
calculated as the ratio of the maximum vol- bitumen.
ume of foam relative to the original volume
of bitumen.
The half-life is a measure of the stability of
the foam and provides an indication of the
rate of collapse of the foam. It is calculated
as the time taken in seconds for the foam to
collapse to half of its maximum volume.

One of the dominant factors influencing foam

characteristics is the amount of water that is
injected into the expansion chamber to cre-
ate the foam, the foamant water. Increasing
the application rate of water creates greater
expansion (higher ER) but leads to more rapid
subsidence or decay, a shorter half-life (1/2),
as illustrated in the graph opposite.
 15 15

14 14

13 Optimum foamant 13
water content
12 12

Half-life (seconds)
Expansion (times)

11 11

10 10
Minimum acceptable
9 9
Expansion Ratio
8 8

7 Minimum acceptable 7
Half-life
6 6
Equal Equal
5 5

4 4
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Water Added (%)

= Expansion
= Half life

Determining the Optimum Foamant Water Content

Foamed Bitumen Characteristics (Minimum Limits)

Aggregate Temperature 10 C to 15 C Greater than 15 C
Expansion Ratio, ER (times) 10 8
Half-life, 1/2 (seconds) 8 6

138 / 139
4.3.6 Active filler

For the purpose of this manual, the term ac- Various types of active filler can be used, sepa-
tive filler is used to define fillers that chemically rately or in combination. The type selected will
alter the mix properties. The types of active filler depend on availability, cost and efficacy with the
used with BSMs are: cement (various types, but actual component materials. Research has shown
not rapid hardening cements), lime, fly ash and that it is almost impossible to predict which active
slagment but excludes natural fillers such as rock filler will prove to be the most effective without ex-
flour. In addition, the term lime always refers to perimentation during mix design. (Testing 100 mm
hydrated lime in this manual. diameter specimens for Indirect Tensile Strength is
the most useful guide for active filler selection, as
The purpose of incorporating active filler in described in Appendix 1)
BSM is to:
When cement is used, the application rate must be
> Improve adhesion of the bitumen to the
limited to a maximum of 1% by mass of dry mate-
aggregate.
rial. When using hydrated lime, the application rate
> Improve dispersion of the bitumen in the mix. may be increased to 1.5% (or more) where the lime
is required to modify plasticity. However, it should
> Modify the plasticity of the natural materials be noted that above these application rates, the
(reduce PI). increase in mix stiffness is compromised signifi-
> Increase the stiffness of the mix and rate of cantly by a loss in flexibility of the material and the
strength gain. benefit of the bitumen is hardly realized.

> Accelerate curing of the compacted mix. Where active fillers are applied, the time delay
between mixing the active filler with the material
and application of the foamed bitumen or bitumen
BSM-emulsion emulsion should be reduced to a minimum (both in
the laboratory and the field). The active filler reac-
Control the breaking time.
tion begins immediately upon contact with moist
Improve the workability (in some cases).
material, promoting adhesion between the fine
particles. The longer the delay between premix-
ing with active filler and applying the bitumen, the
BSM-foam
lower the percentage of fines available for disper-
To assist in dispersing the bitumen splinters. sion of the bitumen in the BSM mix.

Note:

The maximum allowable cement addition for

BSMs is 1%
Where materials with unacceptably high PI bitumen or bitumen emulsion. Pre-treating with
values are encountered, they can be treated with lime must allow for sufficient time for modification
hydrated lime to modify the plasticity, thereby ren- to take place before bitumen treatment (normally 4
dering them acceptable for treatment with foamed hours is sufficient).

4.3.7 Water quality

The quality of the water used to create the foamed for concrete and other road materials should be
bitumen and to dilute a bitumen emulsion is impor- followed in this regard.
tant. The standard requirements of water quality

BSM-emulsion BSM-foam
The pH levels of the water must be checked, Although acceptable foam may be achieved
as must the compatibility of the bitumen emul- using water containing impurities, such prac-
sion and the water. tice should be avoided. Impurities often lead to
scales forming on the walls of the feed pipes
and these eventually dislodge and block the
Note: water injection jets, preventing the bitumen
from foaming.
When diluting bitumen emulsion, always add
the water to the bitumen emulsion to avoid a
premature break.

140/141
4.3.8 Mix design procedure

The mix design procedure involves several steps Additional mix designs using 150 mm diameter
and one or more series of tests, depending on specimens cured under a different regime may
the importance of the road and the magnitude of then be carried out if a higher level of confidence
design traffic. The mix design procedure always is required. The interval of added bitumen for such
starts by testing samples of the material that will mixes is normally less than that applied for the
be stabilised (standard laboratory tests) to deter- initial mix design, thereby allowing a more precise
mine whether they are suitable for treating with assessment of the optimum added bitumen
bitumen and, if not, the type of pre-treatment or content to be made. Other sensitivity tests can
blending required to make them suitable. also be undertaken using the same procedure for
specimen manufacture and curing (e.g. the effect
Once the suitability of the material has been prov- of reducing the amount of active filler to 0.75%).
en, the actual mix design procedure commences
with a series of preliminary tests to determine the Appendix 1 includes the detailed laboratory pro-
need for the addition of an active filler and whether cedures for carrying out a BSM mix design using
the material exhibits a preference for either ce- either bitumen emulsion or foamed bitumen as the
ment or hydrated lime. All further testing is then stabilising agent.
undertaken in accordance with these results (e.g.
if cement is shown to be the preferred active filler Where even higher levels of confidence are
then all mixes will include a nominal 1% cement). required (e.g. for the design of heavy-duty or
strategic pavements, such as runways at major air-
The mix design is then undertaken by mixing a ports), a series of triaxial tests may be carried out
series of samples, each with a different amount of on large specimens (150 mm diameter x 300 mm
bitumen added. The mixed materials are then used high) manufactured at the optimum added bitumen
to manufacture several 100 mm or 150 mm diam- content. The shear properties determined from
eter specimens which are cured dry and tested to such a testing programme feed directly into pave-
determine their respective indirect tensile strengths ment design models. These tests are considered
(ITS) under both unsoaked and soaked condi- specialised and are not covered in this manual;
tions. These results provide an indication of the they are adequately described in the Technical
optimum amount of bitumen that should be added Guideline TG2 (2009).
to achieve a specific level of strength.
 Obtain representative samples

Standard Laboratory Tests

- Sieve analysis (gradings)
- Atterberg Limits (plasticity)
- Moisture/density relationship

Blend sample components in the YES Sample preparation

required proportions Is blending required?
NO
Determine ICL value YES Is pre-treatment with lime
Pre-treat with lime required?
NO
Active Filler Requirements
ITS tests on 100 mm
specimens cured dry

Bitumen emulsion Bitumen stabilising agent Foamed bitumen

NO NO
YES BSM Mix Design YES Are the foaming properties (expan-
Is the material (aggregate)
ITS tests on 100 mm or 150 mm
compatible with the emulsion sion ratio & half-life) acceptable?
specimens, cured dry

NO Do the results meet the ITS

requirements?
YES
Is the level of confidence YES
Compile mix design report
satisfactory?
NO
Additional BSM Mix Design
ITS tests on 150 mm
specimens, cured at EMC

NO Do the results meet the ITS

requirements?
YES
Is the level of confidence YES
satisfactory?
NO
Advanced triaxial testing

The complete mix design procedure is explained in the flow chart above.

142/143
4.3.9 Classification of BSMs

The results obtained from the various mix design To simplify matters and to focus on the more
strength tests are used to judge the anticipated practical aspects of pavement design, two classes
performance of the BSM. Several different clas- of BSMs have been adopted:
sification systems have been developed for these
materials, as well as methods to relate the BSM to Class 1 BSM: 
a standard material (e.g. the equivalency method Materials with high shear strength. These materi-
adopted in California). Structural layer coefficients als are suitable for base layer construction on
derived from these strength tests have also been pavements with a structural capacity in excess
for developed for input into the AASHTO Structural of 3 million ESALs (where analytical modelling is
Number design method (see Section 4.3.12). recommended for pavement design).

To date, the most comprehensive research pro- Class 2 BSM: 

gramme on BSMs was carried out in South Africa Materials with moderate shear strength. These
between 2004 and 2009. Involving both laboratory materials are suitable for base layer construction
testing (including monotonic and repeated-load for pavements with a structural capacity less than
triaxial tests on large specimens) together with a 3 million ESALs (where the empirical methods
detailed long-term pavement performance (LTPP) (SN and PN methods) are adequate for pavement
exercise of 23 different pavements (including design).
several HVS trial sections), this work culminated
in the publication of Asphalt Academys Technical The following table summarises the requirements
Guidelines TG2 (2009). The design methodology for classifying a BSM into one of two classes.
emanating from this work uses the results from
strength tests (ITS and triaxial) to classify the BSM
in to one of three classes (BSM1 being a high
quality material with excellent shear properties
whilst a BSM3 has poor shear properties, suitable
only for lightly trafficked roads). Such classification
feeds directly into the empirical Pavement Number
pavement design method (described in Section
2.6.4).
 Recommended BSM classes based on ITS test results
Class 1 Class 2 Not suitable
RAP and GCS Material blends Poor gravels
Parent material:
RAP/GCS blend Natural gravel Plastic materials
Test result from Mix Design
RAP/GCS/gravel Marginal materials Soils

ITSDRY
Both 100 mm and 150 mm > 225 kPa 125 to 225 kPa < 125 kPa
specimens

ITSWET & ITSSOAK

Both 100 mm and 150 mm > 100 kPa 50 to 100 kPa < 50 kPa
specimens

ITSEQUIL
> 175 kPa 95 to 175 kPa < 95 kPa
150 mm specimens only

Implied shear properties

Cohesion > 250 kPa > 50 < 50

Angle of internal friction > 40 > 25 < 25

Implied parent material

properties

California Bearing Ratio (CBR) > 80% >20% < 20%

Plasticity Index (PI) < 10 < 15 > 15

144/145
4.3.10 Working with BSMs

Safety aspects for BSM-foam personnel receive appropriate training. The same
Bitumen temperatures need to be high (typically safety rules as those documented for hot mix
>160C) for the water reaction to produce an ac- asphalt, are applicable to foamed bitumen.
ceptable foamed. At such high temperatures, bitu-
men must be treated with respect and adequate Fluid Considerations
safety procedures established, similar to those The role of moisture in the stabilised material is
adopted for hot mixed asphalt production. This is similar for BSM-emulsion and BSM-foam in many
well known by asphalt manufacturers who work respects, but there are some differences. The en-
with hot bitumen on a daily basis, but the recycling tire fluid content in the mix (moisture and bitumen)
contractor who undertakes a foamed bitumen needs to be considered. The role of fluid in the two
project for the first time needs to ensure that his types of BSM is explained in the table below.

Role of fluids in BSM

Component BSM-emulsion BSM-foam
Negligible contribution to fluids for
Bitumen Contributes to fluids for compaction
compaction
Reduces absorption of bitumen emulsion Separates and suspends fines making them
water into aggregate available to bitumen during mixing
Acts as carrier for bitumen splinters d uring
Prevents premature breaking
mixing
Moisture in Extends curing time and reduces early
Reduces early strength
aggregate strength
Provides workability of BSM at ambient temperatures
Reduces friction angle and lubricates for compaction
Provides shelf-life for the mix

BSM-emulsion phase is referred to as the breaking time

(Also known as the setting or settling time.)
Changes in moisture content occur in two
distinct phases, namely:
The breaking process with anionic bitumen
Breaking is the separation of the bitumen
emulsions is a mechanical process (evapo-
from the water phase through flocculation
ration), whereas cationic bitumen emul-
and the coalescence of the bitumen droplets
sions produce a chemical break. For dense
to produce films of bitumen on the individual
mixtures, more time is needed to allow for
particles of the material. The rate at which
mixing and placement and slower breaking
the bitumen droplets separate from the water
times are required. As the bitumen emulsion
 breaks, the colour changes from dirty brown Temperature of material and air. The higher
to black. Although this can be observed with the temperature, the quicker the bitumen
the naked eye, it is recommended that a emulsion breaks and cures.
magnifying glass is used.

Curing is the displacement of water and the BSM-foam

resultant increase in stiffness and tensile
strength of the BSM. This is important as a The moisture content of the material reduces
mix needs to acquire sufficient stiffness and due to evaporation and repulsion by the bitu-
cohesion between particles before carrying men. This process is known as curing.
traffic.
As the moisture content reduces, the tensile
Some of the factors which influence the break- strength and stiffness of the material increas-
ing and curing process (known as the setting es. This is important because the completed
process of bitumen emulsions), include: layer of BSM-foam often needs to acquire
Rate of absorption of water by the material. sufficient stiffness and cohesion between the
Rough-textured and porous materials reduce particles before carrying heavy loads.
the breaking and setting time by absorbing
water contained in the bitumen emulsion.
Moisture content of the mix prior to mixing
influences breaking time.
Moisture content of the mix after compac-
tion influences curing rate.
Grading of the material and voids content of
the mix.
Type, grade and quantity of the bitumen
emulsion.
Mechanical forces caused by compaction
and traffic.
Mineral composition of the material. The
rate of cure may be affected by physico-
chemical interactions between the bitumen
emulsion and the surface of the individual
material particles.
Intensity of electrical charge on the material
particles in relation to that of the bitumen
emulsion.
Active filler addition, the amount of cement
or lime.

146/147
Mixing Moisture. The moisture content that will
BSM-foam
provide the best BSM mix is termed the optimum
mixing moisture content (OMMC). This is the Fluffpoint moisture (the moisture content that
moisture in the material plus, for BSM-emulsions, results in the maximum bulk volume of loose
any additional moisture in the bitumen emulsion. mineral aggregate during agitation) should be
OMMC varies with gradation of the material and, used as a target. This value ranges from 70 to
in particular, the size of the fraction smaller than 90% of the optimum moisture content (OMC)
0.075 mm. (determined from modified AASHTO compac-
tion).
BSM-emulsion
A minimum of 1 to 2% moisture is required The targeted mixing moisture content when
in the material prior to adding the bitumen adding foamed bitumen is 75% of OMC.
emulsion.

The water and bitumen in the bitumen emul- Bitumen Supply

sion act as lubricants in BSM-emulsion mixes. When coupling a new tanker load of bitumen sta-
The optimum moisture content (OMC) deter- bilising agent to the recycler, some basic checks
mined from modified AASHTO compaction should be conducted to ensure that the bitumen
should be used for the total mixing fluid con- is acceptable for use. Regardless of whether the
tent. This is explained in the equation below: tanker contains bitumen emulsion or Penetration-
OFC = OMCMOD U = FMC + EWC + RBC grade bitumen, the temperature of the contents
should be checked using a calibrated thermometer
Where (gauges fitted to tankers are notoriously unreli-
OFC = optimum fluids content (%) able).
OMCMOD-U = optimum moisture content using
Mod. AASHTO compaction on
BSM-emulsion
untreated material (%)
FMC = moisture content of Pressure gauge on the spraybar should be
aggregate (%) checked to ensure that the bitumen emulsion
EWC = bitumen emulsion water content is flowing freely and not creating excessive
including water used for dilution backpressure due to a blockage.
as percentage of dry aggregate
(%)
RBC = residual bitumen content as
percentage of dry aggregate (%)
 agitation energy within a relatively short time
BSM-foam
period to achieve a homogenous blend. This pro-
the foaming characteristics of each tanker cess is adequately described in Chapter 3 for both
load of bitumen must be checked using in-place and in-plant treatment.
the test nozzle on the recycler.
Compaction
The quality of foam is a function of bitumen
Special attention needs to be paid to compaction,
operating pressure. The higher the pressure,
as it improves particle contacts and reduces voids.
the more the stream of bitumen will tend to
The density achieved is critical to the ultimate
atomise as it passes through the jet into
performance of the mix.
the expansion chamber, thereby promoting
uniformity of foam. If the bitumen were to
enter the expansion chamber as a stream (as BSM-emulsion
it does under low pressures) the water would
impact on only one side of the stream, creat- The inclusion of bitumen emulsion typically
ing foam, but the other side would remain as improves compactibility of the mix.
unfoamed hot bitumen. It is therefore impera-
tive to set the recycler up and to operate at an
advance speed that will maintain a minimum BSM-foam
operating pressure above 3 bars.
Compaction promotes adhesion of the
bitumen mastic to the coarse particles.
Bitumen delivered to site by tankers that are fitted
with fire-heated flues is sometimes contami-
nated with small pieces of carbon that form on
the sides of the flues whilst heating. Draining the
last few tons from the tanker tends to draw these
unwanted particles into the recyclers system Vibrating padfoot rollers are normally used to com-
and can cause blockages. This problem is easily pact layers of BSM in the field. Such rollers impart
resolved by ensuring the effectiveness of the filter very high energy and therefore, in order to achieve
in the delivery line. Any unusual increase in pres- a specific level of density, require a lower effective
sure will indicate that the filter requires cleaning, a optimum fluids content relative to that determined
procedure that should anyway be undertaken on a in the laboratory. For this reason it is normally pos-
regular basis (e.g. at the end of every shift). sible to compact in the field at the (relatively low)
mixing moisture content. (Roller selection, in terms
Applying Active Filler of type and capacity, is described in the Wirtgen
See Section 4.2.6, Uniformity of application. Cold Recycling Application manual.)

Mixing As with cement stabilised materials, compaction of

Mixing the BSM involves the addition of all addi- BSMs should always aim to achieve the maximum
tives to the parent material and providing sufficient density possible under the conditions prevailing on

148/149
site (the so-called refusal density). A minimum
BSM-emulsion
density is usually specified as a percentage of the
modified AASHTO density, normally between 98% Chemistry plays a significant role in the curing
and 102% for bitumen stabilised bases. A density of BSM-emulsions. Water is an intrinsic com-
gradient is sometimes permitted by specifying an ponent of bitumen emulsions. Breaking of the
average density. This means that the density at bitumen emulsion needs to take place before
the top of the layer may be higher than at the bot- curing as a result of migration and moisture
tom. Where specified, it is normal also to include loss by evaporation.
a maximum deviation of 2% for the density meas-
BSM-emulsion usually requires longer curing
ured in the lowest one-third thickness of the layer.
times than BSM-foam because of the higher
moisture contents.
Hence, if the average density specified is 100%,
then the density at the bottom of the layer must be
more than 98%.

Curing BSM-foam
Curing of BSMs is the process where the mixed
and compacted layer looses moisture through Curing takes place as a result of migration of
evaporation, particle charge repulsion and pore- water during compaction and continues with
pressure induced flow paths. repulsion of moisture by the bitumen and loss
by evaporation.
The reduction in moisture content leads to an
increase in the tensile and compressive strength,
as well as stiffness of the material.

The rate of moisture loss from newly constructed

BSM layers plays a significant role in determin-
ing the performance of the layer. It is in the early
period of repeated loading that the majority of the
permanent deformation takes place in BSM layers.
Where a new BSM layer is to be trafficked immedi-
ately after construction, it is important to keep the
compaction moisture content to a minimum. The
lower the degree of saturation (moisture content)
of the BSM, the greater the resistance to perma-
nent deformation.
 BSM-foam stiffness

ss
fne
n stif
io
uls
M-em
BS

Moisture content

1 3
Time (years)

Concept of Curing and Influence on Mix Stiffness

Curing rates are dependent on the type of treat- Although a BSM should have sufficient stiffness
ment (BSM-emulsion or BSM-foam). Although the and strength to withstand moderate levels of early
use of active filler also has an impact on curing, traffic, the layer will continue to gain strength over
its inclusion in a BSM does not justify extensions several years (i.e. improve its resistance to perma-
in the curing time as cementation is not one of the nent deformation).
desired properties of these materials.

150 / 151
4.3.11 Mechanical tests

Indirect Tensile Strength (ITS) The specimens are cured for 72 hours at 40 C to
The ITS test is an indirect measure of the tensile reach a constant mass (see Appendix 1). ITSDRY
strength and reflects the flexibility and flexural values are determined from these specimens.
characteristics of the BSM. Although this test does The results obtained after soaking specimens for
not produce highly repeatable results, it is the 24 hours at 25 C are termed ITSWET. The ratio of
most economical method available for investigat- ITSWET and ITSDRY, expressed as a percentage, is
ing the effectiveness of the bitumen in a BSM. the Tensile Strength Retained (TSR). An example
In addition, a background of historical data is of the typical trends of ITS test results is shown in
available. the figure below, indicating how the actual results
are analysed against the limits for a specific BSM
100 mm or 150 mm diameter specimens are used classification.
to indicate the optimum bitumen content, the need
for an active filler and, if active filler is required, at 150 mm diameter and 95 mm high specimens
what content. can also be cured to simulate the field moisture
conditions (see Appendix 1). The ITS results from

ITSdry

Min ITSdry
t
ITS
we
ITS

Min ITSwet

Min BC

BSM Binder Content

ITS interpretation for bitumen type and content (Level 1)

specimens tested after this curing are termed > 100% RAP material. A combination of TSR
ITSEQUIL. The results after soaking for 24 hours at <50% and ITSDRY > 500 kPa indicates that the
25 C are termed ITSSOAK. treated material is partly stabilised and partly
asphaltic (continuously bound). In this situa-
The limits for interpreting the various ITS tests are tion, the material probably requires blending
shown in the table in section 4.3.9. with crusher dust to ensure that the stabilisation
process dominates the mix (see Chapter 6).
Note: The TSR value is useful for identifying
problem materials. If the TSR is less than 50%, it is In addition, tests using 150 mm diameter speci-
recommended that active filler be included in the mens should be undertaken to verify the relevant
mix. If such treatment does not increase the ITSWET ITSEQUIL values for BSM classification.
value with a maximum application of 1% lime
or cement, then the nature of the material being
stabilised should be investigated:

> Granular material. A combination of TSR < 50%

and ITSDRY > 400 kPa suggests contamination
(normally attributed to clay or deleterious ma-
terials). In this situation, it is suggested that the
material is pre-treated with hydrated lime and
the tests repeated.

152/153
4.3.12 Pavement design approaches for BSMs

There are three accepted structural design meth- on the accuracy of the input parameters. With
ods for BSMs and these are discussed below. Structural Numbers, the primary input that affects
the result is the selection of an appropriate struc-
Structural Number design method tural layer coefficient for each layer of material.
The AASHTO (1993) design method using These coefficients reflect the engineering proper-
Structural Numbers was described in Chapter ties of the in situ material in the layer.
2, Section 2.6.3. This design method is popular
worldwide and, due largely to its simplicity and The table below includes structural layer coef-
user-friendliness, it is used extensively for the ficients for layers that are constructed using
design of all types of pavements. conventional materials.

As with all design routines, the reliability of the

results obtained using this method is dependent

Typical structural layer coefficients (from AASHTO)

Structural layer coefficient
Material type Characteristic
(per inch)
Elastic modulus
Asphalt surfacing 0.30 to 0.44
2,500 to > 10,000 MPa
Asphalt base Continiously graded (6% voids) 0.20 to 0.38
Graded crushed stone CBR > 80% 0.14
Natural gravel, type 1 CBR 65 to 80% 0.12
Natural gravel, type 2 CBR 40 to 65% 0.10
Soil, type 1 CBR 15 to 40% 0.08
Soil, type 2 CBR 7 to 15% 0.06
Cohesionless sand PI = 0 0.04 to 0.05
Cement-treated crushed stone 1.0 < UCS < 3.0 MPa 0.17
Cement-treated gravel UCS < 1.0 MPa 0.12
 Since bitumen stabilisation is a relatively new
Note:
technology, no relevant structural layer coefficients
The AASHTO design method originated are available for these materials in the AASHTO
from America. Structural layer coefficients design method. The following table shows how the
are normally quoted in terms of per inch of structural layer coefficient of a BSM can be esti-
layer thickness. The various values assigned mated from ITS test results. In addition, this chart
to different materials and the magnitude of can be used as an indicator of the shear properties
the coefficient in per inch terms is well of a BSM as well as gaining a rough estimate of
known, worldwide. It is therefore preferable the anticipated structural layer coefficient based
for countries using the metric system to on the CBR value of the untreated material.
retain the coefficients in per inch terms and
convert the layer thickness from centimetres
to inches (1 inch = 2.54 cm).

Suggested structural layer coefficients

for bitumen stabilised material (BSM)
Structural layer coefficient (per inch) max
0.18 0.23 0.28 0.35
Indirect tensile strength (ITS) after stabilisation
100/150 mm specimens

ITSDRY (kPa) 125 175 225

ITSWET & ITSSOAK (kPa) 50 75 100

150 mm specimens

ITSEQUIL (kPa) 95 135 175

Indicated shear properties

Cohesion (kPa) 50 100 250
Angle of Friction () 25 30 40
Material CBR value before stabilisation (at field density)
(Materials with CBR < 20% not
recommended)
20 40 80

Anticipated application rate of bitumen for stabilisation (% by mass)

2.54.0 2.03.0 1.82.3

154/155
Pavement Number design method their respective Effective Long Term Stiffness (ELTS)
The PN design method was developed as part of values, as described in Section 2.6.4. Some rules
TG2 (2009) guidelines for the design of BSMs. relating to the pavement layers are:
> The load spreading potential of an individual
The PN design method is a knowledge-based (or
layer is a product of its thickness and its ELTS
heuristic) approach and can be used for design
value under loading.
with a reliability of 90% to 95%. As was described
in Chapter 2, Section 2.6.4, the PN method is simi- > The ELTS value of a layer depends on the mate-
lar to the well known Structural Number method rial type and class and on its location in the
(AASHTO 1993) and has the following advantages: pavement structure.
> Data from numerous in-service pavements were > Fine-grained subgrade materials act in a stress-
used to develop the method. The type and softening manner. For these materials, the ELTS
detail of the data suggests the use of a relatively is determined mainly by the material quality
simple method and precludes the use of a and by the climatic region. Owing to the stress
Mechanistic-Empirical design method. softening behaviour, subgrade materials will
> The method gives a good fit to the available field generally soften with decreased cover thickness.
data. > Coarse-grained, unbound materials act in a
> The method is robust, and cannot easily be stress-stiffening manner. For these materials, the
ELTS is determined mainly by the material qual-
manipulated to produce inappropriate designs
ity and the relative stiffness of the supporting
layer. The ELTS of these materials will increase
The PN design method has been verified for
with increasing support stiffness, by means of
design traffic of up to 30 million ESALs based on
the modular ratio limit, up to a maximum stiff-
long-term pavement performance (LTPP) data from
ness which is determined by the material quality.
pavements with BSM base layers. The pavements
being monitored continue to be trafficked and the > BSMs are assumed to act in a similar way to
maximum design traffic will increase with time. coarse granular materials, but with a higher cohe-
Pavements modelled using this method require a sive strength. The cohesive strength is subject to
minimum subgrade CBR value of 3%. In addition, breakdown during loading and thus some soften-
the LTPP exercise included roads with only an ing over time can occur. The rate of softening is
80 kN legal axle limit. The method has therefore mainly determined by the stiffness of the support,
not been calibrated for higher axle loads (e.g. the which determines the degree of shear in the layer.
13 ton legal limit in Greece). However, owing to the higher cohesive strength
in BSMs, these layers are less sensitive to the
The PN design method uses a materials classifica- support stiffness than unbound granular materials
tion system for each layer in the pavement structure and can therefore sustain higher modular ratio
and the design approach is, in effect, an intelligent limits. (If, however, the cement content of a BSM
Structural Number method. The Pavement Number mix exceeds 1 %, the material is assumed to
is the sum of the products of layer thicknesses and behave as a cemented material.)
The above rules-of-thumb introduce several ing) behaviour. These need to be understood in
concepts such as the ELTS, modular ratio limit, order to use the method. (Refer TG2 (2009) for
maximum stiffness and stress-stiffening (or soften- more details.)

Deviator Stress Ratio Design Method be adopted. The Deviator Stress Ratio Method is
Pavement rehabilitation projects that are de- based on the premise, proven through research,
signed to carry > 30 million ESALs design traffic, that the rate of permanent deformation in a BSM
require more advanced structural analysis than the layer is a function of the ratio of the applied devia-
Pavement Number Design Method alone. In such tor stress relative to the maximum deviator stress
cases the Deviator Stress Ratio Method should of the BSM (at failure).

This is shown in the equation below.

1 3
Deviator Stress Ratio =
1.f 3

Where 1 = major principal stress applied to BSM layer from M-E analysis
3 = minor principal stress applied to BSM layer from M-E analysis
1,f = major principal stress at failure from triaxial testing of BSM

In order to carry out a Mechanistic-Empirical (M-E) triaxial testing and pavement monitoring, the fol-
analysis of the pavement structure, a Resilient lowing values are considered to be reasonable for
Modulus value for the BSM is required. Based on this purpose.

Resilient Moduli ranges for BSMs

Material type Bitumen Content of BSM (%) Resilient Modulus MR(MPa)
100% RAP 1.6 to 2.0 1,000 to 2,000
RAP/crushed stone
1.8 to 2.5 800 to 1,500
(50:50 blend)
Graded crushed stone 2.0 to 3.0 600 to 1,200
Natural gravel (PI < 10, CBR>45) 2.2 to 3.5 400 to 800
Natural gravel (PI < 10, CBR>25) 2.5 to 4.0 300 to 600

The preferred way to approach advanced pave- analysed and the design traffic. The following
ment design is to select a deviator stress ratio limit figure illustrates how the pavement life decreases
depending on the importance of the road being as the deviator stress ratio increases.

156/157
 10

1
Permanent Strain (%)

0.1

0.01

0.001

1.00E+04 1.00E+05 1.00E+06 1.00E+07 1.00E+08 1.00E+09

Number of Repetitions N

Deviator stress ratio = 45%

Deviator stress ratio = 40%
Deviator stress ratio = 30%

Recommended deviator stress ratio limits, based count of typical recycling depths, are provided in
on the relationships outlined above and taking ac- the table below.

Deviator Stress Ratio Limit for 10 mm maximum deformation in BSM layer

Design Reliability 95% Design Reliability 80 90%
(Highways with heavy traffic) (Arterials with moderate traffic)
35% 40%
It is recognised that BSMs have stress-dependent tor stress ratio can be used to establish whether
properties. Where dynamic triaxial testing has or not the BSM layer is the critical layer in the
provided the data to develop relationships be- pavement (i.e. the layer that reaches a terminal
tween Resilient Modulus and applied stress for a condition first and precipitates failure in the other
particular BSM, then iterative multi-layer stress- layers).
dependent analyses can be followed. This requires
the BSM layer to be subdivided into 25 mm to The deviator stress ratio approach provides
50 mm sub-layers and each one is then analysed reasonable designs with sufficient conservatism.
for convergence of Resilient Modulus. Such analy- It does, however, require sound pavement engi-
ses will yield more realistic stress distributions in neering knowledge and should only be used by
the BSM layer that are, in turn, used to determine experienced practitioners.
the deviator stress ratio. The calculated devia-

158 / 159
4.4 Summary: Advantages and disadvantages
of cement and bitumen stabilising agents

Cement Stabilisation
Advantages Disadvantages
Availability. Cement can be obtained worldwide, Shrinkage cracking is unavoidable.
always in bags, often in bulk. However, it can be minimised.
Cost. Relative to bitumen, cement is Increases rigidity in flexible pavements.
inexpensive.
R
 equires proper curing and protection from
Ease of application. Cement can always be early trafficking, particularly heavy slow-moving
spread by hand in the absence of bulk spreaders vehicles.
or slurry units.
Acceptance. Cement is well-known in the con-
struction industry. Standard test methods and
specifications are usually available.
Significant improvement of compressive strength
and durability properties of most materials.
 Stabilising with Bitumen (Emulsion and Foamed)
Advantages Disadvantages
Flexibility. Stabilising with bitumen creates Cost. Bitumen is relatively expensive.
a visco-elasto-plastic type of material with
Bitumen emulsions are not normally manu-
improved shear properties, (cohesion and
factured on site. The manufacturing process
resistance to deformation).
requires strict quality control. Emulsifiers are
Ease of application. A bulk tanker is coupled to expensive. Transport costs inflated by haul-
the recycler and the bitumen is injected through ing the water component, not only bitumen.
a spraybar (special type of spraybar for foamed (Foamed bitumen uses standard penetration-
bitumen). grade bitumen. There are no additional manufac-
turing costs and is therefore less expensive than
Acceptance. Bitumen emulsions are rela-
bitumen emulsion.)
tively well-known in the construction industry.
Standard test methods and specifications are Foamed bitumen demands that the bitumen is
available. hot, usually above 160 C. This often requires
special heating facilities and additional safety
Rate of gain of strength. Material can be traf-
precautions.
ficked soon after placing and compaction,
especially with BSM-foam BSM-foam requires strict adherence to grading
requirements, especially the fraction < 0.075 mm.
Durability. BSMs tend to lock-up the finer
(BSM-emulsion is more forgiving in this respect.)
particles by encasulating them in bitumen. This
prevents them from reacting to water and from W
 here the moisture content of the material in the
any pumping potential. existing pavement is close to OMC, saturation
often occurs when emulsion is added.
Curing can take a long time for BSM-emulsion.
Strength development is dictated by moisture
loss.
Availability. The required formulation for an
emulsion that is appropriate for a specific recy-
cling application may not always be available.

160/161
5 Recycling Solutions

5.1 Guidelines for recycling different pavements 165

5.1.1 Lightly trafficked roads (structural capacity: 0.3 million ESALs) 166
5.1.2 Low volume roads (structural capacity: 1 million ESALs) 168
5.1.3 Secondary rural roads (structural capacity: 3 million ESALs) 170
5.1.4 Main rural roads (structural capacity: 10 million ESALs) 172
5.1.5 Interurban highways (structural capacity: 30 million ESALs) 174
5.1.6 Major multi-lane highways (structural capacity:
100 million ESALs) 176
5.2 Alternatives for pavement rehabilitation 178
5.2.1 Existing pavement 180
5.2.2 Rehabilitation requirements 181
5.2.3 Rehabilitation options 182
5.2.4 Maintenance requirements 190
5.2.5 Construction & maintenance costs 192
5.2.6 Energy consumption 195
5.2.7 Relevant comments 199

162/163
All pavement rehabilitation solutions are project This is followed by an example showing differ-
specific. As was described in Chapter 2, each ent solutions for rehabilitating a specific road,
construction site is different; the depth of recycling two of which require the existing pavement to be
and type of stabilisation that is appropriate for a recycled. This exercise includes analyses of the
specific road is dictated by: costs involved in the initial rehabilitation exercise,
maintenance requirements during the service life
> the anticipated traffic over the design period
and the cost of rehabilitating the pavement at the
(the structural capacity requirement),
end of the service life (whole-of-life cost analysis).
> the composition of the existing pavement struc- Also included in this exercise is a section on the
ture, energy consumed by the various construction and
maintenance activities during the entire service
> the materials in the various layers, particularly life. Combining the whole-of-life costs with energy
those at the top of the pavement, consumed is useful for decision making.
> the number of layers overlying the subgrade
(the total cover thickness), and
> the strength of the underlying subgrade.

