4.

Mix deslgn
This chapter deals with all aspects of asphalt mix 4.6 Volumetric Characteristics 49
design including mix properties, selection of materials'
4.6.1, General 49
volumetric design and testing of asphalt mixtures.
4.6.2 Bulk Density 49
The design of specialised mixes such as open graded
4.6.3 Maximum Density 50
asphalt, stone mastic asphalt, fine gap graded asphalt,
4.6.4 Voids in Mineral Aggregate (VMA) 50
recycled mixes and other special applications are also
4.6.5 Air Voids 50
discussed.
4.6.6 Binder Content 51
4.6.7 Voids Filled with Binder (VFB) 51
4.1 lntroduction 41 4.6.8 Binder Film Thickness 51

4.1,.1, General 41
4.1..2 Design Process 41 4.7 Mix Stiffness 51
4.1.3 Australian Provisional Guide to
4.7.1 General 51
Asphalt Mix Design 42
4.7.2 Resilient Modulus Test 52

4.2 Asphalt Mix Behaviour 43

4.8 DeformationResistance 52
4.3 Selection of Mix Type and 4.8.1 General 52
Component Materials 44 4.8.2 Dynamic Creep Test 53
4.8.3 'sfheel Tracking Test 53
4.3.1, General 44
4.3.2 Aggregates 44
4.3.3 Mineral Filler 44 4.9 Fatigue 53
4.3.4 Binder 44
4.9.1 General 53
4.9.2 Fatigue Test 54
4.4 Combining and ProPortioning
of Aggregates 45
4.1O Moisture SensitivitY 55
4.4.1, General 45
4.10.L General 55
4.4.2 Maximising Packing Properties of
4.L0.Z Testing for Moisture Sensitivity 55
Dense Graded Mixes 45
4.4.3 Grading Envelopes 45
4.4.4 Combining Aggregates to a Target 46 4.11 Marshall Stability and Flow 56
Grading

4.5 Mixing, Conditioning and 4.12 Modified Hubbard-Field

Compaction of SamPles 46 Stability 57

4.5.1 General 46
4.5.2 Marshall Compaction 47 4.13 Open Graded Asphalt 58
4.5.3 Hubbard-Field Compaction 47
4.5.4 Gyratory Compaction 47
4.5.5 Preparation of Slabs and Large Samples 47 4.14 Stone Mastic AsPhalt 58
4.5.6 Compaction to Refusal Density 48
4.5.7 Workability 48
4,15 Fine Gap Graded AsPhalt 58

39
4.16 Recycled Asphalt 58

4.17 Cold Mixes 59

4.1a Mix Designs for Specialised

Applications 59
4.18.1 Polymer Modified Binders 59
4.1.8.2 Crumb Rubber Asphalt Mixes 59
4.18.3 Ultra Thin Asphalt 60

References 60

Appendices 62
A4.1 Combined Aggregate Particle Size
Distribution (Rothfuchs Method) 62
A4.2 Adjustment of Mix Designs 63