This chapter provides guidelines to assist design

engineers visualising the type of pavement
structures that can be achieved by recycling.
Typical solutions for different traffic categories are
illustrated, ranging from low volume roads (< 1 mil-
lion ESALs) to heavily trafficked highways (> 100
million ESALs), each with a selection of realistic
conditions in the existing pavement that are in
need of rehabilitation.
5.1 Guidelines for recycling different pavements

The guidelines presented in this section are For traffic categories below 30 million ESALs,
focused on recycling. Six different traffic catego- three different existing pavement scenarios are
ries are shown: considered, essentially dictated by subgrade
strength, depth of cover and the quality of material
> light pavements with a structural capacity of
in the existing layers. The three different subgrade
approximately 300,000 ESLAs that would nor-
conditions assumed are:
mally be applicable to rural access roads;
> Good with an in situ CBR value between
> low volume roads with a structural capacity of
10% and 25%,
1 million ESALs, typical for farm-to-market
roads; > Fair with an in situ CBR value between
3% and 10%. And
> secondary roads with a structural capacity of
3 million ESALs; > Poor with a CBR value of 3% or less.
> main rural roads with a structural capacity of
Each scenario has been carefully selected to be
10 million ESALs;
relevant for the type of pavement that is to be
> interurban highways with a structural capacity rehabilitated and/or upgraded. The same realistic
of 30 million ESALs; and approach has been adopted for traffic categories
in excess of 30 million ESALs; existing pavements
> major multi-lane highways with a structural for such heavy traffic are likely to be reasonably
capacity of 100 million ESALs. substantial with sufficient cover.

Three different recycling applications are

considered for each category:
> recycling the existing pavement without the
ddition of stabilising agents (mechanical
a
modification or reworking),
> recycling with a cementitious stabilising agent
(a minimum thickness of 150 mm has been
adopted for such layers), and
> recycling with a bitumen stabilising agent
(the minimum thickness for such layers when
recycled in-place is 100 mm).

164/165
5.1.1 Lightly trafficked roads (structural capacity: 0.3 million ESALs)

Existing roads falling into this category are normal- material with a CBR in excess of 25%. (Should
ly gravel roads or roads sealed with a light surface this not be the case, then the road needs to be
treatment. As a minimum, the existing road will constructed; it cannot simply be recycled because
include a layer of suitable gravel wearing course there is no structural layer to recycle.)

Mix designs are required to ensure that sufficient

strength is achieved by adding stabilising agents. Existing Pavement Recycle /
The minimum requirements for these lightly traf-
Thick CBR
ficked pavements are: mm %
> Cement stabilisation (denoted as CTB in the
Good support

chart): UCS > 1 MPa 2nd: Import 125 mm

(new layer)
> Bitumen stabilisation (Denoted as BSM in the
chart): ITSDRY > 125 kPa, ITSWET > 50 kPa 1st: Rework 100 mm
150 30

10

Chip seal
Fair support

Bitumen stabilised material (BSM) 2nd: Import 150 mm

(new layer)
Cement stabilised material (CTB)
1st: Rework 100 mm
CBR > 100 New GCS 150 30
CBR > 80 GCS (graded crushed stone)
45 < CBR < 80 (coarse gravel) 7

25 < CBR < 45 (natural gravel)

15 < CBR < 25 (gravel / soil)
10 < CBR < 15 (sandy silty soil)
Poor support

2nd: Import 150 mm

7 < CBR < 10 (silty soil) (new layer)
3 < CBR < 7 (clayey silty soil) 1st: Import 100 mm
Recycle 125 mm
Rules 150 30
1. Recycling with cement:
minimum thickness: 150 mm 3
2. Recycling with bitumen:
minimum thickness: 100 mm
Mechanical modify Recycle with cement or lime Recycle with bitumen
Thick CBR Thick CBR Thick CBR
mm % mm % mm %

125 50

100 Recycle 100 mm

30 Recycle 150 mm 150 CTB with bitumen 100 BSM
50 with cement 50 30
10 10 10

150 50 Import 125 mm

Recycle 150 mm 150 CTB Import 50 mm 100 BSM
with cement Recycle 100 mm
100 30 with bitumen
125 30 100 30
50
7 7 7

150 50
Import 175 mm
125 Recycle 200 mm 200 CTB Import 100 mm 125 BSM
30 with cement Recycle 125 mm
125 with bitumen 125 30
125 30

3 3 3

Chip seals (normally single seals) are an appropriate surface treatment for these pavements

166/167
5.1.2 Low volume roads (structural capacity: 1 million ESALs)

Existing low volume roads will normally include

a gravel wearing course with a surface treatment
(chip seal) overlying a relatively light pavement Existing Pavement Recycle
structure constructed from natural gravels. Thick CBR
mm %

Good support
Import 75 mm
Recycle 125 mm
150 30
Mix designs are required to ensure that sufficient
150 10
strength is achieved by adding stabilising agents.
The minimum requirements for these low volume 10
pavements are:
> Cement stabilisation (denoted as CTB in the Import 100 mm
chart): UCS > 1 MPa Recycle 150 mm
Fair support

> Bitumen stabilisation (Denoted as BSM in the 150 30

chart): ITSDRY > 175 kPa, ITSWET > 75 kPa
150 10

2nd: Import 150 mm

(new layer)
Poor support

Chip seal
150 30 1st: Rework 150 mm
Bitumen stabilised material (BSM)
Cement stabilised material (CTB) 150 10
CBR > 100 New GCS
3
CBR > 80 GCS (graded crushed stone)
45 < CBR < 80 (coarse gravel)
Two-part
25 < CBR < 45 (natural gravel) Existing Pavement
Recycle in
15 < CBR < 25 (gravel / soil)
Thick CBR Step 1
10 < CBR < 15 (sandy silty soil) mm %
Poor support

7 < CBR < 10 (silty soil)

3 < CBR < 7 (clayey silty soil) 150 30 Recycle 200 mm
with lime
Rules
150 10
1. Recycling with cement:
minimum thickness: 150 mm
3
2. Recycling with bitumen:
minimum thickness: 100 mm
/Mechanical modify Recycle with cement or lime Recycle with bitumen
Thick CBR Thick CBR Thick CBR
mm % mm % mm %
125 50
Recycle 150 mm Recycle 100 mm
150 CTB with bitumen 100 BSM
100 30 with cement

150 10 150 10 150 10

10 10 10

150 50
Recycle 150 mm Recycle 100 mm
150 CTB with bitumen 100 BSM
100 30 with cement

150 10 150 10 150 10

7 7 7

150 50

150 30 Recycle 250 mm Recycle 150 mm 150 BSM

with cement 250 CTB with bitumen
150 10 50 10 150 10

3 3 3

recycling: Lime to modify plasticity and bitumen for strength

place with lime Re-recycle with bitumen
Thick CBR Step 2 Thick CBR
mm % mm %

Recycle 100 mm
with bitumen 100 BSM
200 CTB
100 CTB
100 10 100 10
3 3

Chip seals (normally double seals) are an appropriate surface treatment for these pavements

168/169
5.1.3 Secondary rural roads (structural capacity: 3 million ESALs)

Existing roads falling into this category can be invariable have been constructed using a reason-
expected to have a formal pavement structure ably good quality material in the base layer, either
with a light surfacing (either a thin asphalt layer a crushed stone or a natural gravel with a CBR
or competent chip seal). These pavements would value in excess of 80%.

Mix designs are required to ensure that sufficient

strength is achieved by adding stabilising agents. Existing Pavement Recycle
The minimum requirements for these secondary
Thick CBR
road pavements are: mm %
> Cement stabilisation (denoted as CTB in the
Good support

chart): UCS > 2 MPa Import 100 mm

Recycle 125 mm
> Bitumen stabilisation (Denoted as BSM in the
chart): ITSDRY > 225 kPa, ITSWET > 100 kPa, ITSE- 150 80
QUIL > 175 kPa
150 50

20

2nd: Import 150 mm

Chip seal (new layer)
Fair support

Bitumen stabilised material (BSM) 1st: Rework 100 mm

150 80
Cement stabilised material (CTB)
CBR > 100 New GCS 150 50
CBR > 80 GCS (graded crushed stone) 7
45 < CBR < 80 (coarse gravel)
25 < CBR < 45 (natural gravel) 2nd: Import 150 mm
(new layer)
15 < CBR < 25 (gravel / soil)
1st: Import 50 mm
Poor support

10 < CBR < 15 (sandy silty soil) Recycle 200 mm

7 < CBR < 10 (silty soil) 150 80

3 < CBR < 7 (clayey silty soil) 150 50

Rules
150 10
1. Recycling with cement:
minimum thickness: 150 mm
3
2. Recycling with bitumen:
minimum thickness: 100 mm
/Mechanical modify Recycle with cement or lime Recycle with bitumen
Thick CBR Thick CBR Thick CBR
mm % mm % mm %

125 100
Recycle 250 mm Recycle 125 mm
with cement with bitumen 125 BSM
125 80 250 CTB
150 50 150 50

20 20 20

150 100
Import 75 mm
100 Recycle 275 mm 275 CTB Recycle 150 mm
80 with cement with bitumen 150 BSM
50
150 50 100 50 150 50

7 7 7

150 100

Import 100 mm
200 80 Recycle 300 mm 300 CTB Recycle 200 mm
with cement 200 BSM
with bitumen
150 50 100 50 100 50

150 10 150 10 150 10

3 3 3

A competent chip seal (Cape Seal or triple seal) is an appropriate surface treatment for these pavements although the
application of a thin asphalt (30 mm) is often preferred, especially in wet and/or cold regions

170/171
5.1.4 Main rural roads (structural capacity: 10 million ESALs)

Existing roads falling into this category would pavements would invariable have been construct-
normally have been constructed with a reasonable ed using a crushed stone or good quality natural
pavement structure, surfaced with either an thin gravel (CBR > 80%) but may include a layer of
asphalt surfacing or multiple chip seals. These cement stabilised material.

Mix designs are required to ensure that sufficient

strength is achieved by adding stabilising agents. Existing Pavement Recycle
The minimum requirements for these main road
Thick CBR
pavements are: mm %
> Cement stabilisation (denoted as CTB in the
Good support

chart): UCS > 2 MPa 2nd: Import 125 mm

40 AC (new layer)
> Bitumen stabilisation (Denoted as BSM in the 1st: Rework 125 mm
chart): ITSDRY > 225 kPa, ITSWET > 100 kPa, 150 80
ITSEQUIL > 175 kPa, cohesion > 250 kPa, angle of
150 50
friction > 40
20

2nd: Import 150 mm

Asphalt (new layer)
Fair support

40 AC 1st: Import 60 mm
Bitumen stabilised material (BSM) Recycle 125 mm
Cement stabilised material (CTB) 150 80
CBR > 100 New GCS 150 50
CBR > 80 GCS (graded crushed stone)
7
45 < CBR < 80 (coarse gravel)
25 < CBR < 45 (natural gravel) 2nd: Import 150 mm
(new layer)
15 < CBR < 25 (gravel / soil)
1st: Import 185 mm
Poor support

10 < CBR < 15 (sandy silty soil) 40 AC Recycle 250 mm

7 < CBR < 10 (silty soil)
150 80
3 < CBR < 7 (clayey silty soil)
150 50
Rules
1. Recycling with cement: 150 10
minimum thickness: 150 mm
2. Recycling with bitumen: 3
minimum thickness: 100 mm
/Mechanical modify Recycle with cement or lime Recycle with bitumen
Thick CBR Thick CBR Thick CBR
mm % mm % mm %
40 HMA 40 HMA
2nd: Import 125 mm 125 100
125 100 New layer 40 HMA
1st: Recycle 250 mm Recycle 125 mm
125 80 with bitumen 125 BSM
with cement 250 CTB
150 50 90 50 150 50

20 20 20
40 HMA
40 HMA
150 100 2nd: Import 150 mm
New layer 150 100 40 HMA
125
80 Recycle 200 mm
125 1 : Recycle 275 mm
st
275 CTB with bitumen 200 BSM
with cement
150 50 140 50

7 7 7
40 HMA
150 100
40 HMA
2nd: Import 150 mm 40 HMA
250 New layer 150 100
80 Import 50 mm
Recycle 250 mm 250 BSM
125 1st: Recycle 300 mm 300 CTB with bitumen
with cement
150 50 140 50

150 10 150 10 150 10

3 3 3

A 40 mm thick asphalt surfacing is appropriate for these pavements although, in dry regions, a competent surface
treatment (e.g. Cape Seal) is often preferred

172/173
5.1.5 Interurban highways (structural capacity: 30 million ESALs)

Existing roads falling into this category would asphalt base layer and surfacing. A good quality
normally have been constructed with a deep crushed stone or good natural gravel (CBR > 80%)
pavement structure and would usually include an would invariably lie beneath the asphalt layers, or a

Mix designs are required to ensure that sufficient

strength is achieved by adding stabilising agents. Existing Pavement Recycle /
The minimum requirements for these main road
Thick CBR
pavements are: mm %
> Cement stabilisation (denoted as CTB in the
chart): UCS > 2 MPa
> Bitumen stabilisation (Denoted as BSM in the
chart): ITSDRY > 225 kPa, ITSWET > 100 kPa, Pulverise
Good support

ITSEQUIL > 175 kPa, cohesion > 250 kPa, 100 AC and
compact
angle of friction > 40 125 mm
150 80

Asphalt 150 50

Bitumen stabilised material (BSM)

Cement stabilised material (CTB) 20
CBR > 100 New GCS
CBR > 80 GCS (graded crushed stone)
2nd: Import
45 < CBR < 80 (coarse gravel) 150 mm
25 < CBR < 45 (natural gravel) New layer
1st: Pulver-
15 < CBR < 25 (gravel / soil)
Fair support

100 AC ise and

10 < CBR < 15 (sandy silty soil) compact
125 mm
7 < CBR < 10 (silty soil) 150 80

3 < CBR < 7 (clayey silty soil)

150 50
Rules
1. Recycling with cement:
minimum thickness: 150 mm 7
2. Recycling with bitumen:
minimum thickness: 100 mm
 layer of cement stabilised material. a layer of crushed stone material on top of the
Two different options are shown for the pavement recycled/cement stabilised layer, the other with
that is recycled with cement or lime, one with asphalt.

Mechanical Recycle with cement or lime

modify Recycle with bitumen
Crushed stone overlay HMA overlay
Thick CBR Thick CBR Thick CBR Thick CBR
mm % mm % mm % mm %
50 HMA
2nd: Import
90 HMA 125 mm 125 100 80 HMA
(new layer)
1 :st Recycle 50 HMA
125 100 Recycle 250 mm
Recycle
275 CTB 275 mm 275 CTB with
275 mm
with with bitumen 250 BSM
125 80 cement
cement

150 50 125 50 125 50 150 50

20 20 20 20

90 HMA 50 HMA
2nd: Import 50 HMA
150 100 150 mm 150 100 Import
New layer 90 HMA 150 mm
Recycle
275 mm 275 BSM
125 100 1st: Recycle
Recycle with
300 mm bitumen
300 mm 300 CTB 300 CTB
with
125 80 with
cement cement 125 80

150 50 100 50 100 50 150 50

7 7 7 7

These pavements always receive an asphalt surfacing with a minimum thickness of 40 mm

174/175
5.1.6 Major multi-lane highways (structural capacity: 100 million ESALs)

Existing roads falling into this category would gen- good natural gravel (CBR > 80%) would invariably
erally have been constructed with a deep pave- lie beneath the asphalt layers, or a layer of cement
ment structure that includes an asphalt base layer stabilised material. Normally the asphalt will be in a
and surfacing. A good quality crushed stone or distressed state with cracks affecting the full depth

Mix designs are required to ensure that sufficient

strength is achieved by adding stabilising agents. Existing Pavement Recycle /
The minimum requirements for these main road
pavements are: Thick CBR
> Cement stabilisation (denoted as CTB in the mm %
chart): UCS > 2 MPa
> Bitumen stabilisation (Denoted as BSM in the
chart): ITSDRY > 225 kPa, ITSWET > 100 kPa,
Good support

Pulverise
ITSEQUIL > 175 kPa, cohesion > 250 kPa, 150 AC and com-
angle of friction > 40 pact
175 mm
150 80

Asphalt 150 50

Bitumen stabilised material (BSM)

20
Cement stabilised material (CTB)
CBR > 100 New GCS
CBR > 80 GCS (graded crushed stone)
Remove existing
45 < CBR < 80 (coarse gravel)
Thick CBR Step 1
25 < CBR < 45 (natural gravel) mm %
15 < CBR < 25 (gravel / soil) Mill off asphalt and
haul to BSM mixing
Good support

10 < CBR < 15 (sandy silty soil) 150 AC plant

7 < CBR < 10 (silty soil)
3 < CBR < 7 (clayey silty soil) 150 80

Rules 150 50
1. Recycling with cement:
minimum thickness: 150 mm
20
2. Recycling with bitumen:
minimum thickness: 100 mm
 of the layer. Two different options are shown for with asphalt. In addition, an option that involves
the pavement that is recycled with cement or lime, two-part recycling is included.
one with a layer of crushed stone material on top
of the recycled/cement stabilised layer, the other

Mechanical Recycle with cement or lime

modify Recycle with bitumen
Crushed stone overlay HMA overlay
Thick CBR Thick CBR Thick CBR Thick CBR
mm % mm % mm % mm %
50 HMA
Import
180 HMA 150 mm 150 100
(new layer)
90 HMA 50 HMA
Recycle Recycle
175 100 Recycle 250 mm
300 mm 300 CTB 300 mm
300 CTB with 250 BSM
with with
cement bitumen
125 80 cement
50 80
150 50 150 50 150 50 150 50

20 20 20 20

Two-part recycling to achieve deep composite pavement

asphalt Recycle in place with cement Import recycled BSM & overlay
Thick CBR Step 2 Thick CBR Step 3 Thick CBR
mm % mm % mm %
50 HMA
Import &
pave
150 mm 150 BSM
BSM
Recycle
150 80 250 mm
with 250 CTB 250 CTB
150 50 cement
50 50 50 50
20 20 20

These pavements always receive an asphalt surfacing with a minimum thickness of 50 mm

176/177
5.2 Alternatives for pavement rehabilitation

The real challenge for a pavement engineer tasked > to provide an indication of the impact each
with a rehabilitation design is selecting the alterna- different pavement has on the environment, the
tive options for treating a distressed pavement. energy consumed by construction activities in
Often such comparisons do not go further than achieving the required service life for each differ-
simplistic costing exercises, normally confined to ent pavement can be estimated and aggregated.
initial construction costs. In addition, the alterna-
tive options for rehabilitation are often not properly Several other aspects could also be incorporated
formulated. This results in different options having in comparing different pavements over a defined
different life expectations (structural capac- service life, the most important being road-user
ity), making any comparison a futile exercise of costs. However, by tailoring maintenance and/or
subjectivity. Furthermore, the current increasing phased interventions aimed at keeping the func-
awareness of the need to reduce energy consump- tional properties of the pavement at similar levels,
tion in all walks of life brings a new dimension to these may be ignored since they will all make simi-
evaluating construction activities. lar contributions. Hence, including only the cost
of all construction activities and energy consumed
This section aims to illustrate the importance of in the initial provision of an adequate pavement
three primary aspects that should be addressed in structure, the various interventions required dur-
comparing different pavements: ing the service life, as well as rehabilitation at the
> the pavement structures being compared must end of the service life are considered adequate
for comparing different pavement structures and
have similar service lives. The pavement can
portraying the real picture that allows a realistic
be constructed initially to provide sufficient
whole-of-life comparison to be made.
structural capacity for the duration of the service
life with timeous interventions to meet functional
Selection of pavement rehabilitation options can
requirements. Alternatively, a phased construc-
best be explained by way of an example. This sec-
tion approach can be adopted with strengthen-
tion uses the comparison of four different options
ing interventions that will allow the service life to
for rehabilitating a hypothetical pavement to dem-
be achieved;
onstrate a proposed method for incorporating cost
> all costs incurred in achieving the required and energy in determining which option is optimal.
service life must be considered, not solely the A typical pavement structure (for many countries)
initial cost of construction i.e. cradle to cradle. consisting of thick asphalt concrete constructed
Provided the different pavements are properly on layers of granular material has been as-
conceived to provide similar service lives, such sumed. The four rehabilitation options selected
an exercise is relatively straightforward; and for comparison include conventional construction
methods as well as those that incorporate recy-
cling material from the existing pavement. The key
requirements are a structural capacity of 20 million
equivalent 80 kN standard axle loads (ESALs) over
a 20-year service life.
The envisaged maintenance/strengthening The Present Worth of Costs based on the whole-
interventions and rehabilitation requirements of-life cost estimate and different discount rates
after 20-years are then defined and all construc- are then used to evaluate the relative economic
tion activities quantified. These quantities allow a differences.
whole-of-life cost estimate to be calculated, based
on unit rates prevailing in the countrys construc- Finally, the same quantities are used to determine
tion industry. the amount of energy consumed during the service
life, including rehabilitation after 20 years.

178/179
5.2.1 Existing pavement

The figure adjacent shows the structure of a

pavement that is typical for a heavily trafficked
road. Full-depth cracking in the asphalt layers
indicates that the pavement has reached the end
of its service life. The various pavement layers are HMA 150 mm
150 mm of thick asphalt concrete overlying
350 mm of granular material consisting of two
layers; a 150 mm layer of good quality graded
crushed stone (CBR > 80%) above a 200 mm
thick layer of natural gravel (CBR > 45%). CBR > 80 150 mm
The total cover to the underlying subgrade is
therefore 500 mm. The subgrade is assumed to
have an in situ resilient modulus of 85 MPa.

Distress symptoms are typical for such a pave- CBR > 45 200 mm
ment structure at the end of the service life. The
asphalt material has suffered fatigue cracking with
cracks propagating thorough the full thickness of
asphalt. Such cracks allow water to enter the un- CBR 15
derlying crushed stone material causing saturation
and leading to the hydraulic displacement of fines
(pumping) when subjected to heavy traffic loads.
The consequence of pumping is degradation of
the layer and pothole development. This pavement
has reached its terminal state and requires major
rehabilitation.
5.2.2 Rehabilitation requirements

Rehabilitation objectives call for a 20 year service tion course (UTFC) surfacing is required. Such a
life. Anticipated traffic over this period indicates surfacing is expected to provide a service life of
a structural capacity requirement of 20 million between six and eight years. At the end of the
ESALs. To meet the normal requirements for ride 20 year service life, rehabilitation will be required to
quality and skid resistance, an ultra-thin fric- restore structural capacity.

180 / 181
5.2.3 Rehabilitation options

Four alternative design options are considered UTFC layer together with 35 mm of underlying
below, each meeting the structural capacity re- asphalt will carry the design traffic for a further
quirement of 20 million ESALs. The AASHTO 1993 7 years when the same treatment will be required
Design Method (Structural Numbers) provides a to achieve the overall 20 year service life. At that
simple means of evaluating the alternative pave- stage, advanced distress (in the form of bitumen
ment structures, using the following input data to stripping and washboarding) can be expected in
determine the Structural Number requirer (SNREQ): the body of the asphalt, requiring deep milling to
> Subgrade support conditions: CBR 15% address the problem. Rehabilitation requirements
(average 85 MPa / 12 392 psi) are expected to be the same as those described
> Reliability: 90% under Option 2 below.
> Standard deviation: 0.45
> Initial serviceability: 4.2
> Terminal serviceability: 2.5 30 mm UTFC
60 mm HMA
A SNREQ value of 4.63 is obtained by using the
above data as input for an appropriate SN com-
75 mm deep patching
puter programme.

Option 1. Patch and overlay 75 mm aged asphalt

This option is popular in many first-world coun-
tries, primarily due to the speed and simplicity of
construction. Severely cracked sections that fall 150 mm GCS
within the wheel paths are milled out and replaced
with fresh asphalt before an overlay is applied. A
milling machine with a 1 m cut width can be uti-
lised to cut a 75 mm deep strip following the wheel
path. Continuously graded hot mixed asphalt
(HMA) is used as backfilled material, placed by 200 mm gravel
paver and compacted.

To minimise the thickness of asphalt overlay,

a phased construction approach is adopted.
A 60 mm thick asphalt base surfaced with a
30 mm thick UTFC surfacing is estimated to pro-
vide a 7 year life before cracks reflecting from the
underlying fatigued structure demand intervention.
This is timed to coincide with the requirement to
replace the UTFC surfacing as it would reach the
end of its functional life after 7 years. It is optimisti-
cally estimated that milling off and replacing the
 Option 1. Patch and overlay. Structural Number determination

Layer coefficient Drainage Layer thickness (t) Layer contribution

Layer
(CL per inch) coefficient (CD ) (mm / inch) (CL x CD x t)

New UTFC 0.44 1 30 / 1.2 0.53

New asphalt
0.42 1 60 / 2.4 1.01
binder
Patched
0.33 * 1 75 / 3 0.99
asphalt

Aged asphalt 0.22 ** 1 75 / 3 0.66

Old GCS
0.12 *** 1 150 / 6 0.72
base
Gravel
0.10 0.9 200 / 8 0.72
subbase

SNACT 4.63

* Structural layer coefficient of 0.33 reflects the part new / part aged composition of the layer
** Structural layer coefficient of 0.22 reflects the aged (brittle) nature of the asphalt
*** Structural layer coefficient of GCS reduced from 0.14 to 0.12 since pumping has occurred

182 / 183
Option 2. Mill and replace
This rehabilitation method calls for the entire
thickness of distressed asphalt affected by full-
depth cracking to be milled off and removed from
site. The underlying crushed stone base will then 30 mm UTFC
require repairing by in situ reworking (to a nominal
depth of 125 mm) before paving a 150 mm thick
asphalt concrete base, followed by the 30 mm 150 mm HMA
UTFC surfacing.

The figure illustrates the operations required. Rework 125 mm

Reworking the crushed stone base implies that
traffic will have to be diverted for sufficient time
to allow the base material to dry back before the
asphalt base can be paved, followed by the UTFC
surfacing.

The critical layer in this pavement is the com-

bined asphalt base and surfacing layers that will
suffer fatigue cracking due to tensile strain level
developing at the bottom of the asphalt concrete.
Two maintenance interventions are envisaged to
coincide with the anticipated life of the UTFC sur-
facing. After 7 and 14 year intervals, only the UTFC
will need replacing. At the end of the service life,
fatigue cracks would have reached the surface,
allowing water to ingress through to the underly-
ing granular layers, causing the same distress
and failure mechanism that the pre-rehabilitated
pavement has suffered. At that stage, rehabilita-
tion requirements are expected to be the same as
those described under Option 1 above.
 Option 2. Mill and replace. Structural Number determination

Layer coefficient CL Drainage Layer thickness t Layer contribution

Layer
(per inch) coefficient CD (mm / inch) CL x CD x t

New UTFC 0.44 1 30 / 1.25 0.53

New asphalt
0.42 1 150 / 6 2.52
base
Reworked
0.14 1 150 / 6 0.84
GCS base
Gravel
0.1 0.9 200 / 8 0.72
subbase

SNACT 4.61

184 / 185
Option 3. Recycle / cement stabilise and overlay to be slushed that, in turn, calls for a drying out
A standard rehabilitation approach that is popular period before the asphalt binder layer and UTFC
in several parts of the world is shown in the figure surfacing can be applied.
below. It calls for the upper 300 mm of existing
pavement to be recycled in situ, stabilised with Since a cohesionless crushed stone base cannot
cement. Such a blend of recovered asphalt pave- tolerate the action of traffic without ravelling, this
ment material and crushed stone would normally method of rehabilitation calls for all traffic to be
require the addition of some 2.5% (by mass) of diverted away from the works until the asphalt has
cement to achieve an unconfined compressive been applied.
strength (UCS) of 2 MPa.
The critical layer in this pavement is the crushed
After a curing period of 7 days, a new 150 mm stone base. The failure condition assumed is
thick highly densified graded crushed stone base 20 mm of permanent deformation followed by
is constructed from on top of the new subbase. degradation due to moisture-activated distress.
Achieving high levels of density requires the layer
Two maintenance interventions are envisaged
to coincide with the anticipated life of the UTFC
surfacing. After 7 years, only the UTFC will need
30 mm UTFC replacing. After a further 7 years (i.e. 14 years after
the initial rehabilitation), both layers of asphalt will
35 mm HMA
need replacing to ensure that the 20 year service
life is achieved. At the end of the service life,
150 mm GCS deformation in the wheel paths can be expected to
be in the order of 20 mm. At that stage, rehabilita-
tion requirements are expected to be the same as
those described under Option 4 below.

300 mm CTB
 Option 3. Recycle/cement stabilise and overlay. Structural Number determination

Layer coefficient CL Drainage Layer thickness t Layer contribution

Layer
(per inch) coefficient CD (mm/inch) CL x CD x t

New UTFC 0.44 1 30/1.2 0.53

New asphalt
0.42 1 35/1.3 0.54
binder
New GCS
0.14 1 150/6 0.84
base
Recycled
0.17 1 300/12 2.04
CTB
Gravel
0.1 0.9 200/8 0.72
subbase

SNACT 4.67

186/187
Option 4. Recycle / bitumen stabilise emulsion to break sufficiently. Instant cohesion is,
This rehabilitation method calls for the upper however, achieved on compaction when foamed
250 mm of the pavement to be recycled in situ bitumen is used as the stabilising agent.
with the addition of a bitumen stabilising agent, as
illustrated in the figure below. (Assumed applica- Since the asphalt concrete surfacing cannot be
tion rates are 2.2% residual bitumen and 1% ce- applied until the moisture content of the recy-
ment (by mass.)) One of the reasons for this meth- cled base has reduced (< 50% of the optimum is
od becoming popular is the increase in cohesion normally specified), a fog spray of dilute emulsion
of the stabilised material that allows the completed is usually applied to prevent the finished surface
layer to be opened to traffic soon after it has been from ravelling under traffic action.
compacted (normally to a density in excess of
100% of the modified AASHTO T-180 density) and The critical layer in this pavement is the bitumen
finished off. When a properly-formulated bitumen stabilised base. However, the deviator stress ratio
emulsion is used as the stabilising agent, a delay is below 30% implying that the failure condition
of between 2 and 4 hours is required to allow the assumed (20 mm of permanent deformation in the
wheel paths) will not be reached after the repeated
loading of 20 million ESALs. (Note: Permanent
deformation will occur, but the total amount will be
less than 20 mm.) Two maintenance interventions
are envisaged to coincide with the anticipated
30 mm UTFC life of the UTFC surfacing. After 7 and 14 year
35 mm HMA intervals, only the UTFC will need replacing. At
the end of the service life, permanent deformation
will be evident in the wheel paths, and this can be
addressed by milling off and replacing the asphalt
250 mm BSM layers. This will return the pavement to its original
state and restore the structural capacity.
 Option 4. Recycle/bitumen stabilise. Structural Number determination

Layer coefficient CL Drainage Layer thickness t Layer contribution

Layer
(per inch) coefficient CD (mm/inch) CL x CD x t

New UTFC 0.44 1 30/1.2 0.53

New asphalt
0.42 1 35/1.3 0.54
binder
Recycled
0.26 1 250/10 2.60
BSM base
Remaining
0.14 1 50/2 0.28
GCS
Gravel
0.1 0.9 200/8 0.72
subbase

SNACT 4.67

188/189
5.2.4 Maintenance requirements

The figure below summarises the various main- above, that are applicable in this example.
tenance and rehabilitation measures explained

Year 0 7 years 14 years 20 years

Option 1 Rehab
Patch 15% type # 1 Replace Replace
overlay HMA 35 mm 35 mm
HMA HMA
+ UTFC + UTFC Rehab
type # 2

Option 2 Rehab
Mill off, type # 2 Replace Replace
rework base, UTFC UTFC
Rehab
replace HMA type # 1

Option 3 Rehab
Recycle with type # 3 Replace Replace
cement, UTFC 35 mm
overlay GCS, HMA Rehab
thin HMA + UTFC type # 4

Option 4 Rehab
Recycle with type # 4 Replace Replace
Replace
bitumen, UTFC UTFC
35 mm
thin HMA HMA
+ UTFC

Maintenance intervals and rehabilitation required after 20 years

190/191
5.2.5 Construction & maintenance costs

The table below is an example of the exercise nance interventions. This example uses average
that needs to be undertaken in order to obtain the unit rates prevailing in the South African roads
cost of the various construction activities for each industry in 2011.
rehabilitation option and for the different mainte-

Cost estimate for each option (Quatities for 1 km of road 10 m wide) using average unit rates
Schedule Rehab option # 1 Rehab option # 2 Rehab option # 3

Rate Patch/overlay Mill & replace CTB/GCS/HMA

Item Unit
(US$) Quantity Amount Quantity Amount Quantity Amount
Milling m 30,0 281 8,438 1,500 45,000
HMA patch ton 70,0 703 49,219
Reprocess m 13,2 1,250 16,500
GCS m 50,0 3,150 157,500
Prime m 0,5 10,000 5,000 10,000 5,000
Tack m 0,4 21,500 8,600 20,000 8,000 20,000 8,000
30 UTFC ton 80,0 750 60,000 750 60,000 750 60,000
HMA ton 70,0 1,500 105,000 3,000 210,000 875 61,250
20 km haul m km 0,4 5,625 2,250 30,000 12,000 incl 0
Recycle
m 5,5
300 CTB 3,000 16,500
m 6,6
250 BSM
Cement ton 150,0 158 23,625
Bitumen ton 400,0
233,506 356,500 331,875
 Rehab option # 4 Intervention # 1 Intervention # 2
BSM/HMA UTFC only 35 mm HMA + UTFC
Quantity Amount Quantity Amount Quantity Amount
300 9,000 650 19,500

30,000 12,000 10,000 4,000 20,000 8,000

750 60,000 750 60,000 750 60,000
875 61,250 875 61,250
6,000 2,400 13,000 5,200

2,500 16,500
53 7,875
116 46,200
191,825 75,400 153,950

192/193
These costs can then be converted to ratios, as shown in the following table.

Cost per kilometre (zero discount rate)

Rehabilitation Initial 7 year 14 year Rehabilitation
Total Cost
option rehabilitation intervention intervention after 20 years
Option 1 2.3X 1.5X 1.5X 3.6X 9.0X
Option 2 3.6X 0.75X 0.75X 2.3X 7.4X
Option 3 3.3X 0.75X 1.5X 1.9X 7.5X
Option 4 1.9X 0.75X 0.75X 1.5X 5.0X

The effect of different discount rates can then the different options (Options 1, 2 and 3) affected,
be evaluated using the Present Worth of Costs but this exercise emphasizes the true benefit of
(PWoC) model, as shown in the figure below. This adopting a technology that provides improved
figure highlights the importance of considering the performance over the full service life.
time value of money. Not only are the rankings of

7
6
PWoC (X times)

5
4
3
2
1
0
4% 6% 8%
Discount Rate (%)

Option 1 Option 2 Option 3 Option 4

Present Worth of Costs for Alternative Rehabilitation Options

5.2.6 Energy consumption

An increasing awareness of climate change is Several authors have published papers highlighting
making society focus more on energy consump- the savings that can be anticipated from adopting
tion. The construction industry is not exempt and different construction techniques (e.g. recycling
several studies have been undertaken to estimate material from an existing pavement compared to
the amount of energy being consumed, particularly conventional construction processes). Applying
in the construction of roads where large machinery these approaches systematically allows the total
is employed and the quantities of material either energy consumed by all construction activities to
consumed or moved is high. be determined for each option, as shown in the
table on the next page.
Studies have been carried out to evaluate how
much energy is consumed in the production of Note. There are several different sources of energy
construction materials (e.g. bitumen, cement, consumption data for the construction industry,
aggregates, etc.), as well as various construction some dating back to the 1970s when energy
activities (e.g. excavating, transporting, asphalt concerns were first popularised. For illustrative
paving, etc.) purposes, this example uses one source that was
recently published in New Zealand.

194/195
 Estimate of energy consumption for each option (Quatities for 1 km of road

Schedule Rehab Option # 1 Rehab Option # 2 Rehab Option # 3

Rate Patch/overlay Mill & replace CTB/GCS/HMA

Item Unit
(MJ) Quantity Energy Quantity Energy Quantity Energy
Milling ton 5 281 1,406 3,750 18,750
Ingredients ton 348 281 97,788
Mix/Pave ton 320 281 89,920
Crush GCS and import
GCS ton 50 3,150 157,500
Recycle existing pavement
125 GCS ton 11 2,625 28,875
250 BSM ton 11
300 CTB ton 11 6,300 69,300
Stabilising agents
Cement ton 7,000 158 1,102,500
Bitumen ton 6,000
Process & finish off layers

125 GCS m 10 10,000 100,000

150 GCS m 10 10,000 100,000
300 CTB m 10 10,000 100,000
250 BSM m 10

30 UTFC
Ingredients ton 407 750 305,250 750 305,250 750 305,250
Mix/pave ton 320 750 240,000 750 240,000 750 240,000
HMA
Ingredients ton 348 1,500 522,000 3,000 1,044,000 875 304,500
Mix/pave ton 320 1,500 480,000 3,000 960,000 875 280,000
20 km haul ton-km 20 2,812 56,245 7,500 150,000 1,625 32,500

1,792,609 2,846,875 2,691,550

10 m wide) Unit rates (MJ) derived from Patrick et al

Rehab Option # 4 Intervention # 1 Intervention # 2

BSM/HMA UTFC only 35 mm HMA + UTFC
Quantity Energy Quantity Energy Quantity Energy
750 3,750 1,625 8,125

5,250 57,750

53 367,500
116 693,000

10,000 100,000

750 305,250 750 305,250 750 305,250

750 240,000 750 240,000 750 240,000

875 304,500 875 304,500

875 280,000 875 280,000
1,625 32,500 1,500 30,000 3,250 65,000

2,380,500 579,000 1,202,875

196/197
The cumulative energy that is consumed by the compared. For comparison purposes, these data
various construction activities (including main- are summarised in the table below and shown
tenance interventions and rehabilitation at the graphically in the following figure.
end of the service life) for all four options can be

Cumulative energy consumed per kilometre (in GJ )

Rehabilitation
Initial Constr 7 yr maint. 14 yr maint. 20 yr rehab.
option
Option 1 1,793 2,996 4,199 7,046
Option 2 2,847 3,426 4,005 5,798
Option 3 2,692 3,271 4,474 6,855
Option 4 2,381 2,960 3,539 4,742

8,000
7,000
Consumed / km (Gj)
Cumulative Energy

6,000
5,000
4,000
3,000
2,000
1,000
0
Initial Constr. 7 yr maint. 14 yr. maint. 20 yr rehab.
Year

Option 1 Option 2 Option 3 Option 4

Cumulative Energy Consumption for Different Rehabilitation Options

Similar to the trend shown in the costing exer- performance over the full service life with the
cise, the above figure illustrates the true benefit picture changing from that painted by the energy
of adopting a technology that provides improved consumed during the initial construction.
5.2.7 Relevant comments

Environmental considerations in pavement > Whole-of-life analysis using PWoC provides

engineering are no longer esoteric. Increasing more realistic financing requirements for pave-
emphasis on the environmental impact of road ment upkeep over the entire analysis period.
construction and rehabilitation, has led to sufficient
data becoming available for use in analysis and > Energy consumption and its impact on the en-
decision making. Energy consumption figures have vironment can be predicted. The inclusion of an
been used in combination with whole-of-life costs energy consideration may influence the rankings
for four realistic rehabilitation options currently of different rehabilitation options and thus deci-
used globally in road pavements. sion making.

This provides insight into project selection leading

to the following conclusions:

> Initial construction costs alone are inadequate

for selection of rehabilitation alternatives. They
can provide skewed and unrealistic rehabilita-
tion selection, which will lead to unnecessary
wastage of resources.

198/199
6 Recycling 100% reclaimed asphalt
pavement (RAP) material

6.1 RAP material 203

6.1.1 Bitumen binder 203
6.1.2 Grading of the RAP material 205
6.2 Uses for cold recycled RAP material 206
6.2.1 Untreated RAP material 206
6.2.2 RAP material treated with cement 207
6.2.3 RAP material treated with bitumen emulsion 207
6.2.4 RAP material treated with foamed bitumen 211

200/201
Asphalt is manufactured using the best available There are two ways that RAP material can be
aggregates and, since these do not deteriorate treated as a cold recycled material:
over the life time of a pavement, it is not surprising
that RAP is (and has been for decades) one of the > In situ treatment. The existing asphalt can be re-
most recycled materials in the world. claimed and simultaneously treated using cold
in-place recycling technology (also referred
The advent of modern milling machines that to as CIR or partial depth recycling). This
granulate the asphalt in situ has significantly in- process uses a large track-mounted recycler
creased the salvage value of the resulting RAP by to recover the upper portion of asphalt in the
making it a more useful material at no extra cost. existing pavement (normally between 100 mm
This chapter focuses on reusing RAP as a cold and 150 mm cut depth) and simultaneously mix
recycled material. the RAP with additives. The recyclers used for
this process can normally treat a full lane width
in a single pass (e.g. the Wirtgen WR 4200 or
the 2200 CR fitted with a 3.8 m wide cutting
drum/paving screed).
> In plant treatment. The milled asphalt is stock-
piled locally and treated in a Wirtgen KMA 220
mixing plant. The treated RAP material is then
returned to site and paved back in the areas that
were milled. Known as cold in plant recycling,
this technology introduces flexibility into the re-
cycling process by separating the recovery and
reuse operations. In addition, where required,
this process provides the opportunity to crush
and/or screen the RAP material, as well as blend
it with fresh aggregate (e.g. graded crushed
stone) if required, before it is recycled.
6.1 RAP material

The term RAP is given to any (100%) asphalt will be a blend of all component mixes. The two
material (also referred to as asphalt concrete) key features of a RAP material are the bitumen
that is recovered from an existing pavement. The binder (the state, amount and consistency of bitu-
type of asphalt that was originally used together men in the material) and the grading. Also impor-
with variations in the asphalt mix will therefore tant is the tendency for the larger RAP particles to
be reflected in the RAP material. Where multiple break down further when subjected to mixing and
asphalt layers of different mixes are recovered by compaction forces.
milling (or crushed from slabs), the resulting RAP

6.1.1 Bitumen binder

From a cold recycling perspective, it is most Where there is doubt as to whether or not the RAP
important to know whether the bitumen in the RAP can be classified as active or inactive, a repre-
material is active or inactive. In other words, is sentative sample may be tested in a laboratory to
the RAP a black aggregate (inactive) with proper- determine:
ties similar to those of graded crushed stone or is
it sticky material (active) with inherent cohesive- > the amount of bitumen in the RAP material (per-
ness due to the bitumen in the RAP material? This centage by mass), and
is important since the state of the old binder will > the rheological properties of the recovered
have a significant influence on how the recycled bitumen (penetration (Pen), softening point and
material will behave when it is reused. viscosity).

The following observations will indicate that the The results of these tests should be used only as
RAP material may be regarded as inactive: indicators since the tests themselves do not pro-
vide definitive answers; the results are influenced
> Visual appearance: the RAP is a dull grey colour by operator competence. The bitumen extraction
with no black shining surfaces. procedure does not necessarily extract all the
> Brittleness: a chunk of RAP breaks cleanly into bitumen (especially that amount absorbed by the
pieces. aggregate) and the amount of bitumen determined
> Adhesion: pieces of RAP (at ambient tempera- may include some material smaller than 0.075 mm
ture) do not stick the hand when a sample is (filler). Recovered Pen values are notoriously varia-
firmly squeezed. ble, largely due to the influence of the solvent used
to extract the bitumen. This is a delicate test that
demands diligence on the part of the operator if
repeat tests are to yield a similar result.

202/203
In spite of these concerns, the two parameters cate the viscosity of the recovered bitumen, allow-
thus determined provide a meaningful guideline in ing the RAP to be classified as active or inactive:
classifying a RAP material. The Pen value will indi-

Recovered 15 10 5
Penetration:

Active Inactive

Degree of aging

Bitumen state: Viscous Semi-viscous Non-viscous

Squeeze: Sticky Some stickiness No stickiness

The bitumen content of RAP is of little signifi- (A quick and easy way of estimating whether a
cance where the material is classified as inactive. RAP sample is active or inactive is to heat the
However, depending on the intended use, the sample to 70 C and manufacture 100 mm
amount of bitumen in an active RAP material can specimens. Soak the specimens for 24 hours
be an important consideration. The various uses before carrying out ITS tests. If the soaked ITS
for recycled RAP material are discussed below in value is > 100 kPa the RAP should be regarded as
Section 6.2. active.)
6.1.2 Grading of the RAP material

It is important to recognise that the grading of The grading of milled RAP material is influenced by
milled RAP material will always be influenced the condition of the in situ asphalt and the milling
by the purpose of the milling operation. Where operation. The main factors influencing the grading
a contractor is milling for the sole purpose of of the RAP are:
removing asphalt from the road, his focus will be
on production at the lowest possible cost, not on > the composition and uniformity of the existing
the grading of the RAP material. However, where asphalt material
100% asphalt material is recycled in situ, the at- > the condition of the existing asphalt material
tention will be entirely different as grading of the > the temperature of the asphalt in the milled
milled RAP is of primary importance. horizon
> the depth of milling
> the speed of advance of the milling machine
> the rotation speed of the milling drum
> the type of milling drum and condition of the
milling tools
> the direction of cut (up-cutting or down-cutting)

204/205
6.2 Uses for cold recycled RAP material

Untreated RAP material may be used as a sub- pavement materials are expensive and/or scarce.
stitute for conventional material to construct a Furthermore, when treated with a suitable additive
new pavement layer. However, it should always (e.g. bitumen), RAP can be used as a premium
be borne in mind that RAP material is derived material to construct the upper layers of pave-
from good quality aggregates and is therefore a ments that carry heavy loads.
valuable material that should not be squandered
by using it where such quality is not warranted. It The following sections discuss the options avail-
should rather be retained and used as a substitute able for using RAP as a cold recycled material.
(treated where necessary) where conventional

6.2.1 Untreated RAP material

For more heavily trafficked roads, only inactive Such cohesion will resist compaction effort and
RAP should be used as a substitute for graded limit the density achieved during construction.
crushed stone in a base layer. Such a layer of RAP However, under the dynamic action of applied
material is constructed using the same procedures traffic loads, the material will slowly consolidate.
and meeting the same density requirements as To counter this tendency, RAP material that clas-
those applicable to conventional crushed stone sifies as active should always be blended with a
bases. nominal 30% (by volume) of graded crushed stone
to achieve an inactive status.
Due to the viscosity of the residual bitumen, RAP
material classified as active has higher levels of
cohesion than inactive RAP or natural material.
6.2.2 RAP material treated with cement

One of the alternatives for treating RAP material However, RAP is invariably a high quality material
is stabilisation with cement. This is normally under- that can be reused more effectively when treated
taken for the purpose of constructing a semi-rigid with bitumen (e.g. bitumen emulsion or foam bitu-
layer in the pavement that is then covered with men) to take advantage of:
additional pavement layers (usually asphalt).
> the adhesive properties of the existing bitumen
To determine the amount of cement addition that in the RAP which offers improved cohesion; and
is necessary to achieve the required properties, > the viscosity and flexibility of the bitumen in the
a conventional cement stabilisation mix design RAP.
should always be undertaken (as described in
Section 4.2).

6.2.3 RAP material treated with bitumen emulsion

When treating RAP material with bitumen emul- or volumetric analysis of the mix in order to as-
sion, the extended volume of the emulsified bitu- certain the design binder content). In addition, an
men will tend to coat the individual RAP particles. asphalt-type product can be achieved by adding
This can lead to continuous bonding, making the an emulsion containing a rejuvenating agent, espe-
mix asphaltic (i.e. a cold mix asphalt material). The cially when the Pen value of the bitumen recovered
expected behaviour of the treated RAP material from the RAP material > 15.
will be influenced by the following:
Where the RAP material does not fit this category,
> Where the Pen value of the bitumen recovered treating with bitumen emulsion is more likely to
from the RAP classifies the material as active produce a non-continuously bound stabilised ma-
(Pen value > 10), the treated material may be terial (BSM-emulsion) than a cold mix asphalt. A
regarded as asphaltic. mix design should always be undertaken following
> Where the bitumen content of the RAP material the standard procedure outlined in Chapter 4 and
exceeds 5%, the treated material will have more described in Appendix 1. The table below summa-
asphaltic tendencies. rises the selection of mix design method.
> Where the residual bitumen content of the emul-
sion added to the RAP material exceeds 2%,
then asphaltic behaviour can be expected.

Where any two of the above three conditions are

met, treating with bitumen emulsion will create an
asphaltic material and an asphalt-type mix design
should be undertaken (normally a Marshall Design

206/207
Guideline for selecting a mix design method when treating RAP with emulsion

Material behaviour
Influencing factors
Stabilised Type Asphalt Type
1.RAP: Recovered Pen < 10 > 10
2.RAP: Recovered bitumen (%) < 5% > 5%
3. ITSWET specimens manufactured at
< 100 > 100
70 C (kPa)
4.Emulsion: Residual bitumen(%) < 2% > 2%
5.Rejuvenating agent No Yes
Mix design method Appendix 1 Marshall

Bitumen emulsions are usually specially formulated > Break time. The time taken for the bitumen
for treating 100% RAP material. These can include to break out of suspension can be controlled
base bitumen that deviates from the standard through chemical interaction between the emul-
80/100 Pen grade that are normally used to sion and the aggregate.
manufacture emulsions and can incorporate either > Cement addition. A small percentage of cement
harder or softer bitumen. The selection of the is sometimes added to destabilise the emul-
correct type of bitumen emulsion for each sion and trigger the break.
application is essential, as outlined below. > Coating of the aggregate particles. This is nor-
mally assessed visually and provides a primary
Worldwide, medium to slow set (stable grade) indication that the formulation is correct.
cationic bitumen emulsions are almost exclusively > Cohesion of the mix. This is influenced by the
used for treating RAP material. The main factors interaction between the emulsion and the aged
influencing the formulation of a specific bitumen bitumen on the aggregate that sometimes ne-
emulsion are: cessitates the inclusion of a rejuvenating agent.
A cold mix asphalt product is normally appropriate tracted cores are often used to identify such sec-
where distress is confined to the top of the pave- tions. However, more reliable information can be
ment and the thickness of asphalt is sufficient to obtained by evaluating the grading and condition
allow a thin (< 150 mm) layer of asphalt to be of milled RAP that is generated from trial milling at
recycled in situ. Similar to HMA, the performance various locations along the road.
of an asphalt-type RAP material treated with
bitumen emulsion is sensitive to changes in grad- Where a consistent grading cannot be guaranteed,
ing. For this reason it is imperative to carry out a the asphalt should rather be milled off, stockpiled
representative mix design and to identify uniform and screened before being treated in a stationary
sections of asphalt that will be recycled in situ. plant. Where necessary, oversized RAP particles
Visual observations and results from tests on ex- should be crushed and screened.

3800 CR: In situ full lane recycling WR 4200: In situ full lane recycling

208/209
 > Stabilisation Mix Design. As shown in the table
on the previous page, where the grading of
the RAP material is relatively fine and the Pen
value of the recovered bitumen exceeds 10,
high ITSDRY values can be generated, some-
times in excess of 500 kPa. It is for this reason
that a minimum ITSWET value of 100 kPa has
been adopted rather than a TSR percentage
value which would be unnecessarily onerous.
However, the combination of high ITSDRY/low
TSR values can also be used as an indication
that the treatment process is not purely one of
stabilisation but is tending towards the more
KMA 220: Ensuring consistency in the RAP material by continuously bound (asphaltic material). This, in
pre-screening turn, is a warning that the RAP material should
be blended with a nominal 15% (by volume)
The following special considerations should be
crusher dust if a non-continuously bound stabi-
noted for mix designs of RAP material treated with
lised material is required.
bitumen emulsion:

> Marshall Asphalt Mix Design. Various countries

have their own guidelines for Stability and Flow
limits of cold recycled mixes that differ from
HMA limits. (It is not uncommon to obtain high
Stability values for such cold recycled mixes.)
The Basic Asphalt Recycling Manual published
in the USA by the Asphalt Recycling and Recla-
mation Association (ARRA) provides some use-
ful guidance as to appropriate values for these
materials.
6.2.4 RAP material treated with foamed bitumen

Foamed bitumen is a stabilising agent. Bitu- is neither asphaltic nor stabilised. For this reason,
men added to a RAP material in a foamed state RAP material classed as active should be blended
disperses as tiny bitumen splinters that do not with a nominal 30% (by volume) graded crushed
coat the RAP particles with a film of fresh bitumen. stone (normally with a maximum size of 20 mm) to
The reason for adding foamed bitumen is not to prevent adherence between particles as a conse-
rejuvenate the aged bitumen in the RAP material quence of the old bitumen remaining sticky.
(to produce an asphaltic product), but to achieve
a bitumen stabilised material (BSM) with its own However, regardless of the apparent nature of the
characteristics and advantages. RAP material and whether or not it appears to be
similar to that previously treated, a mix design
From a BSM-foam perspective, treating a RAP should always be carried out on representative
material is different from treating all other materials samples to determine the characteristics of the
due to the presence of the aged bitumen. treated product. In particular, the results of ITS
Whereas non-RAP material demands the presence tests (100 mm diameter specimens described in
of a minimum amount of dust (>4% by mass Appendix 1) should be carefully interrogated since
< 0.075 mm) for the bitumen to effectively dis- these provide the best indicators of material be-
perse, RAP material with as little as 1% passing haviour. As discussed under Section 4.3.11, ITSDRY
the 0.075 mm sieve can be successfully treated values in excess of 500 kPa are an indication of
with foamed bitumen. It is understood that this asphaltic behaviour, suggesting that blending with
is due to individual bitumen splinters having suf- graded crushed stone (or crusher dust) is required
ficient heat energy to warm and adhere (as spots) to ensure that a non-continuously bound stabilised
to the aged bitumen in the RAP. product is achieved.

Provided the RAP material can be classed as As with all BSM-foam materials (described in
inactive, treating with foamed bitumen will produce Chapter 4), the bitumen demand for effective
a non-continuously bound bitumen stabilised stabilisation is a function of the grading of the par-
material. If the RAP material falls into the active ent material. RAP materials tend to be coarse and
class, then some adherence between the old therefore demand a relatively low application of
bitumen coating individual RAP particles is likely, foamed bitumen for effective stabilisation, normally
leading to a more continuously bound material that in the range of 1.6% to 2.2% (by mass).

210/211
Hot climates. Where the base course of a heavy As a final word, foamed bitumen stabilised RAP
duty pavement in a hot climate is constructed from materials have proved to be remarkably success-
RAP material treated with foamed bitumen, the ful on numerous projects worldwide and, provided
following aspects need attention: they are properly designed and constructed, can
be used with confidence as a substitute for asphalt
> the shear properties of the mix should be deter- bases.
mined from triaxial tests carried out at repre-
sentative temperatures;
> the axle mass of heavy vehicles needs to be
controlled. Overloading will generate dispro-
portionately high stresses in the base that may
result in accelerated deformation; and
> as an insurance against loss of cohesion and/
or a tendency towards continuous binding, all
mixes should be blended with virgin material
(nominal 15% crusher dust for inactive RAP
or 30% graded crushed stone for active RAP
material (both by volume)).
212/213
Bibliography

AASHTO. Guide for the Design of Pavement Collings D.C. and Jenkins K.J., Whole-of-Life
Structures. American Association of State Highway Analysis of Different Pavements: The Real Picture.
and Transportation Officials Washington D.C., First International Conference on Pavement Preser-
1993. vation, Newport Beach, California, for April 2010
American Concrete Pavement Association. Committee of State Road Authorities (CSRA),
Pavement Analysis Software, base on the 1993 Guidelines for road construction materials.. Pretoria
AASHTO Guide for the Design of Pavement Committee of State Road Authorities Department
Structures. Washington D.C. 1993. of Transport (DoT). (DoT technical recommenda-
tions for highways; draft TRH14). 1985
A guide to the structural design of bitumen-sur-
faced roads in tropical and sub-tropical coun- De Beer, M. 1990. Aspects of the design and
tries. 1993. 4th Edition. Crowthorne, Berkshire: behaviour of road structures incorporating lightly
Transport Research Laboratory (TRL). (Overseas cementitious layers. Ph.D dissertation, University of
road note 31). Pretoria, Pretoria.
Basic Asphalt Recycling Manual. Asphalt Recycling De Beer, M. 1991. Use of the Dynamic Cone Pene-
and Reclaiming Association (ARRA). Annapolis, trometer (DCP) in the design of road structures. In:
Maryland, USA. 2001 Proceedings of the Regional conference for Africa
on soil mechanics and foundation engineering, 10th
Bonfim V. Cold Milling of Asphalt Pavements.
and the International conference on tropical and
Published by Suiang G Oliveira, English version by
residual soils, 3rd, Maseru, September 1991.
Priscila Podboi Adachi & Patricia Pick, Excero,
Sao Paulo, Brazil, 2008 Ebels LJ, and Jenkins KJ. Characterisation of Bitu-
men Stabilised Granular Pavement Material Prop-
Bredenhann SJ and Jenkins KJ. Determination
erties using Triaxial Testing. International Confer-
of Stress-Dependent Material Properties with the
ence on Advanced Characterisation of Pavement
FWD, for use in the Structural Analysis of Pave-
and Soil Engineering Materials ICACPSEM, Athens,
ments using Finite Element Analysis Techniques.
Greece, June 2007
Conference on Asphalt Pavements for Southern
Africa CAPSA 04, Sun City, South Africa, Flexible pavement rehabilitation investigation and
September 2004 design. Pretoria: Committee of State Road Authori-
ties (CSRA), Department of Transport (DoT).
Claessen, A.I.M, and Ditmarch, R. Pavement
(DoT technical recommendations for highways;
evaluation and overlay design. The Shell Method.
draft TRH12). South Africa 1996
Proceedings of the Fourth International Conference
on the Structural Design of Asphalt Pavements. Horak, E. Aspects of deflection basin parameters
Vol.1, Ann Arbor, 1977. used in mechanistic rehabilitation design proce-
dures for flexible pavement in South Africa. PhD
Collings D.C., and Jenkins K.J., The long term
Thesis, University of Pretoria, Pretoria, 1988.
behaviour of bitumen stabilised materials. 10th
Conference on Asphalt Pavements in Southern Interim Report Construction completion for
Africa CAPSA 2011, Drakensberg, South Africa cold-in-place recycling, Placer 80 PM 14.3/33.3,
 California Department of Transportation (Caltrans), Jordaan, G.J. Pavement rehabilitation design
North Region Materials, Marysville, Ca, August based on pavement layer component tests (CBR
2006 and DCP). Research Report 91/241, Department
of Transport, Pretoria, 1994.
Jenkins, K.J. Analysis of a Pavement Layer which
has been treated by Single Pass In Situ Stabilisa- Kleyn, E.G. and Savage, P.F. The application of the
tion, Masters Degree Thesis. University of Natal, pavement DCP to determine the bearing properties
South Africa, 1994 and performance of road pavements. Proceedings
of the International Symposium on Bearing Capac-
Jenkins, K.J. Mix Design Considerations for Cold
ity of Roads and Airfields. Trondheim, Norway,
and Half-warm Bituminous Mixes with emphasis
1982.
on Foamed Bitumen. PhD Dissertation, University
of Stellenbosch, South Africa, 2000 . Loizos, A. and Papavasiliou, V. In situ Characteriza-
tion of Pavement Materials Stabilised with Foamed
Jenkins K.J. and van de Ven M.F.C. Compari-
Asphalt and Cement. International Conference on
sons between In Situ Recycling with Cement and
Advanced Characterisation of Pavement and Soil
Foamed Bitumen or Emulsion on Vanguard Drive
Engineering Materials ICACPSEM 2007, Athens,
in South Africa. First International Symposium on
Greece
Subgrade Stabilization and In Situ Pavement Recy-
cling using Cement ,Salamanca, Spain. 2001. Long, F,M. and Jooste, F.J. Summary of LTPP
Emulsion and Foamed Bitumen Treated Sections.
Jenkins KJ, van de Ven MFC, Molenaar AAA
Technical memorandum compiled on behalf of
and de Groot JLA, Performance Prediction of
SABITA and GDPTRW. Modelling and Analysis
Cold Foamed Bitumen Mixes. Ninth International
Systems, Cullinan, South Africa. (Gautrans report:
Conference on Asphalt Pavements, Copenhagen,
CSIR/BE/ER/2007/0006/B)
Denmark, 2002.
Long, F,M. and Jooste, F.J. A Materials Classifi-
Jenkins K.J. Collings D.C. and Jooste F.J. TG2:
cation and Knowledge Based Structural Design
The Design and Use of Foamed Bitumen Treated
Method for Pavements with Bituminous Stabilised
Materials. Shortcomings and Imminent Revisions.
Materials. Conference for Asphalt Pavements in
Recycling and Stabilisation Conference, NZIHT,
Southern Africa CAPSA 2007, Gaberone,
Auckland, New Zealand, June 2008.
Botswana 2007
Jooste, J.P. The measurement of deflection and
Long F.M. Validation of Material Classification
curvature of road surfaces. CSIR Manual K16, Na-
System and Pavement Number Method. Technical
tional Institute for Road Research, CSIR, Pretoria,
Memorandum. Modelling and Analysis Systems,
1970.
Cullinan. Sabita/Gauteng Department of Public
Jordaan, G.J. Towards improved procedures for Transport, Road and Works, GDPTRW report no.
the mechanistic analysis of cement-treated layers CSIR/BE/IE/ER/2009/0028/C. Pretoria 2009
in pavements. Proceedings of the 7th International
Mathaniya ET, Jenkins KJ, and Ebels LJ. Char-
Conference on Asphalt Pavements, Nottingham,
acterisation of Fatigue Performance of Selected
England, 1992.

214/215
 Cold Bituminous Mixes. International Conference Note: This publication is available as a free
on Asphalt Pavements ICAP, Quebec, Canada, download on the Asphalt Academy website:
August 2006, www.asphaltacademy.co.za
Mulusa W.K., and Jenkins K.J., Characterization Twagira E.M., and Jenkins K.J., Age Hardening
of Cold Recycling Mixes using a Simple Triaxial Behaviour of Bituminous Stabilised Materials. 7th
Test. 6th International Conference of Maintenance International RILEM Symposium ATCBM09. Rho-
and Rehabilitation of Pavements and Technological des Greece, May 2009.
Control. Turin, Italy, July 2009.
Twagira E.M., and Jenkins K.J., Moisture
Paige-Green, P., Netterberg F. And Sampson L. Damage of Bituminous Stabilised Materials
The carbonation of chemically stabilised road con- using a MIST Device. 7th International RILEM
struction materials: Guide to its avoidance. Symposium ATCBM09. Rhodes Greece,
Project Report PR 89/146/1. Council for Scientific May 2009.
and Industrial Research CSIR Transportek,
Theyse, H.L., De Beer, M. & Rust, F.C. 1996.
South Africa, 1990.
Overview of the South African mechanistic design
Paige-Green P. and Ventura D. Durability of analysis method. In: 75th Transportation Research
Foamed Bitumen Treated Basalt Base Courses. Board (TRB) meeting. January 7-11, 1996.
Council for Scientific and Industrial Research,
van Niekerk, A.A. and Hurman, M. Establishing
Transportek Division. Contract Report:
Complex Behaviour of Unbound Road Building
CR-2004/08, Pretoria, South Africa. 2004
Materials from Simple Material Testing, Report,
Patrick, J. and Moorthy, H. Quantifying the Benefits Delft University of Technology, Netherlands, 1995.
of Waste Minimisation in Road Construction.
van Niekerk, A.A., van Scheers, J., and Galjaard,
Recycling and Stabilisation Conference, New Zea-
P.J. Resilient Deformation Behaviour of Coarse
land Institute of Highway Technology, June 2008,
Grained Mix Granulate Base Course Materials
Takapuna Beach, Auckland, New Zealand
from Testing Scaled Gradings at Smaller Specimen
Technical Guidelines TG2, Second Edition, Sizes. UNBAR 5 Conference, University of
Bitumen Stabilised Materials. A Guideline for the Nottingham, 2000.
Design and Construction of Bitumen Emulsion and
van Niekerk, A.A., Mechanical Behaviour and
Foamed Bitumen Stabilised Materials.
Performance of Granular Bases and Sub-bases in
ISBN 978-0-7988-5582-2, Asphalt Academy,
Pavements. Doctoral Dissertation. Delft University
Pretoria, South Africa. May 2009
of Technology, The Netherlands. 2001.
216/217
Appendix 1 L
 aboratory procedures for
stabilised materials

A1.1 Mix design procedure for cement stabilised materials 223

A1.1.1 Sampling and preparation 223
A1.1.1.1 Field sampling 223
A1.1.1.2 Standard soil tests 223
A1.1.1.3 Sample blending 224
A1.1.1.4 Gradings (sieve analyses) 225
A1.1.1.5 Representative proportioning 226
A1.1.1.6 Hygroscopic moisture content 227
A1.1.1.7 Sample quantities 228
A1.1.2 Determination of the moisture/density relationship
of treated material 228
A1.1.3 Manufacture of specimens for testing 229
A1.1.4 Curing the specimens 232
A1.1.4.1 Standard curing 232
A1.1.4.2 Accelerated curing 232
A1.1.5 Strength tests 232
A1.1.5.1 Unconfined compressive strength (UCS) test 232
A1.1.5.2 Indirect tensile strength (ITS) test 234
A1.1.6 Determine the required application of stabilising agent 236
A1.2 Mix design procedure for foamed bitumen
stabilisation237
A1.2.1 Sampling and preparation 238
A1.2.1.1 Field sampling 238
A1.2.1.2 Standard soil tests 238
A1.2.1.3 Sample blending 239
A1.2.1.4 Gradings (sieve analyses) 240
A1.2.1.5 Representative proportioning 242
A1.2.1.6 Hygroscopic moisture content 243
A1.2.1.7 Sample quantities 243
A1.2.2 Active filler requirements 244
A1.2.2.1 Effect of plasticity 244
A1.2.2.2 Determination of active filler requirements 245
A1.2.3 Determination of the foaming properties of bitumen 246
A1.2.4 Treating the sample with foamed bitumen 249
A1.2.5 Manufacture of specimens for testing 254
A1.2.5.1 Manufacture of 100 mm diameter specimens 254
A1.2.5.2 Manufacture of 150 mm diameter specimens 256
A1.2.6 Curing the specimens 257
A1.2.6.1 Curing dry 257
A1.2.6.2 Curing to simulate field conditions 257
A1.2.7 Preparing the specimens for testing 258
A1.2.8 Determination of the indirect tensile strength (ITS) of specimens 259
A1.2.9 Interpretation of the indirect tensile strength (ITS) test results 262
Annexure A1.2.1: Foamed bitumen mix design worksheet 266
Annexure A1.2.2: Foamed bitumen mix design report 267
A1.3 Mix design procedure for bitumen emulsion
stabilisation268
A1.3.1 Sampling and preparation 268

218/219
A1.3.1.1 Field sampling 268
A1.3.1.2 Standard soil tests 269
A1.3.1.3 Sample blending 269
A1.3.1.4 Gradings (sieve analyses) 271
A1.3.1.5 Representative proportioning 273
A1.3.1.6 Hygroscopic moisture content 274
A1.3.1.7 Sample quantities 274
A1.3.2 Active filler requirements 275
A1.3.2.1 Effect of plasticity 275
A1.3.2.2 Determination of active filler requirements 276
A1.3.3 Determine the fluid/density relationship 277
A1.3.4 Treating the sample with bitumen emulsion 278
A1.3.5 Manufacture of specimens for testing 281
A1.3.5.1 Manufacture of 100 mm diameter specimens 281
A1.3.5.2 Manufacture of 150 mm diameter specimens 283
A1.3.6 Curing the specimens 284
A1.3.6.1 Curing dry 284
A1.3.6.2 Curing to simulate field conditions 284
A1.3.7 Preparing the specimens for testing 285
A1.3.8 Determination of the indirect tensile strength (ITS) of specimens 286
A1.3.9 Interpretation of the indirect tensile strength (ITS) test results 289
Annexure A1.3.1: Bitumen emulsion mix design worksheet291
Annexure A1.3.2: Bitumen emulsion mix design report292
A1.4 Testing field samples of bitumen stabilised materials (BSMs) 293
A1.4.1 Field sampling  293
A1.4.2 Sample preparation  293
A1.4.3 Adjust the moisture content  293
A1.4.3.1 Where the Maximum Dry Density (MDD) and Optimum
Moisture Content (OMC) of the material is known with certainty 294
A1.4.3.2 Where the Maximum Dry Density (MDD) and Optimum
Moisture Content (OMC) of the material is not known 294
A1.4.4 Manufacture of 100 mm diameter specimens for testing 296
A1.4.5 Curing the specimens 296
A1.4.6 Determination of Indirect Tensile Strength (ITS) 297

A1.5 Determining the strength of BSM core specimens 300

A1.5.1 Extracting core samples 300
A1.5.2 Cutting core specimens 300
A1.5.3 Curing the core specimens 300
A1.5.4 Determination of bulk density 301
A1.5.5 Determination of the Indirect Tensile Strength (ITS) 301

A1.6 Laboratory equipment requirements 302

A1.6.1 Laboratory equipment for soils testing 302
A1.6.2 Additional laboratory equipment cement (or lime) stabilisation 303
A1.6.3 Additional laboratory equipment for bitumen stabilisation 304

220/221
This appendix includes detailed procedures for carrying out stabilisation mix designs in a laboratory.
Three mix design procedures are described:

Section A1.1: Cement stabilisation (applicable also to stabilisation with hydrated lime)
Section A1.2: Bitumen stabilisation using foamed bitumen
Section A1.3: Bitumen stabilisation using bitumen emulsion

Each section contains all the necessary steps and descriptions for carrying out the relevant mix design,
including sample preparation, specimen manufacture, curing and testing procedures. In addition, all
formulae required for each mix design are included in the relevant section, thereby eliminating any need to
look any further for information. Furthermore, to make these procedures user-friendly and a truly
one-stop source of information, an explanation is provided at the end of each section for interpreting
the various test results.