40
 4.1 lntroduction
4.1.1 General
The process of asphalt mix design involves the choice
of aggregate fype, aggregate grading, binder type, and
determination of a binder content that will optimise
the engineering properties in relation to the desired
behaviour in service.
Asphalt mix performance is also influenced by manu-
facturing and construction standards. The manner in
which the basic building blocks of materials, design,
manufacture and construction impact on the ultimate
properties and performance of an asphalt mix are
illustrated in Figure 4.1.
4.1.2 Design Process
lsphalt mix design involves the following basic steps
that are similar in concept, regardless of the actual
tests and procedures used:
- 1. selection of mix type;
2. selection of component materials;
3. combination of aggregates to meet target grading;
4. selection of target binder content or range;
5. mixing and compaction of asphalt mix to a density
that is representative of in-service conditions;
6. measurement of volumetric properties of compacted
mix;
7. mechanical testing of compacted samples;
8. verification of design properties on samples of manu-
factured asphalt, if required;
9. selection of job mix.
MATERIALS
DESIGN
FIELD
PLACING
ENVIRONMENT
Figure 4.1 Factors affecting asphalt mix properties
4',1
_Procedures for
designing asphalt mixes have been
generally developed around testing of dense graded
mixes and determination of optimum binder content,
although most tests can be used for other mix types
with suitable interpretation of results)
Selection of mix type and component materials
Selection of mix type for particular performance con-
ditions will set the parameters for aggtegate grading,
binder type, volumetric properties, and the general
performance characteristics of the mix.
Selection of materials must be appropriate to the
mix type and performance environment.
Recipe criteria
Recipe criteria may be used to define some or all of the
components and characteristics of an asphalt mix
where characteristics of the mix cannot be readily
defined or optimised by conventional tests. Such cri-
teria can also be used for simple mixes for use in
minor applications. Examples include:
r setting grading parameters for open graded mixes,
gap graded mixes, stone mastic asphalt, ultra thin
mixes and other special mix rypes
. setting binder content ranges for the preceding mix
rypes
. setting the binder rype or additives in any mix rype
where particular attributes are desired
. using mixes of similar composition to those that
have been determined to perform satisfactorily in
practice
. cold mixes.
 \ In many cases, volumetric properties and mechani-
cal performance tests may still be determined to ensure
compliance with voids targets or other criten)
Preparation and compaction of trial mixes
A variety of methods of sample preparation and com-
paction for laboratory testing of asphalt mixes are
used as described in Section 4.5.
t In most cases, trial mixes are prepared that combine
aggregates with different proportions of binder to
allow selection of a binder content that optimises the
desired volumetric properties. If these are not met at a
suitable binder content, the source or proportions of
aggregates may be altered and the process of sample
preparation and compaction is repeated.
Volumetric properties
Volumetric proportions and characteristics are the
basis of asphalt mix design and largely determine the
in-service performance of the mix.
For many applications, the asphalt mix design pro-
point.
cess ends at this
Mechanical testing
Ideally the mechanical properties of asphalt in-situ are
required for pavement design purposes. In-situ con-
ditions will vary with a range of factors including
temperature, loading time, stress conditions and
degree of compaction. Reproduction of in-situ srress
conditions in the laboratory is difficult, hence simpli-
fied tests have been introduced which may indicate
certain aspects of the in-situ behaviour.
Procedures for the mechanical testing of asphalt
mixtures may be considered in three groups:
. fundamental tests for stiffness and deformation
resistance, and fatigue:
- repeated load indirect tensile test
- repeated load dynamic creep test
- repeated flexural bending
simulative tests for deformation resistance and
other properties:
- wheel tracking test
- moisture sensitivity
empirical tests for design of asphalt mixes:
- Marshall test
- Modified Hubbard-Field test
A major objective of asphalt R&D programs in
Australia and overseas in recent times has been the
development and implementation of simple fundamen-
tal and simulative tests for characterisation of asphalt
mixes to supplement or replace empirical tests that did
not directly relate to road performance. In Australia,
this has resulted in the implementation of new design
procedures as described in Section 4.1.3.
In the USA, research programs have resulted in
the implementation of "Superpave". The volumetric
42
design concepts used in Australia are similar to
"Superpave". The main difference in the Australian
procedure is the adoption of locally developed, more
affordable, equipment for compaction and mechanical
testing of asphalt.
An outcome of the implementation of new volumet-
ric design procedures in Australia is that the Modified
Hubbard-Field Method has been totally discontinued
from use, although its description is included here to
enable interpretation of references contained in older
documents.
The Marshall Method of compaction and mechani-
cal testing is still used in many locations due to its sim-
plicity, although gradually being replaced by gyratory
compaction and other mechanical tests.
In some cases, design properties of asphalt mixes
may be verified by mechanical tests on samples of
plant manufactured or field materials before adoption
of a job mix. Such testing should not be confused with
process control and assurance testing of asphalt pro-
duction referred to in Chapter 5.
Job Mix
The final selected mix is called the job mix (or nomi-
nated mix).
The job mix will nominate the type and source of
components, target grading and binder content, and
volumetric properties of the mix. These are used as the
basis for manufacturing process control. Production
tolerances allow for variations in the mix composition
due to changing feed rates, raw material fluctuations,
sampling and testing limitations, etc.
In many cases, manufacturing process control is
applied to grading and binder content onlS although
some agencies also require volumetric testing, and pos-
sibly some mechanical testing, to monitor consistency
of production (see Chapter 6).
Any change in the type or source of components, or
significant variation in the proportion of any com-
ponent, generally requires redesign of the mix and
determination of a new job mix.
4,1.3 Australian Provisional Guide to
Asphalt Mix Design
The "selection and design of asphalt mixes: Ausrralian
provisional guide" was published by Austroads in
1997 (revised 1,998) as APRG Reporr No.18. It repre-
sented the outcome of ten years of co-operative
research by the Australian road agencies and asphalt
industry to develop performance based methods for
the characterisation of asphalt mixes.
The provisional status of the guide recognises the
extent of change and the tentative nature of some
aspects of performance criteria. That provisional sta-
tus will change as greater confidence is developed in
 Mix 5 specimens at Mix specimens at
3 binder contents design bitumen content

Conditionl hatl50C
Conditionl hatl50C

Gyropac at 10, 50, 80,

1 20, 350 cycles

At x compaction cycles

Design bitumen contenl

Figure 4.2 Austroads mix design procedure

(from Austroads se/ectlon and Design of Asphalt Mixes: Australian Provisional Guide)
-

use of the test procedures and interpretation of results. 4.2 Asphalt Mix Behaviour
The aim of the guide is a mix design procedure that:
Asphalt is a visco-elastic material that displays both
. is performance related
elastic and viscous characteristics under load, depend-
. enables the in-service performance of mixes to be
ing on the circumstances. Asphalt's response to load-
predicted
ing is both temperature and time dependent.
. is relatively affordable (in terms of new equipment
A simple model of the behaviour of asphalt, illus-
cost)
trating both elastic and viscous components of behav-
. is rapid and easy to use.
iour, is given in Figure 4.3.
The design procedure for dense graded mixes is
At low service temperatures and short times of
based on three levels as shown in Figure 4.2.
loading elastic behaviour dominates. Elastic stiffness
The first level involves:
under these conditions is used to calculate the critical
. selection of mix type and component materials
strains and fatigue behaviour of asphalt mixes applica-
likely to satisfy the performance requirements
ble to analytical pavement design.
r mixing of materials and conditioning of the mix
Typically, testing for resilient modulus, flexural stiff-
prior to testing to simulate medium term service
ness and fatigue is undertaken at 20"C or 25'C.
conditions
At high temperatures and with long times of load-
. gyratory compaction the gyl. ;tory compactor
-
enables selection of different compaction levels to
ing viscous behaviour dominates. Resistance to plastic
flow is used to determine deformation characteristics
match service conditions as well as simulation of long
term heavy traffic by compaction to "refusal density"
. determination of volumetric properties and selec-
tion of design binder content. Plastic dashpot
(oilfilled - \
Level two testing involves the application of tests to time & Elastic Spring
define structural performance and, if required, mois- temperature
ture sensitivity of the mix. dependent)

Level three testing is used to evaluate the rutting

resistance of mixes intended for use on pavements sub-
jected to high traffic loading.
Figure 4.3 Model of asphalt behaviour

43
or rutting resistance at high pavement service tempera-
tures and slow rates of loading. Tests involving the
plastic deformation properties of mixes, including
stability, dynamic creep or wheel tracking, are gener-
ally performed at either 50"C or 50"C.
4.3 Selection of Mix Type and
Component Materials
4.3.1 General
The selection of mix rype and performance application
will largely determine:
. aggregate quality and shape requirements
. filler requirements
r binder type
. volumetric requirements for the mix including, com-
bined aggregate grading, range of binder content,
and voids requirements.
A guide to the selecion of mix type is provided in
Chapter 2.
It is important to appreciate that asphalt mix design
is generally a compromise between different proper-
ties, for example:
. increasing the stiffness and rutting resistance of the
mix through the use of coarse gradings, rough tex-
tured aggregates and stiffer binders may reduce
workability and reduce flexibility of thin layers;
. improving the durability of mixes for light traffic
applications through the use of low air voids, high
binder contents or softer binders, increases the risk
of rutting and texture loss under heavy traffic.
A guide to the influence of binder stiffness, binder
content, coarse or fine aggregate grading, aggregate
characteristics and design air voids of a dense graded
asphalt is shown in Table 4.1.
Tabll +.r: Relationship Between Binder and Aggregate properties and Air voids
on the Properties of Asphalt
Structural stiffness
Fatigue resistance
(structural layers)
Fatigue resistance
(thin surfacing)
Deformation resistance
Durability
Workability
Skid Resistance
M
4.3.2 Aggregates
Aggregates should be clean, tough, and durable and
have adequate particle shape, strength and polishing
characteristics to meet the particular asphalt require-
ments. Aggregate shape and texture can also affect
workability and deformation resistance of asphalt
mixes.
Characteristics of different aggregate rypes are dis-
cussedin Chapter 3.
4.3.3 Mineral Filler
Fillers may vary in terms of:
r particle size distribution
. chemical reaction to other constituent materials.
AccordinglS not only is the initial selection of a
filler important, the substituting of one type of filler
for another can have a significant effect on the asphalt
mix.
A detailed discussion on filler properties is given in
Section 3.7.
4.3.4 Binder
The required properties and performance of asphalt
significantly affect the choice of binder. The type and
grade of binder (usually bitumen) depends very largely
on the climate and traffic volume to which the pave-
ment will be subjected.
Basic grades of bitumen for general applications in
Australia are Class 170 and Class 320 (B80 and 860
in New Zealand).
Class 600 bitumen (B40) is used in some asphalt
base applications to provide a stiffer mix and reduced
pavement thickness.
Low
Low
 Multigrade bitumen and polymer modified bitumen
binders are used in wearing course mixes and some
structural applications, primarily where increased
deformation resistance or increased flexibility are
required.
Binder characteristics and selection of binders for
specific applications are discussed in detail in Chapter 3.
Other additives such as fibres and naturally occur-
ring asphalt residues (e.g. Gilsonite and Trinidad Lake
Asphalt) may also be used to modify the properties of
bitumen binders.
For cold mix applications, bitumen may be emulsi-
fied or softened with a cutter oil or flux oil.
4.4 Combining and Proportioning
of Aggregates
4.4.1 General
The combined grading, which influences the density of
the mix and the voids in mineral aggregate, is defined
by the percentage, by mass, of particles passing each
of a series of ASA{ZS sieve sizes.
Specifications typically limit the minimum and maxi-
mum percentage of aggregate passing each sieve size.
These limits may be displayed graphically as a com-
bined grading envelope within which the grading of
the nominated mix must fall.
4.4.2 Maximising Packing Properties of
Dense Graded Mixes
Dense graded asphalt mixes use a continuous grading
to facilitate packing of aggregate particles to achieve:
. flexibility
. durability
. structural stiffness
. deformationresistance
. low permeability.
The maximum densiry grading gives the theoretical
densest particle packing and minimum voids in the
mineral aggregate (VMA) condition of the mix (the
total voids within the mass of compacted aggregate).
For a particular maximum aggregate size, the maxi-
mum density grading may be determined by the Fuller
equation:
P = rcoeDy
where
P = total percentage passing a given sieve
d = of sieve opening
size
D = maximum size of aggregate
n = grading exponent.
The original Fuller equation utilised the grading
exponent of 0.5. Further work has determined that an
exponent of 0.45 provides the greatest density and
lowest VMA. Increasing the exponent to, say 0.6
provides an asphalt mix with slightly increased VMA
and coarser texture. Decreasing the exponent also
increases VMA but results in a finer textured mix.
Higher exponents such as 0.8 or 0.9 tend towards an
open graded mix.
The Fuller equation has proved to give impractical
proportions of filler for some mixes. Another equation
has been developed to adjust for the percentage filler
content.
P=(100-F)(d'-0.075")
P- +F
(D" - 0.07 5")
where
F = the percentage of filler
P, d, D and n are as above.
Maximising the density of the mix with a particle
size distribution that closely follows the maximum den-
sity relationship may result in a VMA that is too low. A
low VMA may not provide enough space for the
required air voids and a reasonable volume of binder to
provide cohesion, durabiliry and fatigue resistance in
the mix. Overfilling the VMA with binder in such cir-
cumstances can lead to mixes with low air voids and
poor ruffing resistance (see also Section 4.6).
Deviations from the maximum density curve gener-
ally result in higher VMA. Significant deviations from
the maximum density curve can, however, result in a
reduction in mix stability. It can also result in a signifi-
cant change in finished surface texture by becoming
"gap" graded. It is desirable, therefore, that changes
to the aggregate grading combination to achieve
higher VMA, maintain a continuous distribution and
interlock of aggregate particle sizes.
4.4.3 Grading Envelopes
Aggregate grading envelopes (as used in specifications)
have been developed through accumulated field
experience and maximum density relationships to
define the grading limits within which mixes should be
designed.
Mixes designed within grading envelopes can still
vary widely in performance. The target grading should
not move widely from one.side of the grading enve-
lope to the other but maintain a uniform distribution
of aggregate sizes taking into account the grading of
available materials and any adjustments necessary to
achieve the desired volumetric properties.
A rypical grading envelope and target based on a
maximum density of n = 0.45 is shown in Figure 4.4.
E
 tttttt
tttttl
tttttl Yi I
I
Glading ehvebpelfor nea{ry dury I I
- - - -r- -- - J - - - - t-- --t- - - _ r_
lttttt
_ __ l
'lnixes-l I - -t-
I I
tttttl I I
lllllr I I
t') n = 0.4s'maximum densltv curve I I
'6 I I I t' I I I
o
(g - ---r- rttt-- _l- ___!____t_ ___ ) _ I
T
__t_
I
o- I I
o drading b-nvelo$ ror -oj I
light traffic mixes
I
c')
6 )
I I