Two sections on quality control testing for BSMs are included:

Section A1.4: Testing field samples of bitumen stabilised material (BSMs)
Section A1.5: Determining the strength of BSM core specimens

Finally, Section A1.6 includes comprehensive laboratory equipment lists for:

Section A1.6.1 Soils testing (basic laboratory equipment for all types of stabilisation)
Section A1.6.2 Additional items for cement (or lime) stabilisation
Section A1.6.3 Additional items for bitumen stabilisation
A1.1 Mix design procedure for cement stabilised
materials
A1.1.1 Sampling and preparation

A1.1.1.1 Field sampling

Bulk samples are obtained from test pits excavated as part of the field investigations (or from borrow pits
and quarries where fresh materials are to be imported and stabilised). Each layer in the upper pavement
( 300 mm) must be sampled separately and at least 100 kg of material recovered from each layer that is
likely to be included in any recycling operation and will therefore require a mix design

Note:

Samples taken from layers of bound material

(asphalt and previously stabilised materials)
should be pulverised in situ using a small
milling machine (or a recycler) to simulate the
grading that will be achieved when the pave-
ment is recycled.

A1.1.1.2 Standard soil tests

Carry out the following standard tests on the material sampled from each individual layer or source:

> Sieve analysis to determine the grading (fines washing procedure, ASTM D 422);
> Atterberg limits to determine the plasticity index (ASTM D 4318); and
> Moisture/density relationship (AASHTO T-180).

222/223
A1.1.1.3 Sample blending

Where necessary, blend the materials sampled from the different layers (and/or new material) to obtain a
combined sample representing the material from the full recycling depth. The in-situ density of the various
component materials must be considered when blending materials, as illustrated in the example shown
below.

Existing upper pavement structure

60 mm Asphalt
Depth to be recycled:

(in situ density 2,300 kg / m3)

200 mm

250 mm graded crushed stone (GCS)

(in situ density 2,000 kg / m3)

Blend the materials in proportion to layer thickness and in situ density as follows:

Material Mass / m (kg) Proportion by mass (%) Per 10 kg sample (g)

Asphalt
0.06 x 2,300 = 138 138 / 418 = 0.33 0.33 x 10,000 = 3,300
(60 mm at 2,300 kg / m3)
GCS
0.14 x 2,000 = 280 280 / 418 = 0.67 0.67 x 10,000 = 6,700
(140 mm at 2,000 kg / m3)
Total 418 1.00 10,000

Note:

Repeat the standard soil tests listed above

to determine the grading, plasticity index
and the moisture / density relationship of the
blended sample.
A1.1.1.4 Gradings (sieve analyses)

Plot the grading curve for the sample that will be

Note:
used in the mix designs. Include on the graph the
Recommended gradings envelop from the table This exercise is advisable as it allows a
below. This plot will indicate whether additional preliminary indication to be made of the
blending with freshly imported material may be strength that can be expected after the
required. However, if the plot includes a bulge in material has been treated with cement.
the fractions between the 0.075 mm and 2.0 mm (A poorly graded material is difficult to
sieves (as shown by the red line entitled Avoid in compact and the consequent low density
the graph below), consideration should be given achieved will significantly affect the strength
to blending the sample with sufficient suitable of the stabilised material.)
fine material (e.g. 10% by volume of minus 5 mm
crusher dust) to reduce the magnitude of the bulge.

100
90
80
Percentage Passing

70
60
50
40
30
20
10
0
0,1 1 10
Sieve Size (mm)

Target Gradings Avoid

Recommended grading curves

224/225
 Recommended gradings
Sieve size (mm) Percent passing (%)
Coarse Fine
50 100 100
37.5 85 100
26.5 72 100
19 60 100
13.2 50 100
9.5 42 90
6.7 35 80
4.75 30 72
2.36 21 56
1.18 14 44
0.6 9 35
0.425 7 31
0.3 5 27
0.15 3 21
0.075 2 18

A1.1.1.5 Representative proportioning

Separate the material in the prepared bulk sample into the following four fractions:

i. Retained on the 19.0 mm sieve;

ii. Passing the 19.0 mm sieve, but retained the 13.2 mm sieve;
iii. Passing the 13.2 mm sieve, but retained on the 4.75 mm sieve; and
iv. Passing the 4.75 mm sieve.

Reconstitute representative samples in accordance with the grading determined above (for the bulk
sample) for the portion passing the 19.0 mm sieve. Substitute the portion retained on 19.0 mm sieve with
material that passes the 19.0 mm sieve and retained on the 13.2 mm sieve. The example in the table
below explains this procedure:
 Sieve analysis Quantity of material to be included in a 10 kg sample

Percentage passing Passing 13.2 mm Passing 19.0 mm

Sieve size (mm) (from sieve analysis Passing 4.75 mm and retained and retained
on bulk sample) on 4.75 mm on 13.2 mm

19.0 90.5
(53.6/100 x ((72.3-53.6)/100 x ((100-72.3)/100 x
13.2 72.3 10,000) 10,000) 10,000)
= 5,360 g = 1,870 g = 2,770 g
4.75 53.6

If there is insufficient material passing the 19.0 mm sieve but retained on the 13.2 mm sieve for substitut-
ing that retained on the 19 mm sieve, then lightly crush the material retained on the 19.0 mm sieve to
provide more of this fraction.

A1.1.1.6 Hygroscopic moisture content

Two representative air-dried samples, each approximately 1 kg, are used to determine the hygroscopic
(air-dried) moisture content of the material. (Note: Larger sample size should be used for more coarsely-
graded materials.) Weigh the air-dried samples, accurate to the nearest 0.1 g, and then place them in an
oven at a temperature of between 105C and 110C until they achieve constant mass. The hygroscopic
moisture content (Wair-dry) is the loss of mass expressed as a percentage of the dry mass of the sample.
Determine the hygroscopic moisture using Equation A1.1.1.

(Mair-dry Mdry)
Wair-dry = x 100 [equation A1.1.1]
Mdry
where:
Wair-dry = hygroscopic moisture content [% by mass]
Mair-dry = mass of air-dried material [g]
Mdry = mass of oven-dried material [g]

226/227
A1.1.1.7 Sample quantities

The guidelines shown in the following table should be used to estimate the quantity of material required
for the respective tests:

Test Mass of sample required (kg)

Moisture/density relationship (modified AASHTO T180) 40

Optimum cement addition determination (150 mm specimens ) 120

Standard soil tests (gradings, Atterberg Limits, moisture content, etc.) 20

A1.1.2 Determination of the moisture/density relationship of treated material

This test is carried out using standard compaction effort to determine the Optimum Moisture Content
(OMC) and Maximum Dry Density (MDD) of the cement stabilised material.

Step 1. Weigh out the required mass of stabilising agent for each of five 7 kg samples prepared as
described in Section A1.1.1. The amount of stabilising agent required (expressed as a percent-
age by mass of the dry sample) should be close to the anticipated optimum for the material being
treated. In the absence of previous tests, the following can be used as a guideline:

Subbase layers: 2% for coarse material (> 50% retained on 4.75 mm sieve)
3% for fine material (< 50% retained on 4.75 mm sieve)

Base layers: 3% for coarse material (> 50% retained on 4.75 mm sieve)
4% for fine material (< 50% retained on 4.75 mm sieve)

Step 2. Add the stabilising agent to the raw material and mix immediately prior to the addition of water.
In order to simulate conditions on the road, compaction of the stabilised material is delayed for
one hour after mixing the untreated material with stabilising agent and water. The mixed material
is placed in an air-tight container to prevent loss of moisture and is thoroughly mixed every fifteen
minutes.

Step 3. Determine the OMC and MDD for the stabilised material in accordance with the modified
moisture-density relationship test procedure (AASHTO T-180).
A1.1.3 Manufacture of specimens for testing

The procedure described below is for the manufacture specimens that are of 150 mm in diameter and
127 mm in height. These specimens will be used to determine the Unconfined Compressive Strength
(UCS) and Indirect Tensile Strength (ITS) of the material.

Step 1. Place 20 kg of sample, prepared as described in Section A1.1.1, into a suitable mixing container.

Step 2. Determine the dry mass of the sample using equation A1.1.2.

(Mair-dry)
Msample =

( (
Wair-dry
((
[equation A1.1.2]
1+
100

where:
Msample = dry mass of the sample [g]
Mair-dry = air-dried mass of the sample [g]
Wair-dry = moisture content of air-dried sample [% by mass]

Step 3. Determine the required amount of stabilising agent using equation A1.1.3.

Cadd
Mcement = x Msample [equation A1.1.3]
100
where:
Mcement = mass of lime or cement to be added [g]
Cadd = percentage of lime or cement required [% by mass]
Msample = dry mass of the sample [g]

228/229
Step 4. Determine the percentage water to be added for optimum mixing purposes using equation A1.1.4.

Wadd = WOMC Wair-dry [equation A1.1.4]

where:
Wadd = water to be added to sample [% by mass]
WOMC = optimum moisture content [% by mass]
Wair-dry = water in air-dried sample [% by mass]

The amount (mass) of water to be added to the sample is determined using Equation A1.1.5.

Wadd
Mwater = x (Msample + Mcement) [equation A1.1.5]
100
where:
Mwater = mass of water to be added [g]
Wadd = water to be added to sample (from equation A1.1.4) [% by mass]
Msample = dry mass of the sample [g]
Mcement = mass of lime or cement to be added [g]

Step 5. Mix the material, cement and water until uniform. Allow the mixed material to stand for one hour
with occasional mixing (as described in Section A1.1.2). Manufacture three specimens, each
150 mm diameter and 127 mm in height, using modified AASHTO (T-180) compaction effort.

Step 6. Samples are taken during the compaction process and dried to a constant mass to determine the
moulding moisture content (Wmould) using equation A1.1.6.

(Mmoist Mdry)
Wmould = x 100 [equation A1.1.6]
Mdry
where:
Wmould = moulding moisture content [% by mass]
Mmoist = mass of moist material [g]
Mdry = mass of dry material [g]
Steps 7 to 9. Repeat the above steps for at least three different stabiliser contents

Step 10. Remove the specimens from the moulds either by dismantling the split moulds or, if ordinary
moulds are used, extruding the specimens carefully with an extrusion jack, avoiding distortion to
the compacted specimens.

Step 11. Record the mass and volume of each specimen and determine the dry density using equation
A1.1.7.

(Mspec) 100
DD = x x 1.000 [equation A1.1.7]
Vol Wmould + 100
where:
DD = dry density [kg/m3]
Mspec = mass of specimen [g]
Vol = volume of specimen [cm3]
Wmould = moulding moisture content [%]

Note:

With certain materials lacking cohesion, it

may be necessary to leave the specimens in
the moulds for 24 hours to develop strength
before extracting. When this is necessary,
the specimens in the moulds should be kept
in a curing room or covered with damp cloth
(hessian).

230/231
A1.1.4 Curing the specimens

A1.1.4.1 Standard curing

Cure the specimens for seven days at 95% to 100% relative humidity and at a temperature of 20C to
25C in a suitable curing room.

A1.1.4.2 Accelerated curing

Place each specimen in sealed plastic bags and cure in an oven at 70C to 75C for 24 hours. (Note. If
hydrated lime is substituted for cement as the stabilising agent, the curing regime must be changed to
60C to 62C for 45 hours.)

After the curing period, remove the specimens from the curing room (or plastic bags) and allow to cool to
ambient temperature, if necessary. Specimens for unconfined compressive strength (UCS) tests should be
submerged in water at 22C to 25C for four hours prior to testing.

A1.1.5 Strength tests

After curing, two of the specimens are tested to determine the Unconfined Compressive Strength (UCS)
and the remaining third specimen is tested for Indirect Tensile Strength (ITS). The test procedures are
described below.

A1.1.5.1 Unconfined compressive strength (UCS) test

The Unconfined Compressive Strength is determined by measuring the ultimate load to failure of a speci-
men subjected to a constant loading rate of 140 kPa/s (153 kN/min). The procedure is as follows:
UCS specimen in press after reaching peak load

Step 1. P
 lace the specimen on its flat side between the plates of the compression testing machine.
Position the specimen such that it is centred on the loading plates.

Step 2. A
 pply the load to the specimen, without shock, at a rate of advance of 140 kPa/s until the
maximum load is reached. Record the maximum load P (in kN), accurate to 0.1 kN.

Step 3. C
 alculate the UCS for each specimen to the nearest 1 kPa using equation A1.1.8.

(4 x P)
UCS = x 1,000,000 [equation A1.1.8]
( x d2)
where:
UCS = unconfined compressive strength [kPa]
P = maximum load to failure [kN]
d = diameter of the specimen [mm]

Step 4. P
 lot a graph of the UCS strengths achieved against the percentage stabilising agent added using
the average UCS for the two specimens tested for each different stabiliser content. Ignore any
obvious incorrect result that may have been caused by damage to the specimen before testing.

232/233
A1.1.5.2 Indirect tensile strength (ITS) test
The ITS of a specimen is determined by measuring the ultimate load to failure applied to the diametrical
axis at a constant deformation rate of 50.8 mm/minute. Cured specimens are tested (unsoaked) at a
temperature of 25C ( 2C) using the following procedure:

ITS specimen placed between loading strips

Step 1. Place the specimen onto the ITS jig. (Ensure the correct loading strips are appropriate for the di-
ameter of the specimen.) Position the sample such that the loading strips are parallel and centred
on the vertical diametrical plane.

Step 2. Place the transfer plate on the top bearing strip and position the jig assembly centrally under the
loading ram of the compression testing device.

Step 3. Apply the load to the specimen, without shock, at a rate of advance of 50.8 mm per minute until
the maximum load is reached.

Step 4. Record the maximum load P (in kN), accurate to 0.1 kN.
Step 5. C
 alculate the ITS value for each specimen to the nearest 1 kPa using Equation A1.2.13.

2xP
ITS = x 1,000,000 [equation A1.1.9]
xhxd
where:
ITS = Indirect Tensile Strength [kPa]
P = maximum applied load [kN]
h = average height of the specimen [mm]
d = diameter of the specimen [mm]

Step 6. Plot a graph of the ITS strength achieved against the percentage stabilising agent added.

234/235
A1.1.6 Determine the required application of stabilising agent

The required application rate of stabilising agent is that percentage at which the minimum required criteria
are met.

2.5 300
UCS
CTB class
2.0 250
UCS (MPa)

Mean UCS ITS

ITS (kP)
CTB class
1.5 200
Min ITS

1.0 150

2.0 2.5 3.0

Cement added (%)

The example in the above graph indicates that a cement addition of 2.5% will satisfy the requirements for
the specific class shown for the cement stabilised material.
A1.2 Mix design procedure for foamed bitumen
stabilisation

The mix design procedures for foamed bitumen stabilisation described below are undertaken on repre-
sentative samples of material for the following primary objectives:

> To determine whether the material is suitable for stabilising with foamed bitumen;
> To determine whether an active filler needs to be added in conjunction with foamed bitumen;
> To determine the amounts of foamed bitumen and active filler that need to be applied for effective
stabilisation; and
> To obtain an indication of the behaviour (engineering properties) of the stabilised material.

The various tests that are carried out on both untreated and treated samples are essentially routine tests
that can be undertaken by most laboratories equipped for normal routine soils and asphalt testing.

236/237
A1.2.1 Sampling and preparation

A1.2.1.1 Field sampling

Bulk samples are obtained from test pits excavated as part of the field investigations (or from borrow pits
and quarries where fresh materials are to be imported and stabilised). Each layer in the upper pavement
( 300 mm) must be sampled separately and at least 200 kg of material recovered from each layer that is
likely to be included in any recycling operation and will therefore require a mix design.

Note:

Samples taken from layers of bound material

(asphalt and previously stabilised materials)
should be pulverised in situ using a small
milling machine (or a recycler) to simulate the
grading that will be achieved when the pave-
ment is recycled.

A1.2.1.2 Standard soil tests

Carry out the following standard tests on the material sampled from each individual layer or source:

> Sieve analysis to determine the grading (fines washing procedure, ASTM D 422);
> Atterberg limits to determine the plasticity index (ASTM D 4318); and
> Moisture/density relationship (AASHTO T-180).
A1.2.1.3 Sample blending

Where necessary, blend the materials sampled from the different layers (and/or new material) to obtain a
combined sample representing the material from the full recycling depth. The in-situ density of the various
component materials must be considered when blending materials, as illustrated in the example shown
below.

Existing upper pavement structure

60 mm Asphalt
Depth to be recycled:

(in situ density 2,300 kg / m3)

200 mm

250 mm graded crushed stone (GCS)

(in situ density 2,000 kg / m3)

Blend the materials in proportion to layer thickness and in situ density as follows:

Material Mass / m (kg) Proportion by mass (%) Per 10 kg sample (g)

Asphalt
0.06 x 2,300 = 138 138 / 418 = 0.33 0.33 x 10,000 = 3,300
(60 mm at 2,300 kg / m3)
GCS
0.14 x 2,000 = 280 280 / 418 = 0.67 0.67 x 10,000 = 6,700
(140 mm at 2,000 kg / m3)
Total 418 1.00 10,000

Note:

Repeat the standard soil tests listed in Section

A1.2.1.2 above to determine the grading, plas-
ticity index and the moisture / density relation-
ship of the blended sample.

238 / 239
A1.2.1.4 Gradings (sieve analyses)

Plot the grading curve for the sample that will be

Note:
used in the mix designs. Include on the graph the
Target gradings envelop from the table below. If the This exercise is advisable as it allows a pre-
material is predominantly RAP or Natural Gravel, in- liminary indication to be made of the strength
clude the relevant curve (also from the table below). that can be expected after the material has
This plot will indicate whether additional blending been treated with foamed bitumen. (A poorly
with freshly imported material may be required. graded material is difficult to compact and
However, if the plot includes a bulge in the frac- the consequent low density achieved will
tions between the 0.075 mm and 2.0 mm sieves (as significantly affect the strength, especially
shown by the red line entitled Avoid in the graph under saturated conditions.)
below), the sample should be blended with a suf-
ficient suitable fine material (e.g. 10% by volume of
minus 5 mm crusher dust) to reduce the magnitude
of the bulge.

100
90
80
Percentage Passing

70
60
50
40
30
20
10
0
0,1 1 10
Sieve Size (mm)

Target Gradings Typical RAP Gravel Avoid

Recommended grading curves

 Percentage passing each sieve size (%)
Sieve size (mm) Target gradings Typical RAP
Natural gravel
Coarse Fine material

50 100 100 100 100

37.5 87 100 85 100
26.5 76 100 72 100
19 65 100 60 100
13.2 55 90 50 100
9.5 48 80 42 100
6.7 41 70 35 100
4.75 35 62 28 88
2.36 25 47 18 68
1.18 18 36 11 53
0.6 13 28 7 42
0.425 11 25 5 38
0.3 9 22 4 34
0.15 6 17 2 27
0.075 4 12 1 20

240/241
A1.2.1.5 Representative proportioning

Separate the material in the prepared bulk sample into the following four fractions:

i. Retained on the 19.0 mm sieve;

ii. Passing the 19.0 mm sieve, but retained the 13.2 mm sieve;
iii. Passing the 13.2 mm sieve, but retained on the 4.75 mm sieve; and
iv. Passing the 4.75 mm sieve.

Reconstitute representative samples in accordance with the grading determined above (for the bulk sample)
for the portion passing the 19.0 mm sieve. Substitute the portion retained on 19.0 mm sieve with material that
passes the 19.0 mm sieve and retained on the 13.2 mm sieve. The example in the table below explains this
procedure:

Quantity of material to be included for

Sieve analysis
every 10 kg of sample

Percentage
passing Passing 13.2 mm and Passing 19.0 mm and
Sieve size
(from sieve Passing 4.75 mm retained on retained on
(mm)
analysis on 4.75 mm 13.2 mm
bulk sample)

19.0 90.5
13.2 72.3 (53.6/100 x 10,000) ((72.3-53.6)/100 x 10,000) ((100-72.3)/100 x 10,000)
= 5,360 g = 1,870 g = 2,770 g
4.75 53.6

If there is insufficient material passing the 19.0 mm sieve but retained on the 13.2 mm sieve for substituting
that retained on the 19 mm sieve, then lightly crush the material retained on the 19.0 mm sieve to provide
more of this fraction.
A1.2.1.6 Hygroscopic moisture content

Two representative air-dried samples, each approximately 1 kg, are used to determine the hygroscopic (air-
dried) moisture content of the material. (Note: Larger sample size should be used for more coarsely-graded
materials.) Weigh the air-dried samples, accurate to the nearest 0.1 g, and then place them in an oven at a
temperature of between 105C and 110C until they achieve constant mass. The hygroscopic moisture con-
tent (Wair-dry) is the loss of mass expressed as a percentage of the dry mass of the sample. Determine
the hygroscopic moisture using Equation A1.2.1.

(Mair-dry Mdry)
Wair-dry = x 100 [equation A1.2.1]
100
where:
Wair-dry = hygroscopic moisture content [% by mass]
Mair-dry = mass of air-dried material [g]
Mdry = mass of oven-dried material [g]

A1.2.1.7 Sample quantities

The guidelines shown in the following table should be used to estimate the quantity of material required for the
respective tests:

Mass of sample
Test
required (kg)

Moisture/density relationship (modified AASHTO T180) 40

Determination of active filler requirement (100 mm specimens) 60

Optimum bitumen addition indication (100 mm specimens) 80

Optimum bitumen addition determination (150 mm specimens ) 100

Standard soil tests (gradings, Atterberg Limits, moisture content, etc.) 20

242/243
A1.2.2 Active filler requirements

A1.2.2.1 Effect of plasticity

Foamed bitumen stabilisation is normally carried out in combination with a small amount (1% by mass) of
active filler (cement or hydrated lime) to enhance the dispersion of the bitumen and reduce moisture sus-
ceptibility. The Plasticity Index (PI) of the material is normally used as a guideline for the use of hydrated
lime or cement in the mix:

Plasticity Index: < 10 Plasticity Index: > 10

Carry out ITS tests on 100 mm specimens

Pre-treat the material with hydrated lime (ICL value)
to determine the need to add either cement
(The initial consumption of lime (ICL value) must
or hydrated lime, as described in Section
first be determined using the appropriate pH test.)
A1.2.2.2 below..

Pre-treatment of material with a PI > 10 requires that the lime and water be added at least 2 hours prior
to the addition of the foamed bitumen. The pre-treated material is placed in an air-tight container to retain
moisture. The moisture content is then checked and, if necessary, adjusted prior to adding the bitumen
stabilising agent (as described in Section A1.2.4).

Note:

Where the material is pre-treated with lime,

the following tests for the Determination of
Active Filler Requirements described under
Section A1.2.2.2 below are not necessary.
A1.2.2.2 Determination of active filler requirements

Where the PI < 10, the need for an active filler and the type of active filler (cement or hydrated lime) that
is appropriate for the material must first be determined by carrying out ITS tests on 100 mm diameter
specimens for three different mixes made from the same sample. The amount of foamed bitumen added
to each of the three mixes is constant, using the fractions passing the 4.75 mm and 0.075 mm sieves as
a guideline, as shown in the following table:

Guidelines for estimating optimum foamed bitumen addition

Foamed bitumen addition

(% by mass of dry aggregate)
Fraction passing
Fraction passing 4.75 mm sieve Typical type of material
0.075 mm sieve (%)

< 50% > 50%

<4 2.0 2.0 Recycled asphalt (RA/RAP)

47 2.2 2.4
RA/Graded crushed stone/
Natural gravel/blends
7 10 2.4 2.8

> 10 2.6 3.2 Gravels/sands

The first of the three mixes contains no active filler, 1% cement is added to the second mix and 1%
hydrated lime is added to the third mix, all three mixes being treated with the same amount of foamed
bitumen. Material from each of the three mixes is used to manufacture 100 mm diameter specimens that
are cured and tested to determine the relevant ITSDRY and ITSWET values (described in Sections A1.2.4 to
A1.2.8 below). The Tensile Strength Retained (TSR) value is then used as the primary indicator for whether
an active filler is required.

Where the TSR value for the mix with no active filler added is in excess of 60%, the mix design should be
undertaken with no active filler. (This situation is usually confined to materials consisting of good quality
crushed stone, often including a significant proportion of reclaimed asphalt (RAP) material.)

244/245
Where the TSR value of the mix with no active filler added is less than 60%, the mix with the type of ac-
tive filler that produces a significantly higher TSR value (> 5%) indicates a preference for either cement or
hydrated lime and should be used in the following mix designs. If the TSR values for both active fillers are
of the same order (difference < 5%) then either type of active filler is suitable.

Note:

To determine the sensitivity of the active filler,

additional tests on 100 mm diameter speci-
mens may be undertaken using the preferred
active filler at a lower application rate (e.g.
0.75%). However, to avoid compromising the
flexibility of the mix, the maximum allowable
application rate for active filler is 1.0% and
should only be exceeded when hydrated lime
is applied as a pre-treatment to eliminate
plasticity.

A1.2.3 Determination of the foaming properties of bitumen

The foaming properties of bitumen are characterised by:

> Expansion Ratio. A measure of the viscosity of the foamed bitumen, calculated as the ratio of the
maximum volume of the foam relative to the original volume of bitumen; and
> Half Life. A measure of the stability of the foamed bitumen, calculated as the time taken in seconds
for the foam to collapse to half of its maximum volume.

The objective of carrying out the following procedure is to determine the bitumen temperature and per-
centage of water addition that is required to produce the best foam properties (maximum expansion ratio
and half-life) for a particular source of bitumen. These properties are measured at three different bitumen
temperatures in the range of 160 C to 190 C using the following procedure:
Step 1. H
 eat the bitumen in the kettle of the Wirtgen WLB 10 S laboratory unit with the pump circulating
the bitumen through the system until the required temperature is achieved (normally starting with
160C). Maintain the required temperature for at least 5 minutes prior to commencing with testing.

Laboratory plant WLB 10 S

Step 2. F
 ollowing standard procedures described in the Users Manual for the Wirtgen WLB 10 S, calibrate
the discharge rate of bitumen (Qbitumen) and set the timer on the unit to discharge 500 g of bitumen.

Step 3. S
 et the water flow-meter to achieve the required water injection rate.

Step 4. D
 ischarge foamed bitumen into a preheated ( 75C) steel drum for a calculated spray time for
500 g of bitumen. Immediately after the foam discharge stops, start a stopwatch.

Step 5. U
 sing the dipstick supplied with the Wirtgen WLB 10 S (which is calibrated for a steel drum of
275 mm in diameter and 500 g of bitumen) measure the maximum height the foamed bitumen
achieves in the drum. This is recorded as the maximum volume.

246/247
Step 6. Use a stopwatch to measure the time in seconds that the foam takes to dissipate to half of its
maximum volume. This is recorded as the foamed bitumens half-life.

Step 7. Repeat the above procedure three times or until similar readings are achieved.

Step 8. Repeat steps 3 to 7 for a range of at least three water injection rates. (Typically, values of 2%, 3%
and 4% by mass of bitumen are used.)

Step 9. Plot a graph of the expansion ratio versus half-life at the different water injection rates on the
same set of axes (see the example in the graph below). The optimum water addition is chosen as
an average of the two water contents required to meet these minimum criteria.

Repeat Step 1 to 9 for two other bitumen temperatures (normally 170C and 180C).

The temperature and optimum water addition that produces the best foam is then used in the mix design
procedure described below.

15 24
Expansion
14 Bitumen temperature: 170C ratio 22
13 20
Half-life
12 18
Expansion ratio (times)

Half-life (seconds)

11 16
10 14
9 12
Minimum acceptable expansion
8 10
7 Optimum 8
water addition Minimum acceptable half-life
6 6
5 4
Equal Equal
4 2
1 1.5 2 2.5 3 3.5 4 4.5 5

Water addition (% of bitumen)

Determination of optimum water addition for foaming (example)

 Note:

The absolute minimum foaming properties

that are acceptable for effective stabilisation
(material temperature > 15C) are:
Expansion ratio: 8 times
Half-life: 6 seconds
If these minimum requirements cannot be met,
the bitumen should be rejected as unsuitable
for use.

A1.2.4 Treating the sample with foamed bitumen

Laboratory plant WLB 10 S coupled to WLM 30 pugmill mixer

248/249
Step 1. Place the required mass of sample (between 20 kg and 30 kg, prepared as described in Section
A1.2.1 above) into the Wirtgen WLM 30 pug mill mixer.

Step 2. Determine the dry mass of the sample using Equation A1.2.2.

(Mair-dry)
Msample =

( (
Wair-dry
((
[equation A1.2.2]
1+
100

where:
Msample = dry mass of the sample [g]
Mair-dry = air-dried mass of the sample [g]
Wair-dry = moisture content of air-dried sample [% by mass]

Step 3. Determine the required mass of active filler (lime or cement) to be added using Equation A1.2.3.

Cadd
Mcement = x Msample [equation A1.2.3]
100
where:
Mcement = mass of lime or cement to be added [g]
Cadd = percentage of lime or cement required [% by mass]
Msample = dry mass of the sample [g]
Step 4. D
 etermine the percentage water to be added to achieve the ideal mixing moisture content
(75% of the OMC of the material), calculated using Equation A1.2.4.

Wadd = 0.75 WOMC Wair-dry [equation A1.2.4]

where:
Wadd = water to be added to sample [% by mass]
WOMC = optimum moisture content [% by mass]
Wair-dry = water in air-dried sample [% by mass]

The amount (mass) of water to be added to the sample is determined using Equation A1.2.5.

Wadd
Mwater = x (Msample + Mcement) [equation A1.2.5]
100
where:
Mwater = mass of water to be added [g]
Wadd = water to be added to sample (from equation A1.2.4) [% by mass]
Msample = dry mass of the sample [g]
Mcement = mass of lime or cement to be added [g]

Step 5. M
 ix the material, active filler and water in the mixer until uniform. After mixing, inspect the sample
to ensure that the material is in a fluffed state.

> If any dust is observed, add small amounts of water (nominally 0.25% each time) and remix until
a fluffed state is achieved with no dust visible.
> If the material is sticky with a tendency to pack against the side of the mixer, then the mois-
ture content is too high for mixing with foamed bitumen. Reject the sample. Start again with a
fresh sample using a lower moisture content.

250/251
Step 6. Determine the amount of foamed bitumen to be added using Equation A1.2.6.

Badd
Mbitumen = x (Msample + Mcement) [equation A1.2.6]
100
where:
Mbitumen = mass of foamed bitumen to be added [g]
Badd = foamed bitumen content [% by mass]
Msample = dry mass of the sample [g]
Mcement = mass of lime or cement added [g]

Step 7. Determine the timer setting on the Wirtgen WLB 10 S using Equation A1.2.7.

Mbitumen
T= [equation A1.2.7]
Qbitumen
where:
T = time to be set on WLB 10 S timer [s]
Mbitumen = mass of foamed bitumen to be added [g]
Qbitumen = bitumen flow rate for the WLB10 S [g/s]

Step 8. Couple the Wirtgen WLB 10 S to the WLM 30 mixer so that the foamed bitumen can be
discharged directly into the mixing chamber.

Step 9. Start the mixer and allow it to mix for at least 10 seconds before discharging the required mass of
foamed bitumen into the mixer. After the foamed bitumen has discharged, continue mixing for a
further 30 seconds or until uniformly mixed.
Step 10. Determine the mass of water required to bring the sample to the OMC using Equation A1.2.8.

Mwater
Mplus = [equation A1.2.8]
3
where:
Mplus = mass of water to be added [g]
Mwater = mass of water previously added (equation A1.2.5) [g]

Step 11. Add the additional water and mix until uniform.

Step 12. Transfer the foamed bitumen treated material into an air-tight container and immediately seal.
To minimise moisture loss, manufacture the test specimens as soon as possible by following the
relevant procedure for either 100 mm or 150 mm diameter specimens, as described in sections
A1.2.5.1 and A1.2.5.2 respectively.

Repeat the above steps for at least four mixes with different foamed bitumen contents at 0.2% inter-
vals.
The Guidelines for estimating optimum foamed bitumen addition (Section A1.2.2.2 above) should be
used in determining the mid-point of the range of foamed bitumen to be added to the four samples.

An example. If the material consists of a blend of RAP and crushed stone with 39% and 8% passing
the 4.75 mm and 0.075 mm sieves respectively, the guidelines in Section A1.2.2.2 indicate an optimum
bitumen addition of 2.4%. The amount of foamed bitumen to be added to each sample (all with the same
amount of active filler and at the same moisture content) is:

Sample 1: 2.1%
Sample 2: 2.3%
Sample 3: 2.5%
Sample 4: 2.7%

252/253
A1.2.5 Manufacture of specimens for testing

The procedures described below are for the manufacture of two different sizes of specimen using different
compaction procedures:

Specimen size and compaction effort applied in the manufacturing process

Specimen diameter Specimen height Compaction effort

100 mm 63.5 mm Modified Marshall*
150 mm 95.0 mm Modified AASHTO
* 75 blows per face

The following two questions are often raised:

1. Which specimen size should be manufactured? As described in Section A1.2.8 below, ITSDRY and
ITSWET values are normally determined from 100 mm diameter specimens. 150 mm diameter speci-
mens may be substituted for 100 mm diameter specimens to obtain the same values. However,
when dealing with coarse material, (i.e. where the grading curve tends towards the coarse side of
the recommended grading envelop) it is strongly recommended that 150 mm diameter specimens
are manufactured and tested in place of the smaller 100 mm diameter specimens.
Note. Only 150 mm diameter specimens are used to determine ITSEQUIL and ITSSOAK values.

2. Can other compaction methods be used? The compaction procedures described below are well
known standard procedures that can be carried out in most laboratories, worldwide. Other pro-
cedures may be used (e.g. gyratory compaction, vibrating hammer, vibrating table, etc.) provided
they achieve the same density target of 100% Marshall compaction for the 100 mm diameter
specimens or 100% of the mod AASHTO T-180 density for 150 mm diameter specimens.