cq, I
I
I

I
o f - -t-
c) I
(L I
I
I
I

1 .18 4.75 9.5

Sieve size (mm)

Figure 4.4 Typical grading envelope for 14 mm dense graded mix

4.4.4 Combining Aggregates to a Target
Grading
The proportioning of the selected aggregate fractions,
together with the filler, may be determined by trial and
error, graphically, arithmetically, by computer spread
sheet, or by using computer software such as "Cmix"
by ARRB TR.
Determining the combined grading of the aggre-
gates is achieved arithmetically using the following
formula:
P=Aa+Bb+Ccetc.
where
P = the percentage of material passing a given sieve for
the combined aggregate fractions A, B, C, etc.
A, B, C etc. - percentage of material passing a given
sieve for each of the aggregates A, B, C, etc.
a, b, c etc. - proportion of aggregates A, B, C, etc.
used in the combination and where the total of a, b, c
etc. = 1.00.
laboratory conditioning step that simulates the binder
hardening that occurs during manufacture and placing
of the mix for approximately the first two years of
field service. This step is particularly important for
mechanical tests such as resilient modulus, creep and
fatigue that are influenced by the stiffness of the
binder.
Procedures for mixing, quartering and conditioning
of laboratory samples are described in AS 2891.2.I.
Methods of compaction of samples for volumetric
and mechanical testing include:
. Marshall
. Hubbard-Field
r gyratory
. slab compactor.
Marshall compaction has traditionally been the most
widely used international compaction method for pre-
paration of laboratory specimens. This is now being
phased out in many countries in favour of gyratory
compaction. The use of Hubbard-Field compaction has
been totally discontinued, both in Australia and over-
SCAS.

Proportioning of aggregates can also be carried out
by graphical methods (see Rothfuchs method in
Appendix A4.1).
4.5 Mixing, Conditioning and
Compaction of Samples
4.5.1 General
An important aim of asphalt mix design procedures
that specimens prepared in the laboratory have proper-
ties similar to asphalt placed in the road.
The Austroads mix design procedure includes
The Austroads mix design procedure uses gyratory
compacted specimens. It produces cylindrical samples
suitable for resilient modulus, dynamic creep and
moisture sensitivity resring.
Cylindrical samples are also produced using
Marshall compaction although it is considered that
gyratory compaction achieves alignment of aggregate
particles that is more representative of field placement
of asphalt and is preferred where performance proper-
is ties are to be measured.
Where larger samples are required for wheel track-
ing or fatigue tests, a variety of methods are used
a involving compaction in rectangular moulds or large

45
 slabs that simulate the field placement of asphalt.
Rectangular samples can also be cored for use in tests
that require cylindrical samples.