A1.2.5.1 Manufacture of 100 mm diameter specimens

A minimum of six (6) 100 mm diameter specimens, 63.5 mm in height, are manufactured from each
sample of treated material by applying modified Marshall compaction effort, as described in the following
steps:
Step 1 P
 repare the Marshall mould and hammer by cleaning the mould, collar, base-plate and face of the
compaction hammer. Note: the compaction equipment must not be heated but kept at ambient
temperature.

Step 2. W
 eigh sufficient material to achieve a compacted height of 63.5 mm 1.5 mm (Approximately
1,100 g for most materials). Poke the mixture with a spatula 15 times around the perimeter and 10
times on the surface, leaving the surface slightly rounded.

Step 3. C
 ompact the mixture by applying 75 blows with the compaction hammer. Care must be taken to
ensure the continuous free fall of the hammer.

Step 4. R
 emove the mould and collar from the pedestal, invert the specimen (turn over). Replace it and
press down firmly to ensure that it is secure on the base plate. Compact the other face of the
specimen with a further 75 blows.

Step 5. A
 fter compaction, remove the mould from the base-plate and extrude the specimen by means
of an extrusion jack. Measure the height of the specimen and adjust the amount of material if the
height is not within the 1.5 mm limits.

Note: Coarse materials are often damaged during the extrusion process. It is therefore recom-
mended that the specimens are left in their moulds for 24 hours allowing sufficient strength
to develop before extruding.

Repeat steps 1 to 5 for the manufacture of at least six (6) specimens.

Step 6. T
 ake 1 kg representative samples after compaction of the second and fifth specimen and dry to
a constant mass. Determine the moulding moisture using Equation A1.2.9.

(Mmoist Mdry)
Wmould = x 100 [equation A1.2.9]
Mdry
where:
Wmould = moulding moisture content [% by mass]
Mmoist = mass of moist material [g]
Mdry = mass of dry material [g]

254/255
A1.2.5.2 Manufacture of 150 mm diameter specimens

A minimum of six (6) 150 mm diameter specimens, 95 mm in height, are manufactured from each sample
of treated material by applying modified AASHTO (T-180) compaction effort, as described in the following
steps:

Step 1. Prepare the equipment by cleaning the mould, collar, base-plate and face of the compaction
rammer. (Either split-moulds or standard Proctor moulds may be used, each fitted with a
32 mm spacer placed on the base plate to achieve specimens that are 95 mm (1.5 mm) in
height.)

Note: the modified AASHTO compaction rammer has the following specifications:

Rammer diameter: 50 mm
Mass: 4.536 kg
Drop distance: 457 mm

Step 2. Compact each specimen applying modified AASHTO (T-180) compaction effort (4 layers approxi-
mately 25 mm thick, each receiving 55 blows from the drop rammer.)

Step 3. Carefully trim excess material from specimens, as specified in the AASHTO T-180 test method.

Step 4. After compaction, remove the mould from the base-plate and extrude the specimen by means
of an extrusion jack. Where split-moulds are used, separate the segments and remove the speci-
men.

Note: Coarse materials are often damaged during the extrusion process. It is therefore recom-
mended that the specimens are left in their moulds for 24 hours allowing sufficient
strength to develop before extruding.

Where split moulds are used, it is advisable to leave the specimen in the mould for 4
hours before splitting the mould and extracting the specimen.

Repeat steps 1 to 4 for the manufacture of at least six (6) specimens.

Step 5. Take 1 kg representative samples after compaction of the second and fifth specimen and dry to
a constant mass. Determine the moulding moisture content using Equation A1.2.9 (above).
A1.2.6 Curing the specimens

Two curing regimes are described below. The first is a standard procedure to dry the specimens to con-
stant mass. The second procedure aims to simulate field conditions where the equilibrium moisture con-
tent is approximately 50% of OMC. Both 100 mm and 150 mm diameter specimens may be cured dry (to
constant mass) whereas only 150 mm diameter specimens are cured at equilibrium moisture content.

A1.2.6.1 Curing dry

Place the specimens (either 100 mm or 150 mm diameter) in a forced-draft oven at 40C and cure to
constant mass (normally 72 hours).

To determine if constant mass has been achieved, weigh the specimens and place back in the oven. Re-
move after 4 hours and reweigh. If the mass is constant, then continue with testing. If the mass is not con-
stant, place back in the oven and reweigh after repeated 4 hour intervals until constant mass is achieved.

When constant mass is achieved, remove the specimens from the oven and allow to cool to 25C
( 2.0C).

A1.2.6.2 Curing to simulate field conditions

Place the 150 mm diameter specimens in a forced-draft oven at 30C for 20 hours (or until the moisture
content has reduced to approximately 50% of OMC).

Take the specimens out of the oven, place each in a sealed plastic bag (at least twice the volume of the
specimen) and place back in the oven at 40C for a further 48 hours.

Remove specimens from the oven after 48 hours and take out of their respective plastic bags, ensuring
that any moisture in the bags does not come into contact with the specimen. Allow to cool down to 25C
( 2.0 C).

256/257
A1.2.7 Preparing the specimens for testing

After cooling, determine the bulk density of each specimen using the following procedure:

Step 1. Determine the mass of the specimen.

Step 2. Measure the height of the specimen at four evenly-spaced locations around the circumference
and calculate the average height of the specimen.

Step 3. Measure the diameter of the specimen.

Step 4. Calculate the bulk density of the each specimen using Equation A1.2.10.

4 x Mspec
BDspec = x 1,000,000 [equation A1.2.10]
x d2 x h
where:
BDspec = bulk density of specimen [kg/m3]
Mspec = mass of specimen [g]
h = average height of specimen [mm]
d = diameter of specimen [mm]

Exclude from further testing any specimen whose bulk density differs from the mean bulk density of all six
(6) specimens by more than 2.5%.

Step 5. Place half of the specimens (normally 3) under water in a soaking bath for 24 hours at 25C
( 2 C). After 24 hours, remove the specimens from the water, surface dry and test immediately.
A1.2.8 Determination of the indirect tensile strength (ITS) of specimens

The ITS of a specimen is determined by measuring

the ultimate load to failure applied to the dia
metrical axis at a constant deformation rate of
50.8 mm/minute. Ensure that the temperature
of the specimens is 25C ( 2 C) and follow the
procedure described below:

Specimen mounted in press for the ITS test

Step 1. P
 lace the specimen onto the ITS jig. (Ensure the correct loading strips are appropriate for the di-
ameter of the specimen.) Position the sample such that the loading strips are parallel and centred
on the vertical diametrical plane.

Step 2. P
 lace the transfer plate on the top bearing strip and position the jig assembly centrally under the
loading ram of the compression testing device.

Step 3. A
 pply the load to the specimen, without shock, at a rate of advance of 50.8 mm per minute until
the maximum load is reached.

Step 4. Record the maximum load P (in kN), accurate to 0.1 kN.

Step 5. Record the displacement at break to the nearest 0.1 mm.

Step 6. Break the specimen in half and record the temperature of the specimen at its centre.

258/259
Step 7. Break up one of the unsoaked specimens and dry to a constant mass. Determine the cured mois-
ture content using Equation A1.2.11.

(Mmoist Mdry)
Wspec = x 100 [Equation A1.2.11]
Mdry
where:
Wspec = moisture content of specimen [% by mass]
Mmoist = mass of moist material [g]
Mdry = mass of dry material [g]

Determine the dry density of each specimen using Equation A1.2.12.

Wspec
DDspec = BDspec x 1
100
[Equation A1.2.12]
( (
where:
DDspec = dry density of specimen [% by mass]
BDspec = bulk density of specimen [g]
Wspec = moisture content of specimen [g]

Step 8. Break up one of the soaked specimens and dry to a constant mass. Determine the after-soaking
moisture content using Equation A1.2.11.
Step 9. Calculate the ITS value for each specimen to the nearest 1 kPa using Equation A1.2.13.

2xP
ITS = x 1,000,000 [Equation A1.2.13]
xhxd
where:
ITS = Indirect Tensile Strength [kPa]
P = maximum applied load [kN]
h = average height of the specimen [mm]
d = diameter of the specimen [mm]

Step 10. Use the Worksheet in Annexure A1.2.1 to record the data and the form in Annexure A1.2.2 to
report the results.

Step 11. For dry-cured specimens only, calculate the Tensile Strength Retained (TSR) value for the sample
using Equation A1.2.14.

Ave ITSWET
TSR = x 100 [Equation A1.2.14]
Ave ITSDRY
where:
TSR = Tensile Strength Retained [%]
Ave ITSWET = average ITSWET value [kPa]
Ave ITSDRY = average ITSDRY value [kPa]

Note. To differentiate between the results obtained from the different curing regimes, the terminol-
ogy shown in the table below should be adopted to avoid any confusion.

260/261
 Term Specimen diameter Curing regime Moisture content
ITSDRY 72 hrs unsealed <1%
100 mm or 150 mm
ITSWET 24 hr soak Saturated

20 hr unsealed,
ITSEQUIL 50% of OMC
150 mm only 48 hrs in sealed bag

ITSSOAK 24 hr soak Semi-saturated

A1.2.9 Interpretation of the indirect tensile strength (ITS) test results

Plot the respective soaked and unsoaked ITS test results against the relevant addition of foamed bitumen,
as shown in the example below. If results for two different curing regimes were obtained, each must be
plotted on a separate graph.

The added foamed bitumen that meets the minimum ITS value for the required material classification is
selected as the primary indicator for the minimum amount of foamed bitumen to be added. Engineer-
ing judgement is then used to determine the amount of foamed bitumen that needs to be added in order
to achieve sufficient confidence, based on the variability of the test results (the goodness of fit of the
regression curve or line through the plotted test results).

The example below explains the process.

The table and plot of ITS test results shown below are typical values achieved from a mix design using
100 mm diameter specimens for a natural granular material treated with foamed bitumen. The curve
through the four ITSDRY points approximates the relationship between ITSDRY and added foamed bitumen.
The line through the four ITSWET points approximates the relationship between ITSWET and added foamed
bitumen. The fine dotted lines indicate that the addition of between 2.2% and 2.3% foamed bitumen will
meet the requirements for a Class 1 foamed bitumen stabilised material (ITSDRY > 225 kPa and ITSWET >
100 kPa).
 Added bitumen (%) ITSDRY (kPa) ITSSOAK (kPa) TSR (%)

2.1 181 86 47.5

2.3 206 109 52.9
2.5 262 152 58.0
2.7 244 169 69.3
Indirect tensile strength (ITS) (kPa)

300

ITSDRY
250
Class 1 Min
ITSDRY
200
ITSWET

150

Class 1 Min
100
ITSWET

2.1 2.3 2.5 2.7

Foamed bitumen added (%)

100 mm specimens

The TSR values indicate that the material remains moisture susceptible after foamed bitumen treatment,
especially when the application rate is below 2.3% (i.e. TSR value < 50%). It is therefore important to
ensure that sufficient bitumen is applied to achieve the minimum ITSWET value required to meet the re-
quirements for Class 1 classification.

262/263
Engineering judgement is then applied, based on the understanding that ITS values are not absolute,
together with an appreciation that variability must be expected when recycling material from an existing
pavement. The results obtained from the mix with 2.3% added foamed bitumen fail to meet the Class 1
requirement for ITSDRY whilst both ITSDRY and ITSWET values for the mix with 2.5% added foamed bitumen
are far in excess of the minimum Class 1 requirements. This suggests that a minimum application rate of
2.4% foamed bitumen is required.

To improve on the level of confidence, additional 150 mm diameter specimens can be manufactured
at 0.1% intervals of foamed bitumen addition, ranging from 2.2% to 2.5% added foamed bitumen and
cured at equilibrium moisture content. (Testing specimens at equilibrium moisture content eliminates the
suction forces that elevate ITSDRY values with a consequent reduction in the TSR values.) The table and
graph below show typical results that would be achieved, superimposed with the relevant minimum limits
for ITSEQUIL and ITSSOAK required for a Class 1 foamed bitumen stabilised material (ITSEQUIL > 175 kPa and
ITSSOAK > 100 kPa).
 Added bitumen (%) ITSEQUIL (kPa) ITSSOAK (kPa)

2.2 198 91
2.3 211 132
2.4 205 148
2.5 195 145
Indirect tensile strength (ITS) (kPa)

250
ITSEQUIL
200
Class 1 Min
ITSEQUIL
150
ITSSOAK
Class 1 Min
100
ITSSOAK

50

2.2 2.3 2.4 2.5

Foamed bitumen added (%)

150 mm specimens

This example highlights the improved level of confidence that is achieved in taking the decision to add
2.3% foamed bitumen to ensure that the minimum requirements for a Class 1 bitumen stabilised material
are satisfied.

264/265
Annexure A1.2.1

FOAMED BITUMEN MIX DESIGN - WORKSHEET

Project Date

Sample / Mix No.: Location

Material description :
Maximum dry density Optimum moisture content
Percentage < 0.075mm Grading: Coarse Medium Fine
Plasticity Index

Bitumen Source Bitumen type

Active Filler Type Filler Source

MOISTURE DETERMINATION Specimen manufacture After Curing

Hygroscopic Sample 1 Sample 2 Dry Soaked
Pan No.
Mass wet sample + pan m1
Mass dry sample + pan m2
Mass pan mp
Mass moisture m1-m2 = Mm
Mass dry sample m2-mp= Md
Moisture content Mm/Mdx100=Mh

Mass of air-dried sample placed in the mixer (kg)

Percentage of water added to sample for mixing: Amount of water added :
Percentage water added to sample for compaction Amount of water added :
Total percentage water added: Total water added:

Foamed bitumen addition (%): Active filler addition (%):

Foam water injection rate (%)
Temperatures (C) Material: Bitumen: Water:

SPECIMEN DETAILS
Specimen ID
Date Moulded
Date removed from oven
Date tested
Diameter (mm)

Individual height
measurements (mm)

Average height (mm)

Mass after curing (g)
Bulk density (kg/m 3 )
Average bulk density
Dry density (kg/m 3 )

ITS TEST
Specimen condition Unsoaked (ITSDRY / ITS EQUIL ) Soaked (ITS WET / ITSSOAK )
Maximum load (kN)
Internal temperature (C)
Deformation (mm)
ITS (kPa)
Average ITS (kPa)
TSR (%)
Annexure A1.2.2

FOAMED BITUMEN MIX DESIGN REPORT (Dry curing)

Project Date

Sample number: Location

Material description :
Maximum dry density Optimum moisture content
Percentage < 0.075mm Grading: Coarse Medium Fine
Plasticity Index

Bitumen Source Bitumen type

Active Filler Type Filler Source

FOAMED BITUMEN STABILISED MATERIAL SPECIMENS

Compactive effort mm specimen diameter
Date moulded
Date tested
Foamed Bitumen added (%)
Active filler added (%)
Moulding moisture content (%)

TEST RESULTS
ITS DRY (kPa)
Moisture content at break (%)
Dry Density (kg/m 3)
Average deformation (mm)
Temperature at break (C)

ITS WET (kPa)

Moisture content at break (%)
Dry Density (kg/m 3)
Average deformation (mm)
Temperature at break (C)

Tensile Strength Retained (%)

Material classification

% Foamed Bitumen vs ITS % Foamed Bitumen vs Dry density

420
2150
370
Dry density (kg/m3)

320 2125
ITS (kPa)

270
2100
220
2075
170
120 2050
1,75 2 2,25 2,5 1,75 2 2,25 2,5
Dry Specimens
ITS dry ITS wet Foamed Bitumen added Foamed Bitumen added
Wet specimens

Comments

266/267
A1.3 Mix design procedure for bitumen emulsion
stabilisation

The mix design procedures for bitumen emulsion stabilisation described below are undertaken on repre-
sentative samples of material for the following primary objectives:

> To determine whether the material is suitable for stabilising with bitumen emulsion;
> To determine whether an active filler needs to be added in conjunction with bitumen emulsion;
> To determine the amounts of bitumen emulsion and active filler that need to be applied for effective
stabilisation; and
> To obtain an indication of the behaviour (engineering properties) of the stabilised material.

The various tests that are carried out on both untreated and treated samples are essentially routine tests
that can be undertaken by most laboratories equipped for normal routine soils and asphalt testing.

A1.3.1 Sampling and preparation

A1.3.1.1 Field sampling

Bulk samples are obtained from test pits excavated as part of the field investigations (or from borrow pits
and quarries where fresh materials are to be imported and stabilised). Each layer in the upper pavement
( 300 mm) must be sampled separately and at least 200 kg of material recovered from each layer that is
likely to be included in any recycling operation and will therefore require a mix design.

Note:

Samples taken from layers of bound material

(asphalt and previously stabilised materials)
should be pulverised in situ using a small
milling machine (or a recycler) to simulate the
grading that will be achieved when the pave-
ment is recycled.
A1.3.1.2 Standard soil tests

Carry out the following standard tests on the material sampled from each individual layer or source:

> Sieve analysis to determine the grading (fines washing procedure, ASTM D 422);
> Atterberg limits to determine the plasticity index (ASTM D 4318); and
> Moisture / density relationship (AASHTO T-180).

A1.3.1.3 Sample blending

Where necessary, blend the materials sampled from the different layers (and / or new material) to obtain a
combined sample representing the material from the full recycling depth. The in-situ density of the various
component materials must be considered when blending materials, as illustrated in the example shown
below.

Existing upper pavement structure

60 mm Asphalt
Depth to be recycled:

(in situ density 2,300 kg / m3)

200 mm

250 mm graded crushed stone (GCS)

(in situ density 2,000 kg / m3)

268 / 269
Blend the materials in proportion to layer thickness and in situ density as follows:

Material Mass/m (kg) Proportion by mass (%) Per 10 kg sample (g)

Asphalt
0.06 x 2,300 = 138 138/418 = 0.33 0.33 x 10,000 = 3,300
(60 mm at 2,300 kg/m3)
GCS
0.14 x 2,000 = 280 280/418 = 0.67 0.67 x 10,000 = 6,700
(140 mm at 2,000 kg/m3)
Total 418 1.00 10,000

Note:

Repeat the standard soil tests listed in Section

A1.3.1.2 above to determine the grading, plas-
ticity index and the moisture/density relation-
ship of the blended sample.
A1.3.1.4 Gradings (sieve analyses)

Plot the grading curve for the sample that will be

used in the mix designs. Include on the graph the
Note:
Target gradings envelop from the table below. If
This exercise is advisable as it allows a pre-
the material is predominantly RAP or Natural Gravel,
liminary indication to be made of the strength
include the relevant curve (also from the table below).
that can be expected after the material has
This plot will indicate whether additional blending with
been treated with bitumen emulsion. (A
freshly imported material may be required. However,
poorly graded material is difficult to compact
if the plot includes a bulge in the fractions between
and the consequent low density achieved will
the 0.075 mm and 2.0 mm sieves (as shown by the
significantly affect the strength, especially
red line entitled Avoid in the graph below), the sam-
under saturated conditions.)
ple should be blended with a sufficient suitable fine
material (e.g. 10% by volume of minus 5 mm crusher
dust) to reduce the magnitude of the bulge.

100
90
80
Percentage Passing

70
60
50
40
30
20
10
0
0,1 1 10
Sieve Size (mm)

Target Gradings Typical RAP Gravel Avoid

Recommended grading curves

270/271
 Percentage passing each sieve size (%)
Sieve size (mm) Recommended gradings
Natural gravel Typical RAP
Coarse Fine
50 100 100 100 100
37.5 87 100 100 85
26.5 76 100 100 72
19 65 100 100 60
13.2 55 90 100 50
9.5 48 80 100 42
6.7 41 70 100 35
4.75 35 62 88 28
2.36 25 47 68 18
1.18 18 36 53 10
0.6 12 27 42 6
0.425 10 24 38 4
0.3 8 21 34 3
0.15 3 16 27 1
0.075 2 10 20 0
A1.3.1.5 Representative proportioning

Separate the material in the prepared bulk sample into the following four fractions:

i. Retained on the 19.0 mm sieve;

ii. Passing the 19.0 mm sieve, but retained the 13.2 mm sieve;
iii. Passing the 13.2 mm sieve, but retained on the 4.75 mm sieve; and
iv. Passing the 4.75 mm sieve.

Reconstitute representative samples in accordance with the grading determined above (for the bulk sample)
for the portion passing the 19.0 mm sieve. Substitute the portion retained on 19.0 mm sieve with material that
passes the 19.0 mm sieve and retained on the 13.2 mm sieve. The example in the table below explains this
procedure:

Quantity of material to be included for

Sieve analysis
every 10 kg of sample

Percentage
passing Passing 13.2 mm and Passing 19.0 mm and
Sieve size
(from sieve Passing 4.75 mm retained on retained on
(mm)
analysis on 4.75 mm 13.2 mm
bulk sample)

19.0 90.5
13.2 72.3 (53.6/100 x 10,000) ((72.3-53.6)/100 x 10,000) ((100-72.3)/100 x 10,000)
= 5,360 g = 1,870 g = 2,770 g
4.75 53.6

If there is insufficient material passing the 19.0 mm sieve but retained on the 13.2 mm sieve for substituting
that retained on the 19 mm sieve, then lightly crush the material retained on the 19.0 mm sieve to provide
more of this fraction.

272/273
A1.3.1.6 Hygroscopic moisture content

Two representative air-dried samples, each approximately 1 kg, are used to determine the hygroscopic (air-
dried) moisture content of the material. (Note: Larger sample size should be used for more coarsely-graded
materials). Weigh the air-dried samples, accurate to the nearest 0.1 g, and then place them in an oven at
a temperature of between 105C and 110C until they achieve constant mass. The hygroscopic moisture
content (Wair-dry) is the loss of mass expressed as a percentage of the dry mass of the sample. Determine the
hygroscopic moisture using Equation A1.3.1.

(Mair-dry Mdry)
Wair-dry = x 100 [equation A1.3.1]
100
where:
Wair-dry = hygroscopic moisture content [% by mass]
Mair-dry = mass of air-dried material [g]
Mdry = mass of oven-dried material [g]

A1.3.1.7 Sample quantities

The guidelines shown in the following table should be used to estimate the quantity of material required for the
respective tests:

Mass of sample
Test
required (kg)

Moisture/density relationship (modified AASHTO T180) 40

Determination of active filler requirement (100 mm specimens) 60

Optimum bitumen addition indication (100 mm specimens) 80

Optimum bitumen addition determination (150 mm specimens ) 100

Standard soil tests (gradings, Atterberg Limits, moisture content, etc.) 20

A1.3.2 Active filler requirements

A1.3.2.1 Effect of plasticity

Bitumen emulsion stabilisation is normally carried out in combination with a small amount (1% by mass)
of active filler (cement or hydrated lime) to enhance bitumen adhesion and reduce moisture susceptibil-
ity. The Plasticity Index (PI) of the material is normally used as a guideline for the use of hydrated lime or
cement in the mix:

Plasticity Index: < 10 Plasticity Index: > 10

Carry out ITS tests on 100 mm specimens

Pre-treat the material with hydrated lime (ICL value)
to determine the need to add either cement
(The initial consumption of lime (ICL value) must
or hydrated lime, as described in Section
first be determined using the appropriate pH test.)
A1.3.2.2 below..

Pre-treatment of material with a PI > 10 requires that the lime and water be added at least 2 hours prior to
the addition of the bitumen emulsion. The pre-treated material is placed in an air-tight container to retain
moisture. The moisture content is then checked and, if necessary, adjusted prior to adding the bitumen
stabilising agent (as described in Section A1.3.4).

Note:

Where the material is pre-treated with lime,

the following tests for the Determination of
Active Filler Requirements described under
Section A1.3.2.2 below are not necessary..

274/275
A1.3.2.2 Determination of active filler requirements

Where the PI < 10, the need for an active filler and the type of active filler (cement or hydrated lime) that
is appropriate for the material must first be determined by carrying out ITS tests on 100 mm diameter
specimens for three different mixes made from the same sample. The amount of bitumen emulsion added
to each of the three mixes is constant for each, using the fractions passing the 4.75 mm and 0.075 mm
sieves as a guideline, as shown in the following table:

Guidelines for estimating optimum bitumen emulsion addition (60% residual bitumen)

Bitumen emulsion (Residual bitumen) addition

(% by mass of dry aggregate)
Fraction passing
Fraction passing 4.75 mm sieve Typical type of material
0.075 mm sieve (%)

< 50% > 50%

<4 3.3 (2.0) 3.3 (2.0) Recycled asphalt (RA/RAP)

47 3.7 (2.2) 4.0 (2.4)

RA/Graded crushed stone/
Natural gravel/blends
7 10 4.0 (2.4) 4.7 (2.8)

> 10 4.3 (2.6) 5.3 (3.2) Gravels/sands

The first of the three mixes contains no active filler, 1% cement is added to the second mix and 1%
hydrated lime is added to the third mix, all three mixes being treated with the same amount of bitumen
emulsion. Material from each of the three mixes is used to manufacture 100 mm diameter specimens that
are cured and tested to determine the relevant ITSDRY and ITSWET values (described in Sections A1.3.4 to
A1.3.8 below). The Tensile Strength Retained (TSR) value is then used as the primary indicator for whether
an active filler is required.

Where the TSR value for the mix with no active filler added is in excess of 60%, the mix design should be
undertaken with no active filler. (This situation is usually confined to materials consisting of good quality
crushed stone, often including a significant proportion of reclaimed asphalt (RAP) material.)
Where the TSR value of the mix with no active filler added is less than 60%, the mix with the type of ac-
tive filler that produces a significantly higher TSR value (> 5%) indicates a preference for either cement or
hydrated lime and should be used in the following mix designs. If the TSR values for both active fillers are
of the same order (difference < 5%) then either type of active filler is suitable.

Note:

To determine the sensitivity of the active filler,

additional tests on 100 mm diameter speci-
mens may be undertaken using the preferred
active filler at a lower application rate (e.g.
0.75%). However, to avoid compromising the
flexibility of the mix, the maximum allowable
application rate for active filler is 1.0% and
should only be exceeded when hydrated lime
is applied as a pre-treatment to eliminate
plasticity.

A1.3.3 Determine the fluid/density relationship

The Optimum Fluid Content (OFC) and the Maximum Dry Density (MDD) of the material stabilised with
bitumen emulsion is determined using standard compaction effort.

The OFC for bitumen emulsion stabilised material is the percentage by mass of bitumen emulsion plus
additional moisture required to achieve the maximum dry density in the treated material. As described
below, the OFC is determined by adding a constant percentage of bitumen emulsion whilst varying the
amount of water added.

Step 1. M
 easure out the bitumen emulsion as a percentage by mass of the air-dried material for each of
five prepared samples (following the procedure described in Section A1.3.1.2). The percentage
of bitumen emulsion added is normally between 2 and 3% residual bitumen (e.g. for 3% residual
bitumen, add 5% of a 60% bitumen emulsion).

Step 2. T
 he bitumen emulsion and water is added to the material and mixed until uniform immediately
prior to compaction.

276/277
Step 3. Determine the OFC and MDD for the stabilised material in accordance with the modified moisture-
density relationship test procedure (AASHTO T-180).

A1.3.4 Treating the sample with bitumen emulsion

Prepare the sample and treat with bitumen emulsion using the following procedure:

Step 1. Place the required mass of sample (between 20 kg and 30 kg, prepared as described in Section
A1.3.1 above) into the Wirtgen WLM 30 pugmill mixer.
Step 2. Determine the dry mass of the sample using Equation A1.3.2.

(Mair-dry)
Msample =

( (
Wair-dry
((
[equation A1.3.2]
1+
100

where:
Msample = dry mass of the sample [g]
Mair-dry = air-dried mass of the sample [g]
Wair-dry = moisture content of air-dried sample [% by mass]

Step 3. D
 etermine the required mass of active filler (lime or cement) to be added using Equation A1.3.3.

Cadd
Mcement = x Msample [equation A1.3.3]
100
where:
Mcement = mass of lime or cement to be added [g]
Cadd = percentage of lime or cement required [% by mass]
Msample = dry mass of the sample [g]

Step 4. Determine the amount of bitumen emulsion to be added using Equation A1.3.4.

RBreqd
Memul = x Msample [equation A1.3.4]
PRB
where:
Memul = mass of bitumen emulsion to be added [g]
RBreqd = percentage of residual bitumen required [%]
PRB = percentage residual bitumen in emulsion [% by mass]
Msample = dry mass of the sample [g]

278/279
Step 5. Determine the amount of water to be added to achieve the OFC of the material using
Equation A1.3.5

Mwater =

( (W OFC Wairdry)
100 (
x Msample Memul
[equation A1.3.5]

where:
Mwater = mass of water to be added [g]
WOFC = optimum fluid content [% by mass]
Wair-dry = moisture content of air-dried sample [% by mass]
Msample = dry mass of the sample [g]
Memul = mass of the bitumen emulsion added [g]

Step 6. Mix the material, active filler, bitumen emulsion and water in the mixer until uniform.

Step 7. Transfer the bitumen emulsion treated material into an air-tight container and immediately seal.
To minimise moisture loss, manufacture the test specimens as soon as possible by following the
relevant procedure for either 100 mm or 150 mm diameter specimens, as described in sections
A1.3.5.1 and A1.3.5.2 respectively.

Repeat the above steps for at least four mixes with different residual bitumen contents at 0.2%
intervals.
The Guidelines for estimating optimum bitumen emulsion addition (Section A1.3.2.2 above) should be
used in determining the mid-point of the range of bitumen emulsion to be added to the four samples.

An example. If the material consists of a blend of RAP and crushed stone with 39% and 8% passing
the 4.75 mm and 0.075 mm sieves respectively, the guidelines indicate an optimum bitu-
men emulsion (residual bitumen) addition of 4.0 (2.4)%. The amount of bitumen emulsion
(residual bitumen) to be added to each sample (all with the same amount of active filler and
at the optimum fluid content) is:

Sample 1: 3.5 (2.1)%

Sample 2: 3.8 (2.3)%
Sample 3: 4.2 (2.5)%
Sample 4: 4.5 (2.7)%
A1.3.5 Manufacture of specimens for testing

The procedures described below are for the manufacture of two different sizes of specimen using different
compaction procedures:

Specimen size and compaction effort applied in the manufacturing process

Specimen diameter Specimen height Compaction effort

100 mm 63.5 mm Modified Marshall*
150 mm 95.0 mm Modified AASHTO
* 75 blows per face

The following two questions are often raised:

1. Which specimen size should be manufactured? As described in Section A1.3.8 below, ITSDRY and
ITSWET values are normally determined from 100 mm diameter specimens. 150 mm diameter speci-
mens may be substituted for 100 mm diameter specimens to obtain the same values. However,
when dealing with coarse material, (i.e. where the grading curve tends towards the coarse side of
the recommended grading envelop) it is strongly recommended that 150 mm diameter specimens
are manufactured and tested in place of the smaller 100 mm diameter specimens.
Note. Only 150 mm diameter specimens are used to determine ITSEQUIL and ITSSOAK values.

2. Can other compaction methods be used? The compaction procedures described below are well
known standard procedures that can be carried out in most laboratories, worldwide. Other pro-
cedures may be used (e.g. gyratory compaction, vibrating hammer, vibrating table, etc.) provided
they achieve the same density target of 100% Marshall compaction for the 100 mm diameter
specimens or 100% of the mod AASHTO T-180 density for 150 mm diameter specimens.

A1.3.5.1 Manufacture of 100 mm diameter specimens

A minimum of six (6) 100 mm diameter specimens, 63.5 mm in height, are manufactured from each
sample of treated material by applying modified Marshall compaction effort, as described in the following
steps:

280/281
Step 1 Prepare the Marshall mould and hammer by cleaning the mould, collar, base-plate and face of the
compaction hammer. Note: the compaction equipment must not be heated but kept at ambient
temperature.

Step 2. Weigh sufficient material to achieve a compacted height of 63.5 mm 1.5 mm (Approximately
1,100 g for most materials). Poke the mixture with a spatula 15 times around the perimeter and
10 times on the surface, leaving the surface slightly rounded.

Step 3. Compact the mixture by applying 75 blows with the compaction hammer. Care must be taken to
ensure the continuous free fall of the hammer.

Step 4. Remove the mould and collar from the pedestal, invert the specimen (turn over). Replace it and
press down firmly to ensure that it is secure on the base plate. Compact the other face of the
specimen with a further 75 blows.

Step 5. After compaction, remove the mould from the base-plate and extrude the specimen by means
of an extrusion jack. Measure the height of the specimen and adjust the amount of material if the
height is not within the 1.5 mm limits.

Note: Coarse materials are often damaged during the extrusion process. It is therefore recom-
mended that the specimens are left in their moulds for 24 hours allowing sufficient strength
to develop before extruding.

Repeat steps 1 to 5 for the manufacture of at least six (6) specimens.

Step 6. Take 1 kg representative samples after compaction of the second and fifth specimen and dry to
a constant mass. Determine the moulding moisture using Equation A1.3.6.

(Mmoist Mdry)
Wmould = x 100 [equation A1.3.6]
Mdry
where:
Wmould = moulding moisture content [% by mass]
Mmoist = mass of moist material [g]
Mdry = mass of dry material [g]
A1.3.5.2 Manufacture of 150 mm diameter specimens

A minimum of six (6) 150 mm diameter specimens, 95 mm in height, are manufactured from each sample
of treated material by applying modified AASHTO (T-180) compaction effort, as described in the following
steps:

Step 1. P
 repare the equipment by cleaning the mould, collar, base-plate and face of the compaction
rammer. (Either split-moulds or standard Proctor moulds may be used, each fitted with a
32 mm spacer placed on the base plate to achieve specimens that are 95 mm (1.5 mm) in
height.)

Note: the modified AASHTO compaction rammer has the following specifications:

Rammer diameter: 50 mm
Mass: 4.536 kg
Drop distance: 457 mm

Step 2. C
 ompact each specimen applying modified AASHTO (T-180) compaction effort (4 layers approxi-
mately 25 mm thick, each receiving 55 blows from the drop rammer.)

Step 3. C
 arefully trim excess material from specimens, as specified in the AASHTO T-180 test method.

Step 4. A
 fter compaction, remove the mould from the base-plate and extrude the specimen by means
of an extrusion jack. Where split-moulds are used, separate the segments and remove the speci-
men.

Note: Coarse materials are often damaged during the extrusion process. It is therefore recom-
mended that the specimens are left in their moulds for 24 hours allowing sufficient strength
to develop before extruding.

Where split moulds are used, it is advisable to leave the specimen in the mould for 4 hours
before splitting the mould and extracting the specimen.

Repeat steps 1 to 4 for the manufacture of at least six (6) specimens.

Step 5. T
 ake 1 kg representative samples after compaction of the second and fifth specimen and dry to
a constant mass. Determine the moulding moisture content using Equation A1.3.6 (above).