4.5.2 Marshall Compaction

The Marshall compaction method involves the manu-
facture of cylindrical specimens 102 mm diameter by
63 mm high using a standard compaction hammer and
cylindrical mould as described in AS 2891.5. For most
applications, 50 blows of the hammer are applied to
each face of the cylinder. A higher standard of com-
paction, involving 75 blows per face, is used for air-
field work and some heavy duty road pavements.
Marshall compaction is limited to mixes with an
aggregate size of 28 mm or smaller.
Mechanical testing for stability and flow is des-
cribed in Section 4.11and AS 2891.5.

4.5.3 Hubbard-Field Compaction

The Hubbard-Field method of compaction uses a larger
briquette (i.e. 152 mm diameter by 75 mm high) than
the Marshall Method. This enables its use with mixes
having a maximum aggregate size of 40 mm.
Specimens are prepared by compacting in two lay-
ers in preheated moulds using 40 blows as described in
AS 2891,.6. Final compaction is achieved using a com-
pression machine, and the specimen is cooled.
Mechanical testing for stability is described in
Section 4.12 and AS 2891.6
r .
4.5.4 Gyratory Compaction
The principle of gyratory compaction is illustrated in
Figure 4.6. Compaction is achieved by shearing forces
obtained by the application of a constant vertical com-
pressive force to the asphalt confined in a cylindrical
mould while the mould is rotated about its vertical
axis through a small angle. The angle is maintained
constant throughout the compaction process.
Compaction can be terminated after a set number of
cycles or at a set height representing a predetermined
volume and density of asphalt. Monitoring the height of
the cylinder during the process enables the density to be
estimated at different numbers of cycles. This data can
be used to determine density at different compaction
levels and provide an indicator of mix workability.
Two forms of gyratory compactor are commonly
used in Australia. The Gyropac (Figure 4.5) is used for
routine mix design and monitoring tests. The Servopac
is a more advanced version that includes more options
and control over the compaction process that make it
more suited to research applications.
For routine mix design, the level of compaction
depends on the traffic level as follows:
Figure 4.5 Gyropac
. light traffic 50 cycles
medium traffr,c B0 cycles
e heavy traffic 120 cycles
r 'refusal'density 350 cycles
Test procedures are described in AS 2891..2.2
Voids and density properties are determined as des-
cribed in Section 4.6.
4,5.5 Preparation of Slabs and Large
Samples
Asphalt characterisation tests require specimens of
uniform composition and representative of materials
placed in the road. Criteria include uniformity of den-
sity and particle distribution, particle orientation con-
sistent with in-situ materials, and absence of undue
change in particle size distribution due to the com-
paction process.
The preferred procedure for obtaining specimens
for those characterisation tests that require asphalt
slab specimens involves compaction in rectangular
moulds using a rolling wheel or segmental compactor
as described in Austroads Test Method AST 05.
Other methods of rectangular sample compaction,
or compaction of slabs using scaled down field