282/283
A1.3.6 Curing the specimens

Two curing regimes are described below. The first is a standard procedure to dry the specimens to con-
stant mass. The second procedure aims to simulate field conditions where the equilibrium moisture con-
tent is approximately 50% of OMC. Both 100 mm and 150 mm diameter specimens may be cured dry (to
constant mass) whereas only 150 mm diameter specimens are cured at equilibrium moisture content.

A1.3.6.1 Curing dry

Place the specimens (either 100 mm or 150 mm diameter) in a forced-draft oven at 40C and cure to
constant mass (normally 72 hours).

To determine if constant mass has been achieved, weigh the specimens and place back in the oven. Re-
move after 4 hours and reweigh. If the mass is constant, then continue with testing. If the mass is not con-
stant, place back in the oven and reweigh after repeated 4 hour intervals until constant mass is achieved.

When constant mass is achieved, remove the specimens from the oven and allow to cool to 25C
( 2.0C).

A1.3.6.2 Curing to simulate field conditions

Place the 150 mm diameter specimens in a forced-draft oven at 30C for 24 hours (or until the moisture
content has reduced to approximately 50% of OMC).

Take the specimens out of the oven, place each in a sealed plastic bag (at least twice the volume of the
specimen) and place back in the oven at 40C for a further 48 hours.

Remove specimens from the oven after 48 hours and take out of their respective plastic bags, ensuring
that any moisture in the bags does not come into contact with the specimen. Allow to cool down to 25C
( 2.0 C).
A1.3.7 Preparing the specimens for testing

After cooling, determine the bulk density of each specimen using the following procedure:

Step 1. Determine the mass of the specimen.

Step 2. M
 easure the height of the specimen at four evenly-spaced locations around the circumference
and calculate the average height of the specimen.

Step 3. Measure the diameter of the specimen.

Step 4. Calculate the bulk density of the each specimen using Equation A1.3.7.

4 x Mspec
BDspec = x 1,000,000 [equation A1.3.7]
x d2 x h
where:
BDspec = bulk density of specimen [kg/m3]
Mspec = mass of specimen [g]
h = average height of specimen [mm]
d = diameter of specimen [mm]

Exclude from further testing any specimen whose bulk density differs from the mean bulk density of all six
(6) specimens by more than 2.5%.

Step 5. P
 lace half of the specimens (normally 3) under water in a soaking bath for 24 hours at 25C
( 2 C). After 24 hours, remove the specimens from the water, surface dry and test immediately.

284/285
A1.3.8 Determination of the indirect tensile strength (ITS) of specimens

The ITS of a specimen is determined by measuring

the ultimate load to failure applied to the dia
metrical axis at a constant deformation rate of
50.8 mm/minute. Ensure that the temperature
of the specimens is 25C ( 2 C) and follow the
procedure described below:

Step 1. Place the specimen onto the ITS jig. (Ensure the correct loading strips are appropriate for the di-
ameter of the specimen.) Position the sample such that the loading strips are parallel and centred
on the vertical diametrical plane.

Step 2. Place the transfer plate on the top bearing strip and position the jig assembly centrally under the
loading ram of the compression testing device.

Step 3. Apply the load to the specimen, without shock, at a rate of advance of 50.8 mm per minute until
the maximum load is reached.

Step 4. Record the maximum load P (in kN), accurate to 0.1 kN.

Step 5. Record the displacement at break to the nearest 0.1 mm.

Step 6. Break the specimen in half and record the temperature of the specimen at its centre.
Step 7. B
 reak up one of the unsoaked specimens and dry to a constant mass. Determine the cured mois-
ture content using Equation A1.3.8.

(Mmoist Mdry)
Wspec = x 100 [Equation A1.3.8]
Mdry
where:
Wspec = moisture content of specimen [% by mass]
Mmoist = mass of moist material [g]
Mdry = mass of dry material [g]

Determine the dry density of each specimen using Equation A1.3.9

Wspec
DDspec = BDspec x 1
100
[Equation A1.3.9]
( (
where:
DDspec = dry density of specimen [% by mass]
BDspec = bulk density of specimen [g]
Wspec = moisture content of specimen [g]

Step 8. B
 reak up one of the soaked specimens and dry to a constant mass (at 105C to 110C).
Determine the after-soaking moisture content using Equation A1.3.8.

286/287
Step 9. Calculate the ITS value for each specimen to the nearest 1 kPa using Equation A1.3.10.

2xP
ITS = x 1,000,000 [Equation A1.3.10]
xhxd
where:
ITS = Indirect Tensile Strength [kPa]
P = maximum applied load [kN]
h = average height of the specimen [mm]
d = diameter of the specimen [mm]

Step 10. Use the Worksheet in Annexure A1.3.1 to record the data and the form in Annexure A1.3.2 to
report the results.

Step 11. For dry-cured specimens only, calculate the Tensile Strength Retained (TSR) value for the sample
using Equation A1.3.11.

Ave ITSWET
TSR = x 100 [Equation A1.3.11]
Ave ITSDRY
where:
TSR = Tensile Strength Retained [%]
Ave ITSWET = average ITSWET value [kPa]
Ave ITSDRY = average ITSDRY value [kPa]

Note. T
 o differentiate between the results obtained from the different curing regimes, the terminol-
ogy shown in the table below should be adopted to avoid any confusion.
 Term Specimen diameter Curing regime Moisture content
ITSDRY 72 hrs unsealed <1%
100 mm or 150 mm
ITSWET 24 hrs soak Saturated

24 hrs unsealed,
ITSEQUIL 50% of OMC
150 mm only 48 hrs in sealed bag

ITSSOAK 24 hr soak Semi-saturated

A1.3.9 Interpretation of the indirect tensile strength (ITS) test results

Plot the respective soaked and unsoaked ITS test results against the relevant addition of bitumen emul-
sion, as shown in the example below. If results for two different curing regimes were obtained, each must
be plotted on a separate graph.

The added bitumen emulsion that meets the minimum ITS value for the required material classification is
selected as the primary indicator for the minimum amount of bitumen emulsion to be added. Engineering
judgement is then used to determine the amount of bitumen emulsion that needs to be added in order
to achieve sufficient confidence, based on the variability of the test results (the goodness of fit of the
regression curve or line through the plotted test results).

The example below explains the process.

The table and plot of ITS test results are typical ITS values achieved from a mix design using 100 mm
diameter specimens for a RAP/crushed stone blended material stabilised with a 60% residual bitumen
emulsion.

288/289
 Added bitumen emulsion
ITSDRY (kPa) ITSWET (kPa) TSR (%)
(residual bitumen) (%)
2.8 (1.7) 204 109 53.4
3.2 (1.9) 234 151 64.5
3.5 (2.1) 282 214 75.9
3.8 (2.3) 327 253 77.4

ITSDRY
Indirect tensile strength (ITS) (kPa)

300

250
Class 1 ITSWET
MIN ITSDRY
200

150

Class 1
100
MIN ITSWET

2.8 (1.7) 3.2 (1.9) 3.5 (2.1) 3.8 (2.3)

Bitumen emulsion (residual bitumen) added (%)

100 mm specimens

The curve through the four ITSDRY points approximates the relationship between ITSDRY and added bitu-
men emulsion (residual bitumen). The line through the four ITSWET points approximates the relationship be-
tween ITSWET and added bitumen. The fine dotted lines indicate that the addition of at least 3.0 (1.8)% of
bitumen emulsion (residual bitumen) is required to meet the requirements for a Class 1 bitumen stabilised
material (ITSDRY > 225 kPa and ITSWET > 100 kPa). Based on these ITS test results and the correspond-
ing TSR values, the level of confidence may be increased by selecting a bitumen emulsion addition that
meets the requirements. For example, selecting 3.5 (2.1)% bitumen emulsion (residual bitumen) addition
will significantly increase the probability of the stabilised material achieving the minimum ITSDRY and ITSWET
values under field conditions.
Annexure A1.3.1

BITUMEN EMULSION MIX DESIGN - WORKSHEET

Project Date

Sample / Mix No.: Location

Material description :
Maximum dry density Optimum fluid content
Percentage < 0.075mm Grading: Coarse Medium Fine
Plasticity Index

Emulsion Source Emulsion type

Residual bitumen (%)
Active Filler Type Filler Source

MOISTURE DETERMINATION Specimen manufacture After Curing

Hygroscopic Sample 1 Sample 2 Dry Soaked
Pan No.
Mass wet sample + pan m1
Mass dry sample + pan m2
Mass pan mp
Mass moisture m1-m2 = Mm
Mass dry sample m2-mp= Md
Moisture content Mm/Mdx100=Mh

Mass of air-dried sample placed in the mixer (kg)

Percentage of water added to sample for mixing: Amount of water added :
Total percentage water added: Total water added:

Bitumen emulsion addition (%): Active filler addition (%):

Residual bitumen addition (%)
Temperatures (C) Material: Emulsion: Water:

SPECIMEN DETAILS
Specimen ID
Date Moulded
Date removed from oven
Date tested
Diameter (mm)

Individual height
measurements (mm)

Average height (mm)

Mass after curing (g)
Bulk density (kg/m 3 )
Average bulk density
Dry density (kg/m 3 )

ITS TEST
Specimen condition Unsoaked (ITSDRY / ITS EQUIL ) Soaked (ITS WET / ITSSOAK )
Maximum load (kN)
Internal temperature (C)
Deformation (mm)
ITS (kPa)
Average ITS (kPa)
TSR (%)

290/291
Annexure A1.3.2

BITUMEN EMULSION MIX DESIGN REPORT (Dry curing)

Project Date

Sample number: Location

Material description :
Maximum dry density Optimum fluid content
Percentage < 0.075mm Grading: Coarse Medium Fine
Plasticity Index
Bitumen Emulsion Type
Emulsion Supplier Residual Bitumen (%)
Active Filler Type Filler Source

BITUMEN EMULSION STABILISED MATERIAL SPECIMENS

Compactive effort mm specimen diameter
Date moulded
Date tested
Bitumen emulsion added (%)
Residual bitumen added (%)
Active filler added (%)
Moulding moisture content (%)

TEST RESULTS
ITS DRY (kPa)
Moisture content at break (%)
Dry Density (kg/m 3)
Average deformation (mm)
Temperature at break (C)

ITS WET (kPa)

Moisture content at break (%)
Dry Density (kg/m 3)
Average deformation (mm)
Temperature at break (C)

Tensile Strength Retained (%)

Material classification

% Residual Bitumen vs ITS % Residual Bitumen vs Dry density

420
2150
370
Dry density (kg/m3)

320 2125
ITS (kPa)

270
2100
220
2075
170
120 2050
1,75 2 2,25 2,5 1,75 2 2,25 2,5
Dry Specimens
ITS dry ITS wet Residual Bitumen added Residual Bitumen added
Wet specimens

Comments
A1.4 Testing field samples of bitumen stabilised
materials (BSMs)

A1.4.1 Field sampling

Samples ( 100 kg) of treated material are to be obtained on site:

> Where the material is mixed in situ, the sample is taken immediately behind the recycler from the full
thickness of treated material (before it is compacted).
> Where the material is plant mixed and placed by paver or grader, the samples may be taken either at
the mixing plant or from site following standard sampling procedures.

Place each sample in an air-tight container and seal to prevent moisture loss. The container or bag used
for sampling must be sufficiently large to ensure that the material remains loose and does not compact
inside the container.

Field samples are to be transported to the laboratory within two hours of sampling and the test specimens
manufactured within four hours of being sampled.

A1.4.2 Sample preparation

Prepare the sample by passing through a 19 mm sieve. Discard the fraction retained on the 19 mm sieve.
Place the sample in an air-tight container and check that the temperature is at a temperature of between
22 C and 25 C. If the temperature of the material is not within this temperature range, place the entire
sample in an air cabinet (or similar) until the material is within this range.

A1.4.3 Adjust the moisture content

Since field samples normally lie in the range of 60% to 80% of Optimum Moisture Content (OMC), suffi-
cient water must be added to the sample to bring it to OMC before the test specimens are manufactured.
(Due to the variability of recycled material, it is seldom that the OMC will be known with certainty. The
amount of water required to bring the sample to OMC must usually be determined (for each and every
sample) using the procedure described in Section A1.4.3.2 below.)

292/293
A1.4.3.1 Where the Maximum Dry Density (MDD) and Optimum Moisture Content
(OMC) of the material is known with certainty

> Determine the moisture content of the field sample;

> Adjust the moisture content of 10 kg of the sample to achieve OMC, mix thoroughly and place in an
air-tight container; then
> Proceed to Section 4 and manufacture 100 mm diameter specimens for testing.

A1.4.3.2 Where the Maximum Dry Density (MDD) and Optimum Moisture Content
(OMC) of the material is not known

Adjust the moisture content of the sample to bring it to the OMC.

Step 1. Prepare the Marshall mould and hammer by cleaning the mould, collar, base-plate and hammer
face. (Note: the compaction equipment must not be heated but kept at ambient temperature.)

Step 2. Weigh out 1,100 g of material and place in the mould. Poke the mixture with a spatula 15 times
around the perimeter and 10 times on the surface, leaving the surface slightly rounded. Ensure
that the material does not lose moisture or segregate during placing in the mould and seal the
remaining sample in the air-tight container.

Step 3. Compact the material by applying 75 blows with the compaction hammer. Care must be taken to
ensure the continuous free-fall of the hammer. Remove the mould and collar from the pedestal,
invert the specimen (turn over). Replace it and press down firmly to ensure that it is secure on the
base plate. Compact the other face of the specimen with a further 75 blows.

Step 4. After compaction, remove the mould from the base-plate and extrude the specimen by means of
an extrusion jack.

Step 5. Determine the mass of the specimen.

Step 6. Measure the height of the specimen at four evenly-spaced locations around the circumference
and calculate the average height of the specimen.

Step 7. Measure the diameter of the specimen.

Step 8. C
 alculate the interim bulk density of the specimen using equation A1.4.1:

4 x MSPEC
BDINT = x 1 000 000 [Equation A1.4.1]
x d2 x h
where:
BDINT = interim bulk density [kg/m]
MSPEC = mass of specimen [g]
h = average height of specimen [mm]
d = diameter of specimen [mm]

Then, using Equation A1.4.2, calculate the true bulk density by excluding the total amount of
moisture that was added:

BDINT
BD = x 100

100 ( W
100 + ADD ( [Equation A1.4.2]

where:
BD = bulk density [kg/m]
BDINT = interim bulk density (from Equation 1) [kg/m]
WADD = total moisture added to sample [%]

Step 9. P
 lot the bulk density against water addition. (The first point on the graph is zero water addition.)

Step 10. Weigh out another 1,100 g of material. Add 5.5 ml of water (0.5% by mass) and mix thoroughly
before placing in the mould, following the procedure described in Step 2.

Step 11. Repeat Steps 3 to 9 and compare the bulk density with that achieved from the previous speci-
men. Continue to follow Steps 3 to 10 making specimens with additional water added until the
bulk density reduces from the previous specimen.

294/295
Step 12. The amount of water added to the specimen that returns the highest bulk density (in percent)
is then added to 10 kg of the remaining sample. Thoroughly mix the material and return to the
air-tight container.

Note. If moisture is observed seeping out of the mould before a turning point in the curve is
achieved, then the amount of water addition required to achieve maximum density is the
amount of water added to the material at which seepage occurred, less 0.5%.

A1.4.4 Manufacture of 100 mm diameter specimens for testing

Using specimen height measurements from Step 6 (Section A1.4.3 above) as a guideline, determine the
mass of material required to achieve a specimen height of 63.5 mm (1.5 mm). Then, follow the procedure
described in Steps 2 to 4, using the revised mass of material to manufacture six (6) specimens, each with
a height of 63.5 mm (1.5 mm).

Note. Where the OMC of the sample is known and the moisture adjustment procedure described above
in Section 3 was not carried out, follow steps 2 and 3 (Section A1.4.3) using 1,100 g of sample
prepared at OMC. Measure the height of the specimen (Step 6) and, if necessary, adjust the mass
to achieve a specimen height of 63.5 mm (1.5 mm). Then, follow the procedure described in Steps
2 to 4, using the revised mass of material to manufacture six (6) specimens, each with a height of
63.5 mm (1.5 mm).

After all specimens have been manufactured, use the remaining sample material to determine the mould-
ing moisture content following standard oven drying procedures.

A1.4.5 Curing the specimens

Place the specimens in a forced-draft oven at 40 C and cure to constant mass (normally 72 hours).

To determine if constant mass has been achieved, weigh the specimens and place back in the oven.
Remove after 4 hours and reweigh. If the mass is constant, then continue with testing. If the mass is not
constant, place back into the oven and reweigh at 4 hour intervals until constant mass is achieved.

After constant mass has been achieved, remove the specimens from the oven and allow to cool to 25 C
( 2.0 C).
After cooling, determine the bulk density of each specimen following Steps 5 to 8 described in Section
A1.4.3.

Exclude from further testing any specimen whose bulk density differs from the mean bulk density of all six
(6) specimens by more than 2.5%.

Half of the specimens (3) are then placed under water in a soaking bath for 24 hours at 25 C ( 2 C).
After 24 hours, remove the specimens from the water, surface dry and test immediately.

A1.4.6 Determination of Indirect Tensile Strength (ITS)

The ITS is determined by measuring the ultimate load to failure of a specimen that is subjected to a
constant deformation rate of 50.8 mm/minute on its diametrical axis. Ensure that the temperature of the
specimens is 25 C ( 2 C) and follow the procedure described below:

Step 1. P
 lace the specimen onto the ITS jig. Position the sample such that the loading strips are parallel
and centred on the vertical diametrical plane.

Step 2. P
 lace the transfer plate on the top bearing strip and position the jig assembly centrally under the
loading ram of the compression testing device.

Step 3. A
 pply the load to the specimen, without shock, at a rate of advance of 50.8 mm per minute until
the maximum load is reached.

Step 4. R
 ecord the maximum load P (in kN), accurate to 0.1 kN.

Step 5. R
 ecord the displacement at break to the nearest 0.1 mm.

Step 6. B
 reak the specimen in half and record the temperature of the specimen at its centre.

Step 7. B
 reak up one of the dry and one of the wet specimens and determine the moisture content follow-
ing standard oven drying procedures.

296/297
Step 8. Calculate the ITS value for each specimen to the nearest 1 kPa using Equation A1.4.3:

2xP
ITS = x 1 000 000 [Equation A1.4.3]
xhxd
where:
ITS = Indirect Tensile Strength [kPa]
P = maximum applied load [kN]
h = average height of the specimen [mm]
d = diameter of the specimen [mm]

Then, using Equation A1.4.2, calculate the true bulk density by excluding the total amount of
moisture that was added:

Step 9. Use the form on the following page to report the results.
 5

ITS DETERMINATION for BSM FIELD SAMPLE

PROJECT Position
Sample No.: Date
Description :

FIELD TREATMENT
Maimum dry density (If known) Optimum moisture content
Bitumen type / source Bitumen applied (%)
Active filler type / source Active filler applied (%)

MOISTURE DETERMINATION Specimen manufacture After Testing

Field Moulding Dry Soaked
Pan No.
Mass wet sample + pan m1
Mass dry sample + pan m2
Mass pan mp
Mass moisture m1 - m2 = Mm
Mass dry sample m2 - mp = Md
Moisture content Mm / Md x 100 = Mh

SPECIMEN DETAILS
DRY SOAKED
Date Moulded
Date removed from oven
Specimen ID
Diameter (mm)

Individual Thickness
Readings (mm)

Avgerage Thickness (mm)

Mass after curing (g)

3
Bulk density (kg/m )
3
Avg bulk density (kg/m )

Moisture content (%)

Dry density (kg/m3)
3
Avg dry density (kg/m )

ITS TESTING

Condition Dry (ITSDRY) Soaked (ITSWET)

Date tested
Maximum load (kN)
Temperature (C)
Deformation (mm)
Tensile strength (kPa)
Avg tensile strength (kPa)

298/299
A1.5 Determining the strength of BSM core
specimens

Core samples may be extracted from the full thickness of the completed layer and tested for ITS values.
Due to the relatively low strength of a BSM, 150 mm diameter cores are preferable to the 100 mm
diameter cores normally extracted for HMA. Cores cannot be successfully extracted until the BSM has
developed sufficient strength and the delay period is dictated by the rate of moisture loss from the mate-
rial which is primarily a function of weather conditions and layer thickness. When conditions are warm
and dry, cores can usually be extracted from a 150 mm thick layer of BSM-foam after 14 days. The delay
period for BSM-emulsion is further influenced by the stability of the emulsion and delays of 30 days are
normal.

A1.5.1 Extracting core samples

The core barrel used to extract samples of BSM must be in a good condition. The amount of water added
whilst drilling should be kept to an absolute minimum and the rate of penetration kept sufficiently low to
prevent erosion and damage. After extraction, core samples must be wrapped individually in a soft cloth
and carefully packed for transporting to the laboratory.

A1.5.2 Cutting core specimens

Use a rotary saw fitted with a large diameter diamond-tipped blade to cut specimens from the portion of
the core that suffered least damage during extraction and handling. The thickness (height) of individual
specimens cut from the core is dictated by the diameter:

> 100 mm diameter cores: 63 mm

> 150 mm diameter cores: 95 mm

Where possible, more than one specimen should be cut from each core sample.

A1.5.3 Curing the core specimens

Cure the specimens to constant mass in a forced-draft oven at 40 C (normally 72 hours).

Where the ITSWET value is to be determined, place the specimens in a soaking bath for 24 hours.
A1.5.4 Determination of bulk density

Follow the procedures described in Section A1.2.7 for determining the bulk density of each core speci-
men.

A1.5.5 Determination of the Indirect Tensile Strength (ITS)

Follow the procedures described in Section A1.2.8 for testing the core specimens to determine the ITSDRY
and ITSWET values and the resulting TSR value. These values are then used to determine whether the
material has met the minimum specified requirements.

Note. Where the ITS results for specimens manufactured from field samples are in conflict with those
obtained from core specimens, the results for the core specimens should be taken as being the
correct values.

300/301
A1.6 Laboratory equipment requirements

A1.6.1 Laboratory equipment for soils testing

Description Quantity Description Quantity

Sample Preparation Moisture/density relationship

Riffler (25 mm openings) 1 Modified AASHTO (T-180)
Riffler pans 3 150 mm mould (including base plate, spacer 18
and collar)
Sieves 450 mm diameter
19.0 mm 1 Compaction rammer 1
(4.536 kg mass with 457 mm drop and 50 mm
13.2 mm 1
diameter)
4.75 mm 1
Electronic balance (12 kg 0.1 g) 1
20 litre air-tight containers 20
(Plastic buckets with lids) Mixing basin ( 0.5 m x0.5 m x 0.3 m) 1
50 kg mechanical balance 1 Mixing trowel 1
1 litre plastic measuring cylinder 1
Sieve Analysis (gradings)
Sieves 200 mm diameter Steel straight edge (for trimming) 1
50.0 mm 1
37.5 mm 1 Containers for moisture content 50
25.0 mm 1 (half liter capacity)
19.0 mm 1 Forced-draft drying oven 1
12.5 mm 1 (400 litre capacity)
9.5 mm 1
4.75 mm 1 Optional Equipment
2.36 mm 1 Mechanical compactor with rotating base plate 1
1.18 mm 1
0.60 mm 1
0.30 mm 1
0.15 mm 1
0.075 mm 3
Pan 1 California bearing ratio
Lid 1 150 mm moulds (including perforated base 30
Electronic balance (15 kg 0.1 g) 1 plate and surcharge weights)
Forced-draft drying oven 1 Compaction rammer (2.495 kg mass with 305 mm 1
(minimum 240 litre capacity) drop and 50 mm diameter)
Pans (300 mm ) 10
Swell gauge 1
Sieve brush 1
Soaking bath (2 m x 1 m x 0.4 m) 1
Optional Equipment
Mechanical Sieve Shaker 1 Compression testing machine 1

Atterberg Limits (plasticity)

Casagrande Liquid Limit Device 1
Grooving tool 1
Mixing bowls (100 mm ) 2
Spatula 1
Wash bottle (250 ml) 1
Timer 1
Glass pane (300 mm x 300 mm) 1
Glass jars (100 ml) 50
Drying oven (use oven in Sieve Analysis section) 1
A1.6.2 Additional laboratory equipment cement (or lime) stabilisation

Note: This equipment is required IN ADDITION to the list shown under A1.4.1,
Laboratory equipment for soils testing

Description Quantity Description Quantity

Compaction Ancillary Equipment

150 mm steel split moulds & collar 3 Teltru Thermometer (250C) 3
(for bath and ovens)
Shovel (300 mm) 1
Specimen curing Gloves (heat resistant) 1
250 mm x 350 mm perforated trays 12 500 mm paint brushes 2
Forced-draft drying oven 1 Broom with handle (soft) 1
(400 litre capacity)
Hammer (2 kg) 1
Plastic bags ( 10 litre) 500
Silicon grease (100 g) 1
Electronic balance (10 kg 0.1 g) 1
Rags for cleaning 1
Temperature controlled waterbath 1
(Note. The CBR soaking bath may be used if the String 1
ambient temperature is constant at 25 deg. C) Marker pen (or paint) 1
(for marking specimens)
2
Grain scoop or similar
Indirect Tensile Strength Testing
Electronic thermometer 1
ITS test jig for 150 mm specimens 1
Hand cleaner 1
Compression testing machine * 1
(loading rate 50.8 mm/min)

Unconfined Compressive Strength Testing * Note. The same compression testing machine
150 mm load transfer plate 1 as used in the CBR test may be used if
equipped with adjustable loading rate
Compression testing machine * 1
(loading rate 153 kN/min)

302/303
A1.6.3 Additional laboratory equipment for bitumen stabilisation

Note: T
 his equipment is required IN ADDITION to the list shown under A1.4.1,
Laboratory equipment for soils testing

Description Quantity Description Quantity

Foamed bitumen mix designs only Indirect Tensile Strength Testing

Wirtgen WLB10 S Laboratory unit 1 ITS test jig for 100 mm specimens 1
(complete with air compressor)
ITS test jig for 150 mm specimens 1
Compression testing machine* 1

Mechanical mixing equipment Ancillary Equipment

Wirtgen WLM 30 pugmill mixer 1 Teltru Thermometer (250C) 3
(for bath and ovens)
Shovel (300 mm) 1
Gloves (heat resistant) 1
500 mm paint brushes 2

100 mm specimen manufacture Broom with handle (soft) 1

Marshall compactor (Manual or automatic with 1 Hammer (2 kg) 1

wooden pedestal & hammer) Silicon grease (100 g) 1
100 mm moulds (with collar and base plate) 24 Rags for cleaning 1
Extrusion jack 1 String 1
Vernier calipers (25 mmm) 1 Marker pen(or paint) 1
(for marking specimens)
Grain scoop or similar 2
Electronic thermometer 1
Hand cleaner 1

Specimen curing * Note. The same compression testing machine

as used in the CBR test may be used if
250 mm x 350 mm perforated trays 12
equipped with adjustable loading rate
Forced-draft drying oven 1
(400 litre capacity)
Plastic bags ( 10 litre) 500
Electronic balance (10 kg 0.1 g) 1
Temperature controlled waterbath
(Note. The CBR soaking bath may be used if the 1
ambient temperature is constant at 25 deg. C)
304/305
Appendix 2 D
 etermining structural capacity
from traffic information

A2.1 Terminology associated with traffic 308

A2.2 Traffic loading classification 309
A2.3 Traffic loading estimates 311
A2.3.1 Traffic counts 313
A2.3.2 Static or dynamic weighing procedures 314
A2.4 Determination of design traffic (structural capacity) 315
A2.5 Practical approach to estimating design traffic 317

306/307
A2.1 Terminology associated with traffic

There are three key terms/abbreviations used to describe the traffic that uses a road and these need to
be clearly understood to avoid confusion.

> Annual average daily traffic (AADT) measured in vehicles per day. Sometimes abbreviated to average
daily traffic (ADT). This is a measure of the total daily volume of traffic using a road. It includes all traffic
travelling in both directions and takes no cognisance of the different types of vehicles (cars, trucks) that
make up the traffic spectrum, nor the number of traffic lanes. Although AADT is widely used to describe
traffic volumes, it is not a very useful measure for structural design purposes. An understanding of the
traffic spectrum and split between lanes is essential for determining the pavement structure required.

> Equivalent standard axle load (ESAL). The loading of heavy vehicles is always governed by legislation
and road pavements are designed accordingly. The term legal axle load usually defines the maximum
load permitted on a single axle. This varies from country to country, typically 80 kN to 130 kN.
For pavement design purposes, the axle configuration of a vehicle is also important in determining the
load applied in terms of Equivalent Standard Axle Loads (ESALs) with the standard axle defined
(e.g. 80 kN).

Pavements are designed to carry a certain number of ESALs. This is called the Structural Capacity of a
pavement and is usually expressed in millions (e.g. 5 x 106 ESALs).

> Average daily equivalent (ADE) traffic. This is the most useful information for pavement design since
it defines the number of equivalent standard axle loads that are currently using the road per lane.
Determining this key number is discussed below.
A2.2 Traffic loading classification

Pavements are classified by the number of ESALs that the road is designed to carry during its service life
(i.e. the Structural Capacity). This introduces a time frame and requires that a design life be defined.
Road authorities normally expect a return on their investment in a pavement and, typically, periods rang-
ing from 5 to 30 years are used in such calculations. This return period is then used to define the design
life of the pavement. Predicting the design traffic (or the number of ESALs expected during that period) is
therefore most important as the cost implications of inaccurate data is obvious in terms of layer numbers,
thickness and material composition.

Table A2.1 Typical pavement classification

Class ESALs x 106

T0 < 0.3

T1 0.3 1.0

T2 1.0 3.0

T3 3.0 10.0

T4 10.0 30.0

T5 30.0 100.0

Many different classification systems exist to describe traffic loading. The often-used terms light/me-
dium/heavy are too subjective and cannot be used for pavement design.
A system that classifies the traffic into loading ranges is normally adopted, such as the one illustrated in
Table A2.1 that has been adopted as a regional guideline for all Southern African countries.

308/309
Translating traffic counts into useful design information requires converting data collected on the traffic
spectrum (discussed below) into ESALs. Table A2.2 may be used as a rough guide to determine the
number of ESALs that will be applied to the road surface by different types of heavy vehicles. It should
be noted that light vehicles are not assigned an ESAL factor and are therefore of no consequence from a
pavement design perspective. The ESALs per vehicle data presented in Table A2.2 were derived from sur-
vey data collected in South Africa and may not be representative for other countries with different vehicle
types and traffic spectra. This information therefore needs to be obtained from the relevant road authority
(where available) or from a counting exercise.

Table A2.2 Example of typical ESALs per heavy vehicle

Vehicle type Normal Range Average

2-axle truck 0.3 1.1 0.70

2-axle bus 0.4 1.5 0.73

3-axle truck 0.8 2.6 1.70

4-axle truck 0.8 3.0 1.80

5-axle truck 1.0 3.0 2.20

6-axle truck 1.6 5.2 3.50

7-axle truck 3.8 5.0 4.40

Average 2.5 6.0

Ultra-heavy classes are also defined for extra heavy-duty pavements, such as those constructed for
major runways and mine haul roads. Such pavements, however, are beyond the scope of this manual and
should be regarded as special application pavements.
A2.3 Traffic loading estimates

Available data is used as a basis for estimating the ADE of existing traffic loading. Where only AADT
figures are available, the formula given in equation A2.1 below can be used to provide an initial estimate
of the ADE.

ADE = AADT x fH x f E x fL x fG x fW [equation A2.1]

where:
fH = percentage heavy vehicles in the traffic spectrum;
fE = estimated average ESAL per heavy vehicle; and
fL = lane distribution factor (see Table A2.3)
fG = gradient factor (see Table A2.4)
fW = lane width factor (see Table A2.5)

Determining the relevant lane distribution factor requires careful consideration, especially for multi-lane
divided highways where the slow-lane invariably carries a higher number of heavy vehicles than the mid-
dle- or fast-lanes.

Table A2.3 Lane distribution factor (example) fL

Total lanes
Slow lane Middle lane Fast lane
(Both directions)

2 0.5

4 0.48 0.15

6 0.35 0.3 0.13

310/311
Table A2.3 gives guidelines based on traffic patterns. These factors are likely to vary from country to
country as they typical reflect driver habits, law enforcement practices, etc. The lane-split data in Table
A2.3 reflects normal traffic patterns for inter-city highways. Roads that are predominantly unidirectional
haul routes (e.g. farm-to-market and mine-to-railhead roads) will obviously not fit this pattern.

The gradient factor takes cognisance of the increase in effective wheel loading due to the reduction
in speed of heavy vehicles crawling up steep slopes whilst the lane width factor recognises the load
concentrating effect of confining traffic. Table A2.4 and A2.5 include the factors recommended by the
German highway authorities in their publication RStO 01

Table A2.4 Gradient factor Table A2.5 Lane width factor

Gradient (%) fG Lane width FW

Less than 2 1.0 Less than 2.5 m 2.0

2 to 4 1.02 2.5 m to 2.75 m 1.8

4 to 5 1.05 2.75 m to 3.25 m 1.4

5 to 6 1.09 3.25 m to 3.75 m 1.1

6 to 7 1.14 More than 3.75 m 1.0

7 to 8 1.20

8 to 9 1.27

9 to 10 1.35

More than 10 1.45

(The design codes in some countries exclude the Lane Width Factor since channelised flow is assumed.
Fw is therefore a constant 1.0)

Estimates based on AADT, average percentage heavy vehicles and average number of ESALs per heavy
vehicle (FE) should be treated with care. They are purely estimates and should not be used as primary
input for the design of major pavements. When there is any doubt about the accuracy or relevance of the
traffic data available, a detailed traffic survey should be undertaken to determine the correct ADE per
lane. Various survey methods may be employed, including:
A2.3.1 Traffic counts

These are normally carried out at specific points (such as intersections, etc) over a 12-hour, 18-hour or
24-hour period. The actual count may then be converted to 24-hour periods by applying appropriate
factors to obtain the AADT, where required. The number of heavy vehicles is normally expressed as a
percentage of AADT.

Traffic counts should include the total number of vehicles per day travelling in each lane (in each direction)
separated into vehicle types. The road category (e.g. major highway or rural farm access road) and avail-
able technology will normally dictate whether the count is conducted electronically. Sophisticated meth-
ods are currently available for counting the number of axles (and even weighing-in-motion, see below),
but these are expensive and therefore only used on important highways. Physical counts are still the most
popular and, depending on the level of competence of the people employed, reliable visual observations
can simultaneously be obtained to confirm certain assumptions and ensure a more accurate traffic load-
ing prediction. The following details should be assessed from observations:

> freight on vehicles (empty, half loaded or fully loaded) and nature of loads;
> vehicle type and number of axles per heavy vehicle;
> land use and current development trends; and
> possible traffic attraction once the road is completed.