47

I
 Force
Gyratory
angle
Figure 4.6 Principle of gyratory compaction
compaction procedures may also be used provided
that they achieve the required uniformity of density
and orientation of aggregate particles consistent with
in-situ pavement materials, without crushing of aggre-
gates and significant change in particle size distribu-
tion.
Slab compaction methods may also be used for
preparation of cylindrical specimens obtained by cor-
ing the completed sample.
4.5.6 Compaction to Refusal Density
Refusal density is a measure of the maximum degree
of packing of aggregates in the asphalt mix that is pos-
sible without degradation of the aggregate. It is used
to simulate the in-situ density to which an asphalt mix
can be compacted under extreme conditions of traffic
loading.
Asphalt mixes compacted to an in-situ air voids
level of less than about 2"/" to 3"/o can become un-
stable and prone to rutting as a result of binder almost
totally filling the void spaces and reducing aggregate
particle interlock and friction. Refusal density is used
to identify the potential for such to occur.
In the Australian mix design procedure, refusal
VOIDS JN
MINERAL
AGGBEGATE
iOIAL
AGGREGATE
VOLUME
AGGREGAT AGGREGATE
(PABTICLE
DENSITY ON
A DBY BASIS)
Figure 4.7 Constituents of a compacted dense graded asphalt mix
48
density is determined by compacting to 350 cycles in
the gyratory compactor.
An alternative measure of refusal density is the
British Percentage Refusal Density (PRD) test (BS 598
Part 104). The test uses a vibratory hammer to com-
pact a heated sample in a cylindrical mould. A split
mould is used to enable testing of field core specimens.
The PRD test was originally developed to establish
a reference density for assessing the effectiveness of
field compaction of asphalt designed to recipe specifi-
cations. It was also found that low final voids in the
PRD test correlated with mixes prone to rutting, hence
the use of refusal density as a design tool to identify
mixes with potential for low in-situ voids.
4.5.7 Workability
'Workability
is the ease with which a material may be
handled, placed and compacted to a dense uniform
mat.
The major factors influencing workability are:
. binder viscosity
. binder content
. filler type and content
. nominal size of mix
I aggregate grading
. type of aggregate (crushed or rounded)
r temperature of placing.
There are no defined tests for workability although
an indicator of workability may be obtained from the
behaviour of a mix under gyratory compaction.
Workable mixes are indicated by mixes that
undergo rapid volume change and increase in density
after a few cycles of gyratory compaction and then a
slower rate of increase in density to the refusal density
condition.
Harsh mixes, with poor workability, are indicated
by mixes that do not increase as rapidly in the earlier
stages of gyratory compaction but continue to increase
in density with further compaction cycles. Such mixes
require greater effort to achieve field density targets.
TOTAL
BINDER
_J VOLUME
ABSORBED BINDER
TOTAL VOLUME
OF COMPACTED
ASPHALT
(BULK DENSITY)
VOLUME OF
EXCLIJDING AIR
VOIDS
(r\,4AXlMUtu
DENSITY)
4.6 Volumetric Characteristics also be used as the reference density for determining
relative density of field compaction.
4.6.1 General AS 2891.8 provides three methods of determining
bulk density of compacted asphalt. Each method will
Asphalt mix design is essentially a volumetric process
give slightly different results depending on the surface
although, for convenience of manufacture and process
texture and water permeable voids and this must be
control, it is simpler to measure the mix proportions
considered in both selection of test method and inter-
by mass.
pretation of test results.
Important measures of volumetric properties are:
. bulk density
. maximum densiry Waxing procedure eliminates the influence of water
. permeable voids by sealing the surface with a layer of
Voids in Mineral Aggregate (VMA)
. wax. It is used for dense mixes where it is desirable to
air voids
. absorbed binder
avoid the influence of water permeable voids on the
. measurement of bulk density. An alternative to waxing
effective binder content.
. is the use of silicone to seal the external surface of the
voids filled with binder
. sample against the entry of water. Removal of wax
binder film thickness
and silicone can be difficult for samples that are to be
\Iolumetric relationships are illustrated in Figures
used for further testing.
4.-.4.8 andTable 4.2.

Presaturation method is suitable for dense mixes

4.6.2 Bulk Density with low permeability and internal air voids that are
Bulk densiry is the weight per volume of the com- largely inaccessible to moisture. Significant permeabil-
paced mix including internal air voids. Bulk density is ity can result in a misleading measure of apparent den-
rhe basis for calculation of voids relationships. It may sity. Influence of water permeable voids is reduced by

Absorbed bitumen

Water permeable
porosity, not filled
with bitumen
Aggregate volume
(particle density on
a dry basis)
Aggregate volume
(effective density)

Aggregate volume
(apparent particle
density)

Figure 4.8 Schematic diagram showing VMA, air vords and effective bitumen content in compacted asphalt mixtures