These observations assist in assigning ESALs per vehicle for the various vehicle types
(shown in Table A2.2) as well as improving the accuracy of traffic growth predictions.

Once the surveys are complete, the ADE is determined for each lane using the formula:

ADE = (nJ x (FE)J) [equation A2.2]

where:
nJ = number of vehicles for each vehicle type (J) in the traffic spectrum;
(FE)J = estimated average ESAL per vehicle for each vehicle type (J). (See Table A2.2)

The degree and extent of overloading is an important statistic since pavement design philosophy is based
on the standard axle load. Overloading causes severe pavement damage. Any information on overload-
ing should therefore be obtained, usually from the law enforcement authorities. In the absence of reliable
information, it is advisable to make an assessment by conducting a load survey. A representative sample
of the heavy vehicles should be physically weighed to determine the number (or percentage) of vehicles
that are overloaded, and the degree of overloading. The results can then be extrapolated to the whole
population of heavy vehicles.

312/313
A2.3.2 Static or dynamic weighing procedures

These are physical measurements taken in the field to determine the range of actual axle loads:

> Static weighing is the stationary weighing of vehicles and is thus limited to a sample of vehicles on a
specific road. Care has to be taken that the sample chosen represents the full traffic spectrum and not
only those that are laden. Weighing carried out for law enforcement purposes should therefore not be
used.
> Dynamic weighing is a continuous weighing-in-motion procedure typically over a seven-day period
at a particular site. This method is the most accurate and suitable for traffic estimation and gives the
number of axles in each of the predefined axle mass categories. However due to relative high costs,
this method is seldom justifiable for minor roads.

Once the surveys are complete, the ADE is determined for each lane using the formula:

ADE = (nM x DM) [equation A2.3]

where:
nM = number of axles for each predefined axle mass category (M);
DM = calculated average ESALs per axle mass category (M).

The average ESALs per axle mass category is calculated from:

M D = (PM/SAL)d [equation A2.4]

where:
PM = axle load in kN for each axle mass category
SAL = relevant Standard Axle Load in kN (e.g. 80 kN)
d = damage coefficient. Dependent on both pavement type and material in the various layers.
A value of n = 4 is generally used as an average. Shallow pavements (relatively thin but strong
upper layers) have n-values exceeding 4, whereas less sensitive (deep) pavements have n-values
of less than 4.
A2.4 Determination of design traffic (structural
capacity)

Once the current ADE per lane has been determined, the increase in ADE due to expected traffic growth
during the design period is calculated thus:

ESALstotal = ADE x fJ [equation A2.5]

where:
ESALstotal = structural capacity for design period

J

f = cumulative growth factor
[equation A2.6]

365 x (1 + 0.01i) x [(1 + 0.01i)y 1]

=
(0.01i)

where:
i = anticipated traffic growth rate in percent
y = number of years in the design life

314/315
The cumulative growth factor (fJ) may also be obtained from standard tables, such as the one shown
below in Table A2.6.

Table A2.6 Cumulative Growth Factor f J

Design life y fJ for traffic growth of i per annum
(years) i = 2% i = 4% i = 6% i = 8% i = 10%
5 1,937 2,056 2,181 2,313 2,451
8 3,195 3,498 3,829 4,193 4,592
10 4,077 4,558 5,100 5,711 6,399
12 4,993 5,704 6,527 7,481 8,586
15 6,438 7,601 9,005 10,703 12,757
20 9,046 11,304 14,232 18,039 22,996
25 11,925 15,809 21,227 28,818 39,486
30 15,103 21,290 30,588 44,656 66,044

Note:
The data shown in Table A2.6 does not include all values for the variables i and y. Care should be taken
when estimating in-between values since direct interpolation or extrapolation may produce an inaccurate
result. Equation A2.6 should rather be used to calculate fJ.

When estimating the anticipated traffic growth rate (i), growth influences other than normal economic
growth factors should be identified. Upgrading roads and pavement rehabilitation often attracts traffic that
would normally use alternative routes.
A2.5 Practical approach to estimating design traffic

The procedures described above should always be followed when determining the structural capacity
(design traffic) at project level. However, there is often a need for a rough estimate of the structural capac-
ity requirements for a specific road, to obtain a ball-park figure to guide planning at network level as well
as assessing the inadequacies of existing pavements. In addition, such estimates are usually made at the
start of a project to get a feel for the type of pavement required relative to the level of traffic that needs
to be accommodated.

Where the number of heavy vehicles is known, a very rough estimate of the structural capacity require-
ments can be made using Table A2.7. This table relates the Number of heavy vehicles per lane per day
to Design traffic (structural capacity) in terms of ESALs x 106, with three variables; compound traffic
growth, pavement design life (years) and load factor (average number of ESALs per heavy vehicle).

For example. Where the number of heavy vehicles travelling in one direction of a two-lane road is
100 per day, the average ESALs per heavy vehicle is 2 and the compound traffic growth is 4% per annum,
the structural capacity required for a 10 year design life indicated from Table A2.7 is 0.91 x 106 ESALs.
If the average number of ESALs per heavy vehicle increases to 3.5, the structural capacity requirement
increases to 1.6 x 106 ESALs.

Cautionary note:
Users of this table must recognise that any structural capacity estimate thus derived has severe limita-
tions and can only be used as an indicator. The major shortcoming is the lack of definition of a heavy ve-
hicle, all are assumed to have the same number of 80 kN ESALs. Pavement designs should therefore not
be based on such information; a proper traffic analysis and forecast following the procedures described in
this section should always be undertaken.

316/317
Table A2.7. Guide for estimating Design Traffic (Structural Capacity) in millions of equivalent 80 kN stand-
ard axle loads (ESALs x 106)

Pavement design life (years)

No. of Com-
heavy pound 5 10
vehicles traffic
Vehicle load factor (80 kN ESALs per heavy vehicle)
per day growth
0.6 2 3.5 4.4 0.6 2 3.5

2%
10 4%
6%
8%

2% 0.29
20 4% 0.32
6% 0.36
8% 0.40

2% 0.34 0.43 0.41 0.71

50 4% 0.36 0.45 0.46 0.80
6% 0.38 0.48 0.51 0.89
8% 0.40 0.51 0.57 1.00

2% 0.39 0.68 0.85 0.82 1.43

100 4% 0.41 0.72 0.90 0.27 0.91 1.60
6% 0.44 0.76 0.96 0.31 1.02 1.78
8% 0.46 0.81 1.02 0.34 1.14 2.00

2% 0.58 1.94 3.39 4.26 1.22 4.08 7.13

500 4% 0.62 2.06 3.60 4.52 1.37 4.56 7.98
6% 0.65 2.18 3.82 4.80 1.53 5.10 8.92
8% 0.69 2.31 4.05 5.09 1.71 5.71 9.99

2% 1.16 3.87 6.78 8.52 2.45 8.15 14.27

1,000 4% 1.23 4.11 7.20 9.05 2.73 9.12 15.95
6% 1.31 4.36 7.63 9.60 3.06 10.20 17.85
8% 1.39 4.63 8.09 10.18 3.43 11.42 19.99

2% 3.49 11.62 20.34 25.57 7.34 24.46 42.80

3,000 4% 3.70 12.34 21.59 27.14 8.20 27.35 47.85
6% 3.93 13.09 22.90 28.79 9.18 30.60 53.55
8% 4.16 13.88 24.28 30.53 10.28 34.26 59.96

2% 5.81 19.37 33.91 42.62 12.23 40.77 71.34

5,000 4% 6.17 20.56 35.98 45.23 13.67 45.58 79.76
6% 6.54 21.81 38.17 47.98 15.30 51.00 89.24
8% 6.94 23.13 40.47 50.88 17.13 57.11 99.94

Key: < 0.256 x 106 ESALs

Table A2.7. Guide for estimating Design Traffic (Structural Capacity) in millions of equivalent 80 kN standard axle
loads (ESALs x 106)
 15 20

4.4 0.6 2 3.5 4.4 0.6 2 3.5 4.4

0.28 0.32 0.40

0.27 0.33 0.28 0.40 0.50
0.32 0.40 0.36 0.50 0.63
0.37 0.47 0.36 0.63 0.79

0.36 0.26 0.45 0.57 0.45 0.63 0.80

0.40 0.30 0.53 0.67 0.57 0.79 0.99
0.45 0.36 0.63 0.79 0.72 1.00 1.25
0.50 0.43 0.75 0.94 0.90 1.26 1.59

0.90 0.64 1.13 1.42 0.27 1.13 1.58 1.99

1.00 0.76 1.33 1.67 0.34 1.42 1.98 2.49
1.12 0.27 0.90 1.58 1.98 0.43 1.80 2.49 3.13
1.26 0.32 1.07 1.87 2.35 0.54 1.81 3.16 3.97

1.79 0.39 1.29 2.25 2.83 0.54 2.26 3.17 3.98

2.01 0.46 1.52 2.66 3.34 0.68 2.85 3.96 4.97
2.24 0.54 1.80 3.15 3.96 0.85 3.61 4.98 6.26
2.51 0.64 2.14 3.75 4.71 1.08 9.05 6.31 7.94

8.97 1.93 6.44 11.27 14.16 2.71 11.30 15.83 19.90

10.03 2.28 7.60 13.30 16.72 3.39 14.23 19.78 24.87
11.22 2.70 9.01 15.76 19.81 4.27 18.04 24.91 31.31
12.56 3.21 10.70 18.73 23.55 5.41 18.09 31.57 39.69

17.94 3.86 12.88 22.53 28.33 5.43 22.61 31.66 39.80

20.05 4.56 15.20 26.60 33.44 6.78 28.46 39.56 49.74
22.44 5.40 18.01 31.52 39.6+2 8.54 36.08 49.81 62.62
25.13 6.42 21.41 37.46 47.09 10.82 54.28 63.14 79.37

53.81 11.59 38.63 67.60 84.99 16.28 67.82 94.98 119.41

60.16 13.68 45.61 79.81 100.33 20.35 85.39 118.69 149.21
67.32 16.21 54.03 94.56 118.87 25.62 108.24 149.44
75.38 19.27 64.22 112.39 141.28 32.47 90.46 189.41

89.68 19.32 64.38 112.67 141.64 27.14 113.04

100.27 22.80 76.01 133.02 33.91 142.32
112.19 27.02 90.05 42.70
125.63 32.11 107.03 54.12

> 250 x 106 ESALs

318 / 319
Appendix 3 G
 uidelines for compiling specifi-
cations for recycling projects

A3.1 Scope 323

A3.2 Materials 325
A3.2.1 In-situ pavement material 325
A3.2.2 Imported natural or processed material 325
A3.2.3 Stabilising agents 326
A3.2.4 Water for construction 327
A3.3 Plant and equipment 328
A3.3.1 Recyclers 329
A3.3.2 Equipment for compaction and finishing 331
A3.3.3 Tankers for the supply of bituminous stabilising agents 332
A3.4 Construction 333
A3.4.1 General limitations and requirements 333
A3.4.2 Requirements before recycling commences 334
A3.4.3 Addition of stabilising agents 336
A3.4.4 Recycling 339
A3.4.5 Subgrade instability 340
A3.4.6 Compaction and finishing 341
A3.5 Trial sections 342
A3.6 Protection and maintenance 343
A3.7 Construction tolerances 344
A3.7.1 Surface levels 344
A3.7.2 Layer thickness 344
A3.7.3 Widths 345
A3.7.4 Cross-section 345
A3.7.5 Surface regularity 345
A3.8 Routine inspection and tests 346
A3.8.1 Application rate of stabilising agents 347
A3.8.2 Strength of the stabilised material 348
A3.8.3 Density achieved 349
A3.9 Measurement and payment 350
A3.9.1 Measurement items 350
A3.9.2 Example of a typical schedule of quantities 353

320/321
Pavement rehabilitation by recycling is a relatively young technology. As a consequence, few standard
specifications have been compiled that adequately cover the requirements for construction. These guide-
lines have been included to assist in the compilation of relevant contract documentation. They follow the
normal format and headings of a standard specification, giving guidance and examples of important fea-
tures that need to be included to avoid the conflict and claim procedures that are normally encountered
when specifications are ambiguous.
A3.1 Scope

Specifications for the recycling portion of the works must cover all operations in connection with the
construction of a new pavement layer using predominantly the material recycled from upper layers of
an existing road. These operations include:

> breaking down and recovering material in the upper layers of existing road pavements;
> changing the nature of the recovered material by the addition of imported material;
> the provision and application of stabilising agents and water; and
> mixing, placing, compacting and shaping to achieve a new pavement layer.

Pavement composition is seldom uniform over long sections of road. In addition to variations in the thick-
ness of individual layers, the quality of the material quality used in the original construction often varies.
Maintenance, upgrading and rehabilitation measures that are applied during the service life of the road
also introduce further variability into the upper portion of the pavement. Where there is more than one
type of recycling operation to be carried out, each needs to be fully described. For example, if the length
of the project to be rehabilitated by recycling is 32.9 km and there are seven different pavement types
each with a different treatment, these need to be detailed as shown in the example table below.

Table showing the different recycling operations required (example)

Recycling
No. Start End Length (m) Treatment required for rehabilitation
depth (mm)
Pre-treat with 2% lime. Within 24 hours,
1 29+900 32+900 3,000 175
stabilise with 3.0% foamed bitumen
Recycle with 1% cement and
2 32+900 35+000 2,100 200
2.7% foamed bitumen
Add 75 mm layer of crusher dust before
3 35+000 38+600 3,600 150
recycling with 2.5% foamed bitumen
Recycle with 3% cement as a new subbase.
4 38+600 49+400 10,800 300 Import 125 mm screened RAP and recycle
150 mm with 2% foamed bitumen
Pre-treat with 2% lime. Within 24 hours,
5 49+400 54+800 5,400 225
stabilise with 3.0% foamed bitumen
Add 75 mm layer of crusher dust before
6 54+800 61+000 6,200 150
recycling with 2.5% foamed bitumen
Recycle with 3% cement as a new subbase.
7 61+000 62+800 1,800 300 Import 125 mm screened RAP and recycle
150 mm with 2% foamed bitumen

322/323
The general description of the pavement to be recycled under the contract, the depth and type of recy-
cling (e.g. 175 mm deep treated with 1% ordinary portland cement and 2.5% bitumen emulsion), as well
as the end-product required are must be clearly defined.

In addition, the design requirements for each type of recycled material treated with stabilising agents
needs to be specified. This is normally included as a table showing the minimum requirements in terms
of strength parameters and density of the compacted layer. The following table shows an example from a
typical recycling project that includes both cement and foamed bitumen stabilisation:

Application rate of
Minimum strength Minimum
stabilising agent
Material ori- requirements density
(% by mass)
Section gin/depth of
% of mod
recycling (mm) Foamed ITSDRY
Cement UCS (MPa) AASHTO
bitumen (kPa)
T-180
km1+200 to 2+800 In-situ/250 2.5 200 1.5 98
km1+200 to 2+800 Imported/125 1 2.0 225 n/a 102
km2+800 to 8+600 In-situ/200 1 2.5 175 n/a 100
km8+600 to 12+800 In-situ/300 3.0 200 1.5 98
km8+600 to 12+800 Imported/150 1 2.0 225 n/a 102

The results of mix designs coupled with assumptions made in determining the pavement design are used
as guidelines in compiling such a table. The relevant methods to be used for determining strength (test
methods) also need to be specified.

A word of caution to those compiling such specifications: The values specified must be attainable in the
field and it is therefore important to liaise with those responsible for the field investigations and the design
procedures. For example, specifying a density requirement of 104% of the modified AASHTO density for
a recycled layer overlying an unstabilised material is impractical. Likewise, specifying a strength require-
ment similar to the maximum achieved from laboratory mix designs is unrealistic.

In addition, the following clause is normally included to place responsibility where it belongs, namely with
the contractor who is in control of the works:

It shall be incumbent on the contractor to organise and execute his operations such that these require-
ments are met.
A3.2 Materials

A3.2.1 In-situ pavement material

Details of all pavement investigations that were undertaken by those responsible for designing and speci-
fying the rehabilitation requirements need to be shown, normally in an appendix. These investigations
include:

> detailed description of the existing pavement structures that are to be recycled;
> the results of tests undertaken, indicating the grading, plasticity and other relevant properties of the
material to be recycled from the upper pavement layers; and
> relevant in-situ moisture contents of the various materials in the existing pavement, measured at the
time the investigations were undertaken.

The following disclaimer is normally included:

This information is offered in good faith but, in the circumstances pertaining to sampling and testing
procedures and the type of information furnished, no guarantee can be given that all the information is
correct or representative of the in-situ conditions at the time of construction. Any reliance placed by the
contractor on this information shall therefore be at his own risk, and he shall undertake his own separate
testing programme to determine the conditions prevailing at the time of construction.

A3.2.2 Imported natural or processed material

Where the project requirements call for imported material to be blended with that recycled from the exist-
ing pavement, the reason for such import needs to be given together with clear specifications for the type
and quantity of material to be imported. Material is normally imported for one or more of the following
reasons:

> to change the grading of the recycled material;

> to effect mechanical modification;
> to supplement the recycled material for shape correction purposes; and/or
> to increase the total pavement thickness.

Contractors need to understand the reasoning behind such requirements in order to formulate their most
cost effective offer since this exercise embodies a host of alternatives (e.g. purchase material from a com-
mercial source or establish their own crushing facility).

In addition, any special conditions under which the material is to be imported must be given (e.g. import
before or after the existing pavement is pre-pulverised). The type of material (e.g. graded crushed stone
with CBR >100% within a specified grading envelope) and amount to be imported must be clear (e.g.
30% by volume of new layer. Alternatively, spread as a nominal 75 mm layer (after compaction) on the
existing road surface).

324/325
A3.2.3 Stabilising agents

The type and quality of all stabilising agents that are to be used on the project must be clearly speci-
fied together with any relevant standards governing their manufacture (e.g. Ordinary portland cement
conforming to the requirements of BS 12.) and use. In addition, any special requirements (e.g. handling
and storage) must be stated. For example, the following paragraph is normally included when cement is
specified as a stabilising agent:

From the time of purchase to the time of use, all cement shall be kept under cover and protected from
moisture, all in accordance with the manufacturers or suppliers recommendations. All consignments
of these materials shall be used in the same sequence as that of their delivery on site. Stocks stored in
excess of three months shall not be used in the works without authorisation.

Where foamed bitumen stabilisation is specified, the type of bitumen to be used must be specified (e.g.
80/100 Penetration Grade) and the following requirements are normally included under this section of the
specifications:

Bitumen for foamed bitumen stabilisation shall be heated, stored and applied strictly in accordance with
the requirements detailed below. (List the specific requirements, e.g. maximum temperature of 195C).
All bitumen for stabilisation shall be delivered to the site in bulk tankers. Each bulk tanker shall be issued
with a Certificate of Loading that contains the following information:

> haulage units identification details;

> product identification (e.g. 150/200 Pen grade bitumen);
> name of the bitumen supplier;
> the relevant batch number and date of manufacture;
> assized weighbridge certificate indicating the net mass of product;
> the temperature at which the product was loaded into the tanker;
> the date, time and place of loading;
> comments concerning any abnormalities of the state of the tanker at the time of loading (e.g. internal
cleanliness, details of the previous load carried and whether there was any residual product from the
previous load; and
> details of any chemical or other substance added to the product before, during or after the loading
procedure (e.g. anti-stripping agent).
A3.2.4 Water for construction

Any limitations on the quality of water to be used in the construction needs to be specified. The following
paragraph is normally included:

Water shall be clean and free from detrimental concentrations of acids, alkalis, salts, sugar and other
organic or chemical substances. If the water used is not obtained from a public drinking water main, tests
may be required to prove its suitability.

326/327
A3.3 Plant and Equipment

To protect against poor workmanship resulting from the use of inappropriate machinery, the following
paragraphs are typical of those normally included in recycling specifications:

All plant and equipment shall be supplied and operated in such a manner as to recycle in situ pavement
material to the specified depth and construct a new layer, all in accordance with the requirements of the
specifications. All plant and equipment deployed on the site shall be of adequate rated capacity and in
good working order. Obsolete, poorly maintained, or dilapidated plant will not be allowed on site.
The minimum compliance requirements for plant and equipment to be used for the recycling work are
given in the following sub-clauses. The contractor shall provide the Engineer with details and technical
specifications of all the plant and equipment to be used for the recycling work at least two weeks prior
to the first proposed usage.
A3.3.1 Recyclers

The following is normally included to ensure that the recycler used on site is adequate:

Recycling shall be effected by utilising a purpose-built recycler to recover the material in the upper layers
of the existing pavement and blend together with any imported material pre-spread as a uniform layer on
the existing road surface. The machine employed shall be capable of achieving the required grading and
consistency of mix in a single pass. As a minimum, the recycler shall have the following features:

> It shall be factory-built by a proprietary manufacturer having a demonstrable track record and manufac-
turing history in the particular type of equipment;
> If older than 10 years, the machine shall be certified by the manufacturer or manufacturers authorised
agent to confirm operational fitness-for-purpose dated not more than 3 months earlier than the date on
which it commences work on the project;
> The milling drum shall have a minimum cut width of 2 metres with the capability of changing the speed
of rotation. The machine shall be capable of recycling to a maximum depth (specified in these docu-
ments) in a single pass;
> A level-control system that maintains the depth of milling within a tolerance of 10 millimetres of the
required depth during continuous operation;
> The milling drum shall rotate within an enclosed chamber inside which water and stabilising agents are
added to the recovered material at the rate required to achieve compliance with the specified require-
ments during a continuous operation.
> All spray systems fitted to the recycler shall be controlled by micro-processor to regulate the flow rate
with the speed of advance of the machine. All spray systems will also have the ability to allow variable
widths of application; and
> The recycler shall have sufficient power to mix the recycled material together with all additives to pro-
duce a homogenously mixed material during continuous operation.

Additional specifications are included to cover the type of stabilising agent that is applied. The following
are examples for stabilising with cement, bitumen emulsion and foamed bitumen.

328/329
Additional requirements when stabilising with cement.
Where the cement stabilising agent is not applied directly on the surface of the road prior to recycling, the
recycler shall be fed with cement slurry that is produced in a separate mobile mixing unit pushed ahead of
the recycler. Such a mixing unit shall have the following minimum features:

> the capability of supplying the cement slurry at the required rate to comply with the specified cement
application rate during continuous operation;
> capable of regulating the application rate of cement slurry in accordance with the speed of advance of
the recycler and volume of material during continuous operation;
> provide uniform application of cement slurry to the recycled material to produce a homogenous mix-
ture; and
> a micro-processor controlled method for monitoring cement usage during operation that can be validat-
ed by simple physical measurement for control purposes.

Additional requirements when stabilising with bitumen emulsion.

In addition, the recycler shall have the following capabilities:

> To supply the bitumen emulsion at the specified application rate during continuous operation;
> To regulate the application rate of bitumen emulsion in accordance with speed of advance of the recy-
cler and volume of material being recycled;
> To provide uniform application of the bitumen emulsion to the recycled material to produce a homog-
enous mixture; and
> A method for monitoring bitumen emulsion application during operation that can be reconciled by sim-
ple physical measurement for control purposes.

Additional requirements when stabilising with foamed bitumen.

In addition, the recycler shall have the following features:

> a series of expansion chambers mounted equidistant on the spraybar (maximum spacing 200 mm) for
creating the foamed bitumen;
> the capability of providing a constant supply of foamed bitumen at the specified application rate during
continuous operation;
> capable of regulating the quality of foamed bitumen and regulating the application rate in accordance
with speed of advance of the recycler and volume of material being recycled;
> provide a uniform application of the foamed bitumen across the width of application to produce a
homogenous mixture;
> a method for monitoring bitumen application during operation that can be reconciled by simple physical
measurement for control purposes;
> functioning temperature and pressure gauges on the bitumen supply line for monitoring purposes;
> a means of demonstrating that all expansion chambers are producing foamed bitumen at any time dur-
ing the operation (no blockages); and
> a means of providing a representative sample of foamed bitumen at any stage during normal operations
(test nozzle).

The following clause is normally added at the end of this section:

The mixed material shall exit from the mixing chamber in a manner that prevents particle segregation and
be continuously placed back in the excavation created by the recycler as it advances. Spreading and
placing to form the new layer shall be carried out by a motor grader only after the primary compaction has
been achieved (unless placed by a screed mounted on the rear of the recycling machine).

A3.3.2 Equipment for compaction and finishing

To prevent the incorrect application of compaction equipment and guard against the bridging phenom-
enon, it is advisable to include the following:

Initial compaction of the recycled material shall be undertaken using a single-drum vibrating roller oper-
ated only in high-amplitude vibration mode. The static mass of the roller to be used shall be in determined
by the thickness of the recycled layer, in accordance with the following table:

Thickness of compacted layer Minimum static mass of roller (tons) Drum type

< 150 mm 12 smooth

150 mm to 200 mm 15 smooth or padfoot

200 mm to 250 mm 18 padfoot

> 250 mm 20 padfoot

The operating speed of the primary roller shall never exceed 3 km/hr and the number of passes applied
over the full width of each cut shall be sufficient to achieve at least the specified layer density in the lower
two-thirds of the layer. Where a compactometer is specified for controlling density, it shall be fitted to the
primary roller.

330/331
A3.3.3 Tankers for the supply of bitumen stabilising agents

The following clause should be included when stabilising with either bitumen emulsion or foamed
bitumen:

Only tankers with a capacity exceeding ten thousand (10,000) litres shall be employed to supply the recy-
cler with bitumen stabilising agents. Each tanker shall be fitted with two recessed pin-type tow hitches,
one in front and the at the rear, thereby allowing the tanker to be pushed from behind by the recycler, and
to push a water tanker in front. No leaking tanker will be permitted on the site. In addition, each tanker
shall be equipped with:

> A functioning thermometer to show the temperature of the contents in the bottom third of the tank; and
> A rear feed valve, with a minimum internal diameter of 75 mm when fully opened, that is capable of
draining the contents of the tank.

Where foamed bitumen is applied, the following additional bullets are included:

> all-round cladding to retain heat; and

> a heating system capable of raising the temperature of the contents of the tank by at least 20 C
per hour.
A3.4 Construction

Typical clauses normally included in this section are reproduced below.

A3.4.1 General limitations and requirements

> Weather Limitations

No work shall be undertaken during misty or wet conditions, nor shall any work commence if there is a
risk that it may not be complete before such conditions set in. Similarly, work shall not be undertaken if
the ambient air temperature is below 5 C. No further work, other than finishing and compaction, will be
permitted if the air temperature falls below 10 C during operations.

Spreading of powdered chemical stabilising agents (lime and cement) on the road ahead of the recycler
shall not be permitted when windy conditions adversely affect the operation.

> Accommodation of Traffic

The contractor shall be responsible for the comfortable passage of public traffic over sections of the
road on which he has occupation and shall at all times take the necessary care to protect the public
and to facilitate the flow of traffic.

> Time Limitations

The maximum time period between mixing the recycled material with a stabilising agent and compact-
ing the placed material shall be determined by the type of stabilising agent that is employed:

cement: Three (3) hours;

hydrated lime: Eight (8) hours if kept moist;
bitumen emulsion: 24 hours (or before the emulsion breaks);
foamed bitumen: 24 hours; and
proprietary products: As per manufacturers instructions.

332/333
A3.4.2 Requirements before recycling commences

> Production Plan

Prior to the start of work every day, the contractor shall prepare a production plan detailing his propos-
als for the forthcoming days work. As a minimum, this plan shall include:

a sketch plan showing the overall layout of the length and width of road intended to be recycling
during the day, broken into the number of parallel cuts required to achieve the specified width and the
overlap dimensions at each longitudinal joint between cuts;
the sequence and length of each cut to be recycled before starting on the adjacent or following cut;
an estimate of the time required for milling, compacting and finishing each cut. The time required to
recycle each cut should be indicated on the sketch plan; and
the location where quality assurance tests are to be taken.

Unless stated to the contrary, longitudinal joints shall be planned to coincide with each and every
change in cross-fall across the road width, regardless of the implications on overlap width.

> Referencing the horizontal alignment

Prior to commencing with the recycling work, the existing horizontal alignment shall be referenced us-
ing a series of pegs (or poles) placed on either side of the road. These pegs (or poles) shall be posi-
tioned outside the working area at a constant distance from, and at right-angles to the centre-line, and
shall be used to reinstate the centre-line after recycling operations are complete. The distance between
successive pegs (or poles) shall not exceed 20 m on curves, or 40 m on tangents (straights).

> Preparing the Surface

Before any recycling work commences, the surface of the existing road shall be prepared by:

cleaning all vegetation, garbage and other foreign matter from the full road width, including any adja-
cent lanes or shoulders that are not to be recycled;
removing any standing water;
premilling where high-spots are to be removed (if required); and
accurately pre-marking the proposed longitudinal cut lines on the existing road surface.

In addition, the contractor shall record the location of all road marking features (e.g. extent of barrier
lines) that will be obliterated by recycling.
> Surface shape and level requirements
Unless otherwise stated, design drawings will not be issued detailing the final level requirements for
the surface of the rehabilitated road. Where the grade line and cross-sectional shape of the existing
road are not excessively distorted, it shall be the contractors responsibility to conduct his operations in
such a manner as to ensure that the surface levels of the completed recycled layer are in sympathy with
those that existed prior to recycling. Where surface defects are to be corrected and/or modifications
made to the grade line, instructions will be issued detailing the new surface level requirements. These
may be achieved prior to recycling by either premilling to remove in-situ material or by importing mate-
rial and accurately spreading on the existing road surface.

> Addition of imported material

Where the design calls for material to be imported as make-up material for the purpose of shape
correction, the prescribed material shall be imported and spread on the existing road surface prior to
recycling. The method of placing and spreading the imported material shall be such as to achieve the
required surface levels and may therefore require the use of a paver, motor grader or other such plant.
Should the thickness of imported material exceed the intended recycling depth, then the requirements
for shape correction will have to be modified by regrading the road surface on either side of the low
point.

Where the design calls for material to be imported for the purpose of altering the grading of the
recycled material, or effecting mechanical modification, the prescribed material shall be imported and
spread on the surface of the existing road as a layer of uniform thickness prior to recycling.

> Pre-milling
Where required, pre-milling shall be undertaken using a milling machine (not a recycler) to:

Remove material from the road. Isolated high spots shall be removed and/or minor modifications
made to crest vertical curves by accurate milling. The material resulting from such milling operations
shall be loaded onto trucks and removed from site.
Break down (pulverise) thin layers of asphalt. Badly cracked asphalt layers (full-depth crocodile
cracks at intervals < 100 mm), and/or sections where thin asphalt overlays are delaminating, shall
be pre-milled immediately in advance of the recycling operation. To ensure that the milling operation
achieves the required degree of pulverisation, the depth of milling shall be constantly monitored and
adjusted so that the bottom of the milling drum remains within the lower half of the cracked / delami-
nating asphalt layer.

The pulverised asphalt material generated from such pre-milling shall remain on the road, behind the
milling machine, where it shall be spread across the width of recycling and rolled with a smooth-drum
roller.

334/335
 Where an acceptable degree of pulverisation cannot be achieved, the machine shall be operated in
reverse (i.e. down-cutting) with the same controls applied to the depth of milling. If such reverse mill-
ing fails to produce an acceptable degree of pulverisation, the offending asphalt layer shall be milled
off and removed.

> Pre-pulverising existing pavement material

Pre-pulverising shall only be undertaken for the purpose of:

- breaking down excessively hard material;

- loosening the material across the road width so that it can be cross-mixed by grader;
- exposing the loosened (fluffed-up) material to the atmosphere to promote drying; or
- loosening the material in the existing pavement so that it can be loaded and removed from site.

The depth of pre-pulverising shall be carefully controlled throughout the operation to ensure that the
cut horizon always remains at least 50 mm above the bottom of the subsequent recycling / stabilisation
horizon.

Unless the objective of pre-pulverising is to dry the material, a water tanker shall be coupled to the re-
cycler and sufficient water added to allow the material to be compacted to a minimum density of 95%
of the mod AASHTO density. Except where the material is to be cross-mixed, it shall be compacted im-
mediately behind the recycler before using a grader to pre-shape the material in accordance with final
level requirements. Where cross-mixing is ordered, the material shall be bladed by grader across the
specified width to achieve a uniform blend of material before being compacted and shaped.
A3.4.3 Addition of stabilising agents

The type of stabilising agent and the required application rate, expressed as a percentage of the mass
of the material to be stabilised, will normally be determined from mix design tests carried out prior to the
work commencing and issued to the Contractor as an instruction.

> Chemical stabilising agents (cement and lime)

The method of applying chemical stabilising agents shall be at the contractors discretion and may be
either:

spread as a uniform layer of dry stabilising agent on the prepared road surface prior to recycling; or
fluidised as a slurry by premixing with water and pumped to the recycler for injection through a spray-
bar into the mixing process; or
premixed in a batch plant and spread on the road surface together with any imported material.

Dry stabilising agents shall be spread uniformly over the full width of road to be recycled during each
pass of the recycler, either by means of a mechanical spreader at the prescribed rate of application as
a continuous process, or by hand. Where spreading is done by hand, pockets or bags of the stabilising
agent shall be spaced at equal intervals along each individual cut. The bags shall be emptied and the
contents spread evenly over the entire area of cut, excluding any overlap.

Mechanically operated mixers shall be used for the manufacture of slurry from dry powdered stabilising
agents and water. The mixer shall be equipped with a screen with openings not exceeding 5 mm and
shall be capable of producing a slurry of uniform consistency and constant water content at the rate
required for stabilisation.

> Bitumen stabilising agents

The bitumen stabilising agent shall be added to the recycling process by pumping from a mobile
bulk tanker that is pushed ahead of the recycler. Where foamed bitumen is applied, tankers shall be
equipped with a built-in thermometer and heating facilities to ensure that the bitumen is maintained
within 5 C of the specified application temperature. Any bitumen that has been heated above the
maximum specified temperature shall not be used and shall be removed from the site.

A one-litre sample of bitumen stabilising agent shall be taken from each tanker load and retained in a
sealed tin as a provision for later testing.

The following sentence should be included when working with foamed bitumen:

Not later than two (2) minutes after starting to recycle with each new tanker load, the foaming
characteristics of the bitumen shall be checked using the test nozzle on the recycler.

336/337
> Addition of fluid stabilising agents
The pumping system required to inject a fluid stabilising agent into the mixing process shall be con-
trolled by the same micro-processor system that monitors travel speed for the control of water addition.

> Controlling the moisture content of recycled material

Sufficient water shall be added during the recycling process to meet the moisture requirements speci-
fied below. Water shall be added only by means of the micro-processor control system on the recycler
and particular care shall be taken to prevent any portion of the work from excessive wetting. Any por-
tion of the work that becomes too wet will be rejected and the contractor shall be responsible for cor-
recting the moisture content by drying out and reprocessing the material, together with fresh stabilising
agent where cementitious stabilising agent is employed, all at his own expense.