Table 4.2: Typical Proportions for Dense Graded Asphalt

By Volume

tr"
By Mass

4'
saturating the specimen in water before measuring
mass displaced by water and mass in air after removal
of surface water from the specimen.
Mensuration method refers to determination of the
volume of the specimen by direct measurement of the
external dimensions. It is generally used for open
graded and other porous mix types.
In dense graded mixes, bulk densiry tends to
increase to a maximum value and then decrease with
increasing binder content.
4.6.3 Maximum Density
Maximum density, also termed voids free bulk density,
is the density of the mix excluding air voids.
Testing is performed by determining the volume of
water or methylated spirits displaced by a loose
sample of mix after removal of all air.
Maximum density determined in this manner is also
-sometimes referred to as maximum theoretical densitS
i.e. the maximum density to which a mix could be theo-
retically compacted if there were no air voids present.
In Australia, the term maximum theoretical density is
usually applied to the calculation of density from the
combined density of all the components, and not the
measurement of maximum density by test.
The difference between maximum density and bulk
density is the air voids in the mix.
Maximum density is sometimes used as the refer-
ence density for calculating relative density of field
compacted samples as an alternative to bulk density.
A further use of maximum density testing is in qual-
ity control of asphalt production. It is a relatively
simple test that gives warning of changes in propor-
tions of components having different density without
the necessity for complete analysis of the mixture.
4,6.4 Voids in Mineral Aggregate (VMA)
The voids in mineral aggregate (VMA) is the total
volume of voids within the mass of compacted aggre-
gate. It is a function of the particle size distribution,
the shape and surface texture and packing characteris-
tics of the aggregate particles.
VMA is the combination of air voids in the com-
pacted mix and the volume occupied by effective
binder (excluding any binder absorbed into the aggre-
gate).
It is important that the VMA be large enough to:
. allow space for sufficient binder so that the mix will
be durable
. allow sufficient air voids in the compacted mix to
ensure the mix is stable.
Typical values of VMA in dense mixes can be 13%
to 20oh, depending on the nominal size of the mix.
50
The VMA affects the performance of the mix as fol-
lows:
VMA too low:
r potential for over-filling of voids with binder result-
ing in flushing or bleeding or instability
. insufficient binder for cohesion and durability.
VMA too high:
. high air voids
. increased binder volume to satisfy air voids require-
ments resulting in a mix that is more costly to pro-
duce.
Increases in VMA may be obtained by:
Changing the particle size distribution. Moving
"
away from a maximum density curve or moving to
a higher exponent of maximum density (see Section
4.4) will increase VMA. Generally, it is preferable to
maintain a uniform particle size distribution than
introduce gaps in the grading.
. Reducing the minus 0.075 mm (filler) fraction. This
will also increase binder film thickness but reduce
stiffness of the mix.
. Changing one or more of the aggregate fractions for
one that has different shape and/or surface texture
characteristics and hence different packing proper-
ties.
Decreases in VMA may be obtained by:
. Changing the particle size distribution to move
closer to a maximum density relationship.
. Increasing the filler content.
. Changing one or more of the aggregate fractions for
one that has different shape and/or surface texture
characteristics and hence different packing proper-
ties.
4.6.5 Air Voids
The air voids content in a mix is a function of:
r VMA (aggregate particle size distribution)
. binder content
. level of compaction.
The air voids content of a mix affects the stability
and durability. In general, mixes should be designed to
have the lowest practical air voids value in order to
reduce:
. the ageing (oxidation) of the binder
r the possibility of water penetration, and hence strip-
ping of binder from the aggregate.
If the air voids content of asphalt in service is too
low (less than 27" to 37"), plastic flow may occur
resulting in flushing, bleeding, shoving or rutting of
the pavement.
The air voids content also influences other proper-
ties such as mix stiffness, fatigue strength and worka-
bility.
 4.6.6 Binder Content
In general, the optimum binder content is a balance
between being high enough to ensure durability and
life for the pavement but not so high that the mix
becomes unstable.
The binder content has a major influence on the
volumetric and mechanical properties of an asphalt
mix. A small change can significantly affect the perfor-
mix.
v / +.1.t General
mance of the
The optimum design binder content is dependent
on:
. aggregate type
. aggregate particle size distribution
. compaction level of the mix
. design air void content.
Asphalt performance depends on its effective binder
content, excluding any binder absorbed uy'rrr.
gate, and not its total binder contenr.
"t;;;:
The proportion of binder absorbed by aggregat.
depends on the porosity of the aggre*"*
of binder. Typically, the amount of binder
"";fi;-r";; fy '
"brorb.j ;
assresate is 0.3 to 0.7 times the water
"bt;;;;
the aggregate (see sectio n 3.4.4and 3.4.5).
Binder absorption is generally determined from rhe
measured maximum density of the mix (4.5.3) ";;; and a :t"t:t:
theoretical maximum density determined il ' a,1 fow temperatures' and/or
combined bulk density of aggregates, filler Jil;;.
Accurate measurement of the density .f ;;;.;;;
and filler is an essenti aI part of estimation of absorbed
binder.
4.6.7 Voids Filled with Binder (VFB)
Voids filled with binder (VFB) is the proportion of
VMA occupied by effective binder. This is generally in
the range of 65Y" to 80%.
At low VFB values, approximating 60"h, the mixes
become dry lacking coh.rio.r, durability and fatigue
resistance. Mixes with low VFB may also be more per-
meable.
When VFB approximates g5% or more, mixes can
become unstable and susceptible to rutting.
High vFB is generally associated with high vMA
while a low VFB is an indicator of low VMA. For
making adiustments to VMA, see section 4.6.4.
4.6.g Binder Film Thickness
Binder film thickness is calculated as a function of the
surface area of the aggregates and filler, and the effec-
tive binder content.
A minimum binder film thickness may be specified
to ensure that there is sufficient binder in the asphalt
mix to ensure adequate cohesion, durability, fatigue
resistance, and resistance to the effects of moisture.
51
I
Binder film thickness may be increased by:
. changing the particle size distribution to increase
the proportion of coarse to fine material;
. increasing the VMA to allow more binder for the
same air void content.
4.7 Mix Stiffness
n.,r,
Stiffness indicates the response to load of the mix and
may be used to indicate the load spreading abilitv of
the asphalt pavement'
Modulus is the ratio of applied stress to strain' It is
used to define the stiffness of a mix and is primarily
dependent on:
the mix
:r "!:.nf
stiffness of the binder
' volumetric composition' including effective binder
and air voids in the compacted mix
::1""t
teTperature
of loading
: :]lt
r degree of compaction'
t: fullv elastic material' but displays
, 1:t111 -l:l "
propertres:
' short duration of loading (i'e' high speed traffic);
while displaying increasing viscoelastic characteris-
.
ttttt.
. ,
at high temperatures, and/or
r long duration of loading (i.e. low speed traffic).
The stiffness of an asphalt mix may be increased by:
. decreasing the binder content
increasing the angularity of the aggregate
' adjusting the filler content
r uSlng a harder binder.
Resilient modulus is the ratio of stress to recover-
able (resilient) strain, and is the standard for the
measurement of modulus'
The resilient modulus is necessary for pavement
designers. The Austroads "Guide to the Structural
Design of Road Pavements" provides a procedure for
estimating stiffness modulus of asphalt, developed by
Shell, using a combination of binder stiffness, esti-
mated using the Shell Nomographs, and the volumet-
ric properties of the mix. Procedures for estimating
design modulus from laboratory measured modulus
have also been developed.
Typical laboratory values for various Australian
mixes are shown in Table 4.3.
The Australian mix design procedure provides for
the measurement of mix stiffness using the resilient
modulus test. Mix stiffness can also be determined in
the fatigue test (Sectibn 4.9)
 for Typical Australian Dense Graded Asphalts
Table 4.3: Modulus Values (MPa)
Determined on Laboratory-Manufactured Samples Using the Indirect Tensile Test
Procedure with Standard Test €onditions' (from APRG, 1997\