At the time of compaction, the type of stabilising agent applied shall govern the moisture content of the
recycled material:

i) Cementitious stabilising agents

The moisture content during compaction shall never exceed 75% of the saturation moisture content of
the natural material (before stabilising), calculated at Maximum Dry Density. The moisture content at the
specified degree of saturation shall be determined using the following formula:

Wv = Sr x {(Xw/Xd) (1,000/Gs)}

where:
Wv = moisture content of the material at the specified degree of saturation (%)
Sr = specified degree of saturation (%)
Xw = density of water (kg/m3)
Xd = maximum dry density of the natural material (kg/m3)
Gs = apparent density of the material (kg/m3)
 ii) Non-cementitious stabilising agents and material recycled without stabilising agents
The moisture content during compaction shall not exceed the optimum moisture content, nor shall it be
less than 70% of the optimum moisture content.

iii) Bitumen emulsion stabilising agents

The total fluid content of the material during compaction shall not exceed the optimum fluid content.
The total fluid content shall be determined by summing the total amount of bitumen emulsion applied
(not only the water fraction) to the in-situ moisture content before mixing, plus any other water applied
independent of the water fraction of the emulsion.

A3.4.4 Recycling

The recycling machine shall be set up and operated to ensure the following key requirements are met:

> Grading of the recycled material

The forward speed of the recycling machine, rate of rotation of the milling drum and the positioning of
the gradation control beam shall be set so that the in-situ material is broken down to an acceptable
grading. The contractor shall take all necessary steps to ensure that the grading that results from the
recycling process conforms to those established during the Trial Section, as described in Clause A3.5
below.

> Addition of water and fluid stabilising agents

The micro-processor control system for the addition of water and fluid stabilising agents shall be set
and carefully monitored to ensure compliance with the requirements for compaction moisture and sta-
biliser content. Where practical, bulk bitumen tankers shall be dipped at the end of each cut to check
actual usage against the calculated theoretical demand.

> Control of cut depth

The actual depth of cut shall be physically measured by dipping from a stringline pulled between survey
control poles at least once every 100 m along the cut length using the same references on the survey
control poles that will be used to achieve final surface levels.

> Overlap on longitudinal joints

To ensure complete recycling across the full width of the road, longitudinal joints between successive
cuts shall overlap by a minimum of 150 mm. Cut lines pre-marked on the road surface shall be checked
to ensure that only the first cut is the same width as the milling drum. All successive cut widths shall be
narrower than the drum width by at least 150 mm. The recycling machine shall be steered so as to

338/339
 accurately follow the pre-marked cut lines. Any deviation in excess of 100 mm shall be rectified imme-
diately by reversing to where the deviation commenced and reprocessing along the correct line, without
the addition of any further water or stabilising agent.

The overlap width shall be confirmed before starting each new cut sequence and any adjustments
made to ensure that the amount of water and fluid stabilising agent to be added is reduced proportion-
ately by the width of the overlap.

> Continuity of stabilisation (lateral joints)

The contractor shall ensure that between successive cuts (along the same longitudinal cut line) no gaps
of unrecycled material remain, nor are any untreated wedges created where the milling drum first enters
the existing material. The exact location at which each cut terminates shall be carefully marked. This
mark shall coincide with the position of the centre of the mixing drum at the point at which the supply
of stabilising agent ceased. To ensure continuity of the stabilised layer, the next successive cut shall be
started at least 0.5 m (500 mm) behind this mark.

> Speed of advance

The speed of advance shall be checked and recorded at least once every 200 m of cut to ensure con-
formity with the planned production rate and ongoing compliance with the recycling process. Accept-
able tolerance limits shall be dependent on the type of recycler and material being recycled, but shall
not be less than 6 m/min nor greater than 12 m/min.

A3.4.5 Subgrade instability

Where subgrade instability is identified either by preliminary investigations, or during the recycling
process, it shall be treated by:

> recovering the material in the pavement layers overlying the unstable material by either milling or exca-
vating and loading into trucks for transport to temporary stockpile;
> excavating the unstable material to the prescribed depth, and removal to spoil;
> treating the exposed roadbed, as specified; and
> backfilling the excavation using both temporary stockpiled and imported material.

Backfilling shall be undertaken in layers not thicker than 200 mm after compaction, and shall continue in
successive layers utilising appropriate material for each layer. When backfilling is complete, recycling shall
continue.
A3.4.6 Compaction and finishing

> Initial compaction

The recycled material shall be rolled initially using a heavy vibrating roller to achieve the specified com-
paction. Rolling shall commence immediately behind the recycler and shall follow the predetermined
sequence as described in Clause A3.5 below.

Density control shall be applied to:

the average density achieved over the full layer thickness; and
the density of the lowest two-thirds of the layer, which shall not be less than the average density for
the full layer thickness, minus 2%.

Alternatively, where refusal density is specified, the following should be included:

The recycled material shall be rolled initially using a heavy vibrating roller fitted with an integrated com-
pactometer system. Compactive effort shall be applied to the recycled material immediately behind the
recycler in high amplitude vibrating mode only. Successive sections, each no longer than 100 m, shall
be compacted and rolling on each section shall continue until the integrated compactometer system
indicates that the maximum density achievable has been reached before the roller moves forward to the
next section.

> Level and shape control

Processed material shall be spread by the recycler to fill the cut void. Such spreading may be achieved
either by a screed attached to the rear of the recycler (track-mounted machines), or by applying suf-
ficient pressure to the rear door of the recycling chamber to ensure that the material spreads across
the full cut width (tyre-mounted machines). The spread material shall then be initially compacted (as
described above) before cutting final levels with a motor grader using references on the survey control
poles as a guide.

> Final compaction, watering, finishing and curing

After final shaping, a smooth-drum vibrating roller operating in low amplitude vibration mode shall
complete the compaction process. The road surface shall then be treated with a light application of
water, or diluted bitumen emulsion where specified, and rolled with a pneumatic-tyred roller to achieve
a close-knit texture. The surface of the completed recycled layer shall be kept continuously damp by
frequent light watering.

The final completed layer shall be free from:

surface laminations (or biscuits);
portions exhibiting segregation of fine and coarse aggregate; and
corrugations, or any other defects that may adversely affect the performance of the layer.

340/341
 A bitumen surfacing (asphalt or seal) shall not be applied to the surface until the moisture content of
the material in the upper 100 mm horizon of the recycled layer is below 50% of the optimum moisture
content of the material.

> Opening to traffic

Unless otherwise instructed, the full road width shall be opened to traffic outside normal daylight work-
ing hours. All temporary road signs, delineators and other traffic control facilities shall be in place before
the road is opened to traffic.

A3.5 Trial sections

Trial sections should always be specified since they force the contractor to rehearse before commencing
with the recycling work. The following clause is usually included:

At the start of the project the contractor shall assemble all items of plant and equipment that he proposes
to use for cold in-place recycling, and shall process the first section of road to be rehabilitated in order to:

> demonstrate that the equipment and processes that he proposes to employ are capable of constructing
the recycled layer in accordance with the specified requirements;
> determine the effect on the grading of the recycled material by varying the forward speed of the recy-
cling machine and the rate of rotation of the milling drum; and
> determine the sequence and manner of rolling necessary to obtain the minimum compaction require-
ments.

A Trial Section shall be at least 200 m in length of full lane-width or half-road width. Should the contractor
make any alterations in the methods, processes, equipment or materials used, or if he is unable to comply
consistently with the specifications due to changes in the in-situ material, or for any other reason, he may
be required to undertake further demonstrations before continuing with the permanent work.
A3.6 Protection and maintenance

The following standard clause is usually included:

The contractor shall protect and maintain the completed recycled layer until the next layer or surfacing is
applied. In addition to frequent light watering to prevent the surface from drying out, maintenance shall
include the immediate repair of any damage to or defects in the layer and shall be repeated as often as
it is necessary. Repairs shall be made so as to ensure that an even and uniform surface is restored after
completion of the repair work. The cost of such repairs shall be borne by the contractor except where the
damage is the result of fair wear and tear from early trafficking. Damage caused by prolonged trafficking
as a result of a delay in applying the next layer or surfacing shall not be deemed to be fair and tear if such
a delay was due to circumstances under the control of the contractor.

342/343
A3.7 Construction tolerances

Construction tolerances specified are usually similar to those for new construction, as shown in the fol-
lowing example:

Only where premilling to achieve the prescribed shape prior to recycling is specified will both surface
levels and layer thickness be subjected to statistical analyses. The completed recycled layer shall comply
with the construction tolerances given below:

A3.7.1 Surface levels

The following is a typical specification for surface levels:

A lot size shall be at least 50 levels taken in a random pattern. The lot will comply with the requirements
specified if it meets the following tolerances:

> H90 20 mm (ie. at least 90% of all surface levels measured are within 20 mm, plus or minus, of the
specified levels); and
> Hmax 25 mm (ie. individual spot levels shall not deviate by more than 25 mm from the specified levels).

A3.7.2 Layer thickness

Since the thickness of the stabilised layer is one of the most important determinants of ultimate perfor-
mance, relatively tight tolerances should be specified, as shown in the following example:

A lot size shall be at least 20 layer thickness measurements. The lot will comply with the requirements
specified if it meets the following tolerances:

> D90 10 mm (ie. at least 90 % of all thickness measurements are equal to or greater than the specified
thickness minus 10 mm);
> Dmean Dspec (Dspec/20) (ie. the mean layer thickness for the lot shall not be less than the specified layer
thickness, minus the specified layer thickness divided by twenty); and
> Dmax < 15 mm (ie. no individual layer thickness measurement shall be less than the specified thickness
minus 15 mm).
A3.7.3 Widths

Nowhere shall the width of the recycled layer be less than the width specified.

A3.7.4 Cross-section

When tested with a 3 m straightedge laid at right-angles to the road centre-line, the surface shall not devi-
ate from the bottom of the straightedge by more than 10 mm. At any cross section, the difference in level
between any two points shall not vary from their difference in level computed from the required cross-
section by more than 15 mm.

A3.7.5 Surface regularity

The following is a typical specification for surface regularity:

When testing the finished recycled layer with a standard rolling straightedge, the number of surface
irregularities shall not exceed:

> six (6) for the average number of irregularities per 100 m equal to or exceeding 6 mm when taken over
300 m to 600 m lengths; and
> eight (8) for the number of irregularities equal to or exceeding 6 mm when taken over a single 100 m
section.

Any individual irregularity measured with the rolling straightedge, or a 3 m straightedge laid parallel to the
road centre-line, shall not exceed 10 mm. However, where it can be shown that irregularities are caused
by factors beyond the control of the contractor (e.g. damage caused by early trafficking), this requirement
may be relaxed.

344/345
A3.8 Routine inspection and tests

The onus rests with the contractor to produce work that conforms in quality and accuracy of detail to all
the requirements of the specifications and drawings. The contractor is therefore responsible for instituting
a quality control system to ensure positive control of the works and to demonstrate that specified require-
ments have been met.

Excluding the geometric details (described above under Construction Tolerances), a further three key
parameters of the recycled work requires careful control, namely:

> the application rate of stabilising agents;

> strength achieved in the recycled/treated material; and
> density achieved in the new recycled layer.

These are dealt with below in separate sub-sections.

A3.8.1 Application rate of stabilising agents

Due to variability in existing distressed pavements, recycled material is seldom homogeneous. Normal
tests to confirm that the required amount rate of stabilising agent has been applied are therefore often
inappropriate as they produce misleading results. For example, calculating the bitumen content as a
percentage by mass of a material recycled with foamed bitumen cannot provide the answer to how much
foamed bitumen was actually added because:

> the amount of bitumen extracted from such a mix is likely to vary considerably due to the presence of
bitumen in the existing pavement (in existing asphalt, additional surfacing layers and in patches);
> expressing bitumen content as a percentage of the total mass of the mix is a concept borrowed from
asphalt technology as it assumes the aggregate conforms to a uniform grading. Since recycled mate-
rial has a variable grading, the inclusion or exclusion of one large piece of aggregate will significantly
change the percentage of bitumen in the sample, making the result meaningless.

Alternative control measures are therefore required and these are normally based on consumption checks,
or physical measurements of the actual quantity of stabilising agent applied compared with the specified
application rate.

> Cement stabilisation

Where cement is spread by hand on the existing road surface prior to recycling, checking the application
rate is relatively straightforward provided the surface is pre-marked for individual bags.

Where a spreader is used to apply the cement on the existing road surface prior to recycling, the standard
canvas-patch test, or similar, is normally used to check the application rate.

Where cement is applied by means of slurry injection, the actual consumption of cement (and water)
can be obtained from the computer that controls the slurry mixing unit. In addition, weighbridge certifi-
cates should be obtained from bulk cement deliveries and verified against daily usage indicated by
the computer.

> Bitumen stabilisation

The actual consumption of both bitumen emulsion and foamed bitumen is best controlled by ensur-
ing that all tankers delivering product to the recycler are provided with weighbridge certificates for each
load. The mass of material stabilised with each tanker (estimated from the length of cut multiplied by the
application width, depth of cut and assumed material density) cut then be reconciled with the amount
consumed, provided always that the tanker is drained.

346/347
A3.8.2 Strength of the stabilised material

A bulk sample (200 kg) is normally taken from behind the recycler, at least one per 2,500 m2 of pavement
recycled. This material is placed in a sealed container and taken immediately to the laboratory for testing.
Normal tests include:

> moisture content;

> moisture/density relationship to determine the maximum dry density (used also for determining the
percentage compaction achieved in the field;
> strength determination from manufactured briquette specimens. (Note that the moisture content of the
material is always adjusted to the optimum moisture content before manufacturing the briquettes with
standard compactive effort. In addition, the temperature of the material being compacted should be
similar to that in the field.)

Other tests that are sometimes required include sieve analyses and plasticity determination.

Where there is reason to suspect that the required strength has not been achieved, a 150 mm diameter
core can be extracted from the layer and tested. This is usually undertaken some 2 to 4 weeks after the
layer was constructed.

See Appendix 1 for detailed test procedures.

A3.8.3 Density achieved

Determining the field density of a layer constructed from recycled material is seldom a straightforward
exercise due to two features of recycled material:

> variability of the recycled material that affects the maximum dry density value against which field den-
sity is compared; and
> bitumen in that portion of the material recycled from existing asphalt and/or bitumen surfacing that
affects the moisture content reading of nuclear gauges.

Bulk samples therefore need to be taken at each and every test location and tested in the laboratory to
determine the maximum dry density of the material and the actual field moisture content. This invariably
increases the workload of the site laboratory and often causes a delay in obtaining results.

In addition, where the specified density is not achieved, there may be reason to suspect that poor un-
derlying support may be a contributing factor making it practically impossible to obtain a higher density
than that already achieved. This is a potential source of conflict that requires additional tests to resolve
(normally a limited DCP survey).

The integrated compactometer system fitted to the primary roller has recently been adopted on recycling
projects to indicate when refusal density has been achieved. This simple system offers a solution to the
problems described above and is therefore recommended for controlling field density. The following is an
example of a specification for such equipment:

The density required is the refusal density and shall be defined as the maximum density achievable
in the field, as indicated by an integrated compactometer system. Such a system shall be fitted to the
single-drum vibrating roller used for initial compaction behind the recycler. Rolling shall continue until the
unit indicates that no further densification is being achieved under additional passes of the roller.
This information shall be stored in the systems computer, downloaded on a daily basis and used to
generate a comprehensive compaction record indicating the level of compaction that was actually
achieved every 2 m of cut, together with proof that maximum densification had been achieved.

Check tests using normal test methods are still required on the completed layer, but at a much-reduced
intensity, normally 6 tests per days work.

348/349
A3.9 Measurement and payment

To avoid any arguments, the following should always be included:

The description of certain pay items states that quantities will be determined from authorised dimen-
sions. This shall be taken to mean the dimensions as specified or shown on any drawing or written
instruction given to the contractor, without any allowance for tolerances. If the work is constructed in
compliance with the authorised dimensions, plus or minus any tolerances allowed, quantities will be cal-
culated from the authorised dimensions regardless of the actual

A3.9.1 Measurement items

Each item to be used for the measurement (and payment) of the recycling work requires a full definition.
The following are examples of typical items:

Item Unit
A3.01 Prepare the existing road surface prior to recycling square metre (m2)

The unit of measurement shall be the square metre of existing road surface that is to be rehabilitated by
recycling, calculated from the authorised width dimension multiplied by the actual length as measured
along the centre-line of the road.

The rate tendered shall include full compensation for all work necessary to clean the road of all water,
vegetation, garbage, and other foreign matter and for the removal, transporting and disposal of all result-
ing debris, as specified.

Item Unit
A3.02 R
 ecycling in all in-situ pavement materials
for the construction of new pavement layers:
a) ........mm (specify) thick completed layer:
i) Road width of 5.0 m or less cubic metre (m3)
ii) Road width greater than 5.0 m, but less than 6.0 m. cubic metre (m3)
iii) etc. for road width increments of 1 m.
b) etc. for each specified layer thickness.

The unit of measurement shall be the cubic metre of completed pavement layer constructed by recycling
the in-situ pavement material, regardless of the hardness or type of such material, and with or without the
inclusion of imported material. The quantity shall be calculated from the authorised dimensions for width
and thickness of the completed layer, multiplied by the actual length as measured along the centre-line of
the road. The authorised width shall not be increased to include any allowance for the specified minimum
overlap between adjacent cuts, nor for the number of cuts required to cover the full road width.

The rates tendered shall include full compensation for setting out the works, for recycling all types of
material in the existing pavement structure to the specified depth together with any stabilising agents
and/or imported material that may have been incorporated, for the supply and addition of water, for mix-
ing, placing and compacting the material, for the reworking of all material in overlapping adjacent cuts
regardless of the number of cuts or width of overlap necessary to cover the full road width, for all curing,
protection and maintenance of the layer, and for conducting all process and acceptance control inspec-
tions, measurements and tests.

Item Unit
A3.03 E
 xtra over Item A3.02
for layers of asphalt material within the recycled portion
of the existing pavement, where the average thickness of asphalt is:
a) More than 50 mm, but less than or equal to 75 mm cubic metre (m3)
b) More than 75 mm, but less than or equal to 100 mm cubic metre (m3)
c) Etc. for increments of 25 mm.

The unit of measurement shall be the same as for item A3.02, the extra over measurement applying to the
full layer thickness regardless of the relative proportions of asphalt and other material that constitutes the
material in the total thickness of the milled layer. No additional payment shall be made where the asphalt
thickness is less than or equal to 50 mm.

The rates tendered shall include full compensation for all additional direct and indirect costs incurred as a
result of recycling material that includes layers of asphalt thicker than 50 mm. These additional costs shall
include, but shall not be limited to, the extra wear and tear on plant and equipment, additional ground-
engaging tools, additional costs incurred where the rate of advance dictates that the pavement is first
premilled before stabilising, and all allowances for delays caused by the resulting slow rate of production.

Item Unit
A3.04 Imported material for addition to the cold in-place recycling process:
a) C
 rushed-stone products from commercial sources:
i) Size and description of crushed product ton (t)
ii) Etc. for each product size and type.
b) Natural gravels and sands from commercial sources:
i) Size and description of natural product cubic metre (m3)
ii) etc. for each type of natural material.
c) etc. for each different imported material.

350/351
The unit of measurement for crushed-stone products purchased from commercial sources shall be the
ton of material brought onto the site and incorporated into the recycled material. Measurement shall be
based on weighbridge tickets. The unit of measurement for natural material, purchased from commercial
sources or obtained from borrow pits, shall be the cubic metre, measured as 70 % of the struck volume of
the haul vehicle.

The tendered rate shall include full compensation for procuring, furnishing and spreading the imported
material on the existing road as a level-correcting layer, or as a layer of uniform thickness, for hauling
the material from the point of supply to its final position on the road, for watering and light rolling where
required, and for any wastage.

Item Unit
A3.05 A
 ctive filler:
a) Ordinary Portland cement ton (t)
b) Etc. for each type of chemical stabilising agent specified.

The unit of measurement shall be the ton of active filler actually consumed in the recycling process.
Measurement shall be based on weighbridge tickets where the supply is bulk, or on agreed counts where
the supply is in pockets or bags.

The tendered rate shall include full compensation for procuring and providing the active filler, for its
addition to the recycling process, including all transport, handling, storage under cover where required,
rehandling and spreading, or fluidising into a slurry and pumping into the process, for all wastage and
safety measures necessary during handling, and for the disposal of any packaging.

Item Unit
A3.06 B
 itumen stabilising agents:
a) Bitumen emulsion:
i) 60% residual bitumen, slow-set cationic ton (t)
ii) Etc. for each different type of bitumen emulsion.
b) Foamed bitumen, produced from:
i) 80/100 penetration-grade bitumen ton (t)
ii) etc. for each different type of bitumen.

The unit of measurement shall be the ton of bitumen stabilising agent actually consumed in the recycling
process. Measurement shall be based on physical dips of bulk tankers, undertaken before and after the
bituminous stabilising agent is applied, supported by weighbridge tickets issued for each tanker-load at
the point of supply.
The tendered rate shall include full compensation for procuring and providing the bitumen stabilising
agent, for its addition to the recycling process, for any chemicals or other additives introduced, for the
water added to achieve foaming, where required, for all transport, heating, handling, storage, and ap-
plying by pumping into the process, for all wastage and for taking all safety measures necessary during
handling.

A3.9.2 Example of a typical schedule of quantities

The following schedule is an example of a schedule of quantities for a recycling project that includes the
items listed above. The quantities relate to a 30 km long road, 7.3 m wide that is recycled to a depth of
175 mm (existing pavement includes an asphalt layer 80 mm thick) stabilised with 1% cement and 4%
bitumen emulsion.

Item Description Unit Quantity Rate Amount

Prepare the existing road surface prior

A3.01 m2 220,000
to recycling

Recycling in all in-situ pavement

materials for the construction of new
A3.02 pavement layers: m3 40,000
a) 175 mm thick layer:
i) Road width >7.0 m but less than 7.5 m

Extra over Item A5.02 for layers of asphalt

material within the recycled portion of the
existing pavement, where the average thick-
A3.03 m3 40,000
ness of asphalt is:
a) More than 75 mm, but less than or equal
to 100 mm

Chemical stabilising agents:

A3.04 ton 1,000
a) Ordinary Portland cement

Bituminous stabilising agents:

A3.05 a) Bitumen emulsion ton 4,000
i) 60% residual bitumen, cationic slow set

352/353
Appendix 4 T
 he principles of economic
analysis

A4.1 Introduction 357

A4.2 Comparability of cost on a time basis  358
A4.3 Economic evaluation techniques 360
A4.3.1 Present worth of cost (PWoC) technique  361
A4.3.2 Benefit/Cost (B/C) ratio technique  362
A4.3.3 Internal rate of return (IRR) technique  363
A4.3.4 Net present value (NPV) technique 364
A4.4 Period of analysis and the terminal and residual value of
transport facilities  365

354/355
The following is an extract from Guidelines for Conducting the Economic Evaluation of Urban
Transportation Projects: WJ Pienaar, University of Stellenbosch.
A4.1 Introduction

Economic evaluation is the conceptual framework for the assessment of all gains (benefits) and losses
(costs) of investment projects regardless of to whom they accrue within a country. A benefit is regarded
as any gain in utility emanating from the operation and use of a facility and a cost is any loss of utility as-
sociated with the implementation of a project, where utility is measured in terms of opportunity costs. (The
term economic evaluation does not include financial and social evaluation).

The primary purpose of the economic evaluation of urban transport projects based on economic ef-
ficiency and the implementation of subsequent recommendations is to minimise total transport cost,
provided transport needs are effectively met. The total transport cost of a project comprises recurrent and
non-recurrent costs. Non-recurrent costs comprise the initial cost (planning cost plus the opportunity cost
of establishing a facility). Recurrent costs are incurred continuously throughout the service life of a facility
and consist of the facility user costs and facility maintenance costs. An increase in non-recurrent costs
generally gives rise to a decrease in recurrent costs, and vice versa. Minimisation of transport cost can
therefore be achieved by determining the optimum trade-off between the cost categories.

There are three evaluation criteria on which the viability of a project may be based:

> Absolute advantage which may be determined by the net present value technique;
> Relative advantage which is usually determined through either the benefit/cost ratio technique or the
internal rate of return technique; and
> Minimum total cost which may be determined by the present worth of costs technique.

356/357
A4.2 Comparability of cost on a time basis

An economic evaluation of alternative transport projects requires individual cost evaluations on a common
time basis, since money has a continuous time value. Thus, an amount X would be more valuable now
than, say, in a years time. This greater relevance of present power of disposal over funds, compared with
eventual power of disposal over the same amount, is referred to as time preference propensity. Although
the rate of inflation has an influence on the intensity of time preference propensity, the prevalence of infla-
tion itself is not the cause that money has a time value. Even in non-inflationary periods there is still a time
preference attached to money that is related to a communitys average income from savings and invest-
ment. Thus, an amount kept in the purse where it cannot yield a return, is being denied the opportunity to
grow in an alternative way. The average time preference of an amount may therefore be equated with its
opportunity or alternative cost, as reflected in the average return on capital over a period.

Because transport facilities, such as roads, have a service life of several years, even decades, the evalu-
ation period generally extends over twenty or more years. Facility maintenance cost and facility user
cost incurred over a period therefore manifestly become less relevant as increasingly distant values are
incorporated in the evaluation process. The method of determining the present worth of future funds is
known as discounting. The rate used to compute the present worth of a cost in year n is referred to as the
discount rate, and represents the continuous time value of money. An economic assessment is possible
only after all future values have been expressed in equivalent terms, in other words after they have been
reduced to their worth at a common point in time by means of a representative discount rate.

The basic formula for discounting a single future amount to its present worth becomes more comprehen-
sible if discounting is regarded as the reverse of compound interest computation or, in other words, the
conversion of present worth into future worth by making use of a specific rate of interest.

If, say, the time value of an amount = i percent per annum, its present worth (PW) in a years time will
increase to PW(1 + i/1 00).
After two years its worth would equal: PW(1 +i/100)(1 +i/100) = PW(1 +i/100)2.
After three years the PW would have increased to an amount equalling PW(1 + i/100}3, and so on, until
after n years it would equal PW(1 + i/100)n.
Since discounting is the reciprocal of interest computation, the calculation applied to PW(1 + I/100)n in or-
der to discount it for a number of years at i percent would have to be inverted. In other words, one would
have to multiply it by 1/(1 + i/100)n to obtain its present worth.
The term (1 +i/100)n is known as the interest function, while its opposite, namely
1/(1 + i/100)n, is known as the discount function or the present worth factor.

The following formula is used to compute the PW of a. future amount (FA) at the end of year n at a
discount rate of i percent per annum:

PW = FA/(1 + i)n (equation A4.1)

where:
PW = present worth (value in year nil)
FA = future amount at the end of year n
i = annual discount rate as a fraction of 100
n = discount period in years

358/359
A4.3 Economic evaluation techniques

Proposed projects to be evaluated can be divided into two groups:

> Mutually exclusive proposals. Exclusive proposals are alternative methods of fulfilling the same func-
tion. The choice of anyone of the proposals will therefore exclude all the others. The cost-benefit analy-
sis of mutually exclusive proposals involves the selection of the most efficient, i.e. most cost-effective
alternative.

> Independent projects. These fulfil different functions and are therefore not alternatives for one another.
Examples of independent projects are: a proposed modal transfer facility in suburb X, a proposed
street-widening in suburb Y, and a proposed traffic interchange in suburb Z. More than one independ-
ent project can be selected for implementation. In fact, it is possible that all independent projects may
be selected if they are all economically justified and sufficient funds are available. The economic evalu-
ation of independent projects involves the ranking of the economically justified projects in terms of their
economic merit.

Various techniques, all based on the principle of discounted cash flows, can be used for cost-benefit
analysis. Four of the more commonly used techniques are explained below, namely:

> Present worth of cost (PWoC) technique;

> Benefit/cost ratio (B/C) technique;
> Internal rate of return (IRR) technique; and
> Net present value (NPV) technique.

In terms of their underlying philosophy, these techniques can be classified into two groups. For the first
group, only the cost of each alternative is calculated, the argument being that the alternative with the least
cost would be superior. The PWoC technique falls in this group.

In the second group of techniques, both benefits and costs of alternatives are calculated. Benefits are
defined as savings in recurring costs relative to the null alternative (i.e. the existing situation or present fa-
cility whose improvement or replacement is being investigated). The underlying philosophy of techniques
falling into this group, is that an alternative will be economically variable if benefits exceed costs. The cost
of a project can be defined as the opportunity cost of economic resources sacrificed in implementing
(providing) the project. The method of identifying the best alternative depends on the specific technique.
Three techniques fall into this group: the NPV, the B/C ratio, and the IRR technique.
A4.3.1 Present worth of cost (PWoC) technique

This technique selects the lowest cost alternative among mutually exclusive projects. All economic costs
(i.e. the opportunity costs) associated with the provision, maintenance and use of each possible alterna-
tive are discounted to their present worth. Given the objective of economic efficiency, the alternative that
yields the lowest PWoC is regarded as the most cost-effective (beneficial) proposal. This method can be
expressed as follows:

PWoC = Ca + PW(M + U) (equation A4.2)

where:
PWoC = present worth of cost
Ca = all initial costs incurred in constructing a facility, minus the discounted residual
value at the end of the analysis period
PW(M + U) = present worth of all maintenance and road user costs during the period under review.

Note that in the case of the null alternative (i.e. the existing facility whose possible replacement or upgrad-
ing is being investigated, and against which the other mutually exclusive alternatives are measured),
PWOC = PW(M + U).

360/361
A4.3.2 Benefit/Cost (B/C) ratio technique

This technique selects the most advantageous alternative by determining the ratio between project ben-
efits (i.e. annual savings relative to the null alternative) and discounted initial project costs. The anticipated
benefit during the analysis period is determined by subtracting the present worth of an alternatives
forecasted road user cost, plus road maintenance costs, from the present worth of the existing facilitys
forecasted user cost, plus maintenance cost.

The ratio between the sum of the discounted benefits and the sum of the initial project costs is obtained
by dividing the former by the latter. All proposals with a ratio value greater than one are viable, while
the one with the highest ratio value is economically the most advantageous. However, when mutually
exclusive alternatives are compared, incremental analysis must be used to identify the most economical
alternative. The method can be expressed as follows:

B/C = [PW (MO + UO) (PW (Ma + Ua)]/Ca (equation A4.3)

where:
B/C = benefit/cost ratio
o = existing facility
a = alternative under consideration
A4.3.3 Internal rate of return (IRR) technique

This technique calculates the anticipated internal rate of return of each alternative relative to the null alter-
native. The distinctive feature of this technique is that its application does not entail a singular discounting
procedure with one prescribed rate only.

Annual savings (returns) for the period under review are discounted to the beginning of the period. The
sum of these discounted amounts are compared with the discounted initial cost. Different rates of return
are selected iteratively and applied until at a certain rate the sum of the annual discounted returns equals
initial costs. This is then referred to as the (anticipated) internal rate of return. The alternative with the larg-
est internal rate of return can be regarded as the most advantageous, although the actual criterion is to
compare the rate thus obtained with the prevailing real discount rate. If it exceeds the prevailing discount
rate, the alternative is economically viable. However, when mutually exclusive alternatives are compared,
incremental analysis must be used to identify the most economical alternative.

The method can be expressed as follows:

IRR = r when {PW(Mo + Uo) PW(Ma + Ua)} = Ca

(equation A4.4)

where:
r = rate at which the left-hand and right-hand sides of the equation are equal.

362/363
A4.3.4 Net present value (NPV) technique

This technique selects an alternative among the mutually exclusive projects that has the greatest net
present value. The discounted initial cost of an alternative is subtracted from the sum of the discounted
annual savings that the alternative will achieve in comparison with the existing facility. All alternatives with
benefits reflecting a positive net present value are viable, while the alternative with the highest present
value is the most advantageous.

The technique can be expressed thus:

NPV = PW(Mo + Uo) PW (Ma + Ua) Ca

(equation A4.5)
A4.4 Period of analysis and the terminal and
residual value of transport facilities

The opportunity cost principle excludes the possibility of a transport facility possessing an additional
value over and above the opportunity cost of its land reserve during its service life, the reason being that
investment in transport facilities is regarded as sunken (i.e. it is taken that the development itself has
no possible alternative application). For this reason it is desirable that the analysis (evaluation) period
should stretch over the entire design life or planned lifespan of a facility. However, there are sound practi-
cal reasons why the analysis period should sometimes be shorter than the planned service life. Due to
risk/future uncertainties, all projections and forecasts for periods exceeding twenty years are extremely
speculative owing to, for example, the difficulty of predicting future traffic volumes, modal split, changes
in technology, land use, demographic features, etc.

The periods for which forecasts and projections are made can be extended beyond 20 years on condition
that they remain reliable. However, it is recommended that analysis periods should not exceed 30 years
(usually a 20 year analysis period is used), even though the logical solutions to certain transport problems
may be durable projects with service lives substantially longer than 30 years. To be fair and just (realistic)
with regards to investment in durable projects, a departure is made from the pure but limiting opportunity
cost rule by introducing a second best notional valuation rule in the form of a residual value. This value
purports to represent an artificial opportunity cost of a project during its service life and is widely, if not
universally, accepted as a legitimate convention and justified departure from the opportunity cost rule.
Had this not been the case, very few durable transport projects would ever have been implemented on
account of the outcome of an economic evaluation.

Should it be found that a project has an expected terminal value (i.e. the r salvage value of the facility at
the end of its design life) or a residual value (i.e. the salvage value of the facility during its design life), such
values should be discounted and deducted from discounted initial cost. The reason why this is done is
that an IRR and a B/C are returns or yields relative to the investment cost. For example, if a project has
a design or service life of 30 years and, after termination of this period, the land is again utilised for the
same economic activity as had been the case before the 30 year period, the cost of the initial investment
is equal to the opportunity cost of facility development (i.e. direct planning, design and construction) plus
the benefit which is being forfeited by not using its land reserve in its alternative application during the
service life. In other words, the investment cost in terms of which return or yield is calculated is the op-
portunity cost attached to developing the facility as well as the opportunity cost of engaging the land for
transportation purposes up to that point when it is again released for alternative use.

364/365
366/367
 Wirtgen GmbH
Reinhard-Wirtgen-Str. 2 53578 Windhagen Germany
Phone: +49 (0) 26 45 / 131-0 Fax: +49 (0) 26 45 / 131-392
Internet: www.wirtgen.com E-Mail: [email protected]

Illustrations and texts are without obligation. Technical details are subject to change without notice.
Performance data depend on operating conditions.
No. 2467915 EN-01/16 by Wirtgen GmbH 2016 Printed in Germany