cl. 170

c|.320
cI.500
Multigrade

SBS

EVA

'Standard test conditions are 40ms rise time and 25'C test temperature

4.7.2 Resilient Modulus test J
The resilient modulus of asphalt is measured by apply-
ing a dynamic force across the vertical diameter of a
cylindrical specimen, positioned on its side between
two concave loading platens (diametral loading). This
produces an "indirect tensile" loading pattern in the
specimen, and the resilient modulus may be calculated
from measurements of the peak applied force and the
resultant peak horizontal diametral displacement. By
varying the rate of loading and temperature, the test
may also be used to determine resilient modulus for
different traffic speeds and climatic conditions.
Indirect tensile test apparatus is shown in Figure
4.9. Testing is undertaken in a temperature controlled
cabinet (Figure 4.10).

4.8 DeformationResistance
4.8.1 General
Permanent deformation in asphalt usually manifests
itself as rutting that results from gradual consolidation
and lateral deformation of the asphalt in the wheel
paths.
The worst conditions for rutting are:
. high temperatures
. heavily loaded vehicles
. channelised traffic
. slow moving vehicles.
The deformation of an asphalt layer can be related
to a combination of the following causes:
. densification of the mix
r plastic deformation.
Densification involves a change in volume of the
mix, and can be controlled by:
r aggregate particle size distribution and shape
. binder content
. appropriate field compaction.
Plastic deformation is predominantly a function of the
binder, or the mixture of the mineral filler and binder. It
Figure 4.9 lndrrect tensile testing jig will be largely controlled by the binder properries.

n

Figure 4.10 Temperature controlled cabinet for indirect tensile and
dynamic creep testing
Mixes with inappropriate aggregate particle size
distribution or rounded particle shape, or very low
voids content may be unstable resulting in uncon-
trolled plastic deformation.
A large number of devices have been developed for
measurement of deformation characteristics of asphalt
using variations of static and dynamic creep or wheel
tracking test devices. The Australian mix design proce-
dure provides for two different methods of test. The
dynamic creep test is performed on cylindrical samples
in the MATTA apparatus. The wheel tracking test is
performed on slabs of asphalt using a rolling wheel.
Compaction to refusal density (Section 4.5.6) is used
as a further criteria for avoiding excessive rutting.
4.8.3 Wheel Tracking Test
'Wheel
tracking has been found to correlate well with
rutting performance in field tests and is widely
accepted internationally. It has been found to be a
more reliable predictor of rutting performance than
the unconfined dynamic creep test.
The test procedure is described in AST 01 and con-
sists of the following:
. Laboratory prepared samples of 300 mm x 300 mm
with thickness of 50 mm or 75 mm depending on
nominal mix size, or thickness of 35 mm to 110
mm from field samples.
Testing consists of repeated application (42 back
and forward motions per minute) of a standardised
wheel (200 mm diameter) under a standard load
(700 N) at a standard temperature of 60"C. A stan-
dard air voids content (5%) is used when compar-
ing different mixes.
The test is concluded after 10 000 cycles or when
the rut depth exceeds 15 mm.
Results are interpreted in terms of the steady track-
ing rate (mm per 1000 load passes after initial 4000
passes) and final rut depth or number of passes to
reach maximum 15 mm depth.
4.9 Fatigue
,4.9.1 General
Fatigue is a phenomenon caused by the repeated bend-
ing of a material such that tensile stresses are induced
in the material. This process gradually weakens the
material (i.e. loss of stiffness) to the point where a
crack develops in the zone of maximum stress.

4.8.2 Dynamic Creep Test

Creep can be defined as the slow and progressive
deformation of a material after a time under constant
StTCSS.

In the dynamic creep test (AS 1289.12.1), the cylin-

drical specimen is mounted on its end between two
polished disk platens. A dynamic axial force is repeat-
edly applied, and the variation in vertical specimen
height is recorded to determine the permanent strain.
This test gives an indication of susceptibility of an
asphalt mix to rutting. The minimum creep slope can
be used to rank mixes. A typical curve from a dynamic
creep test is shown in Figure 4.11.
Correlation with full scale tests has shown that the
dynamic creep test provides good ranking of deforma-
tion resistance for binder variations in similar mix types
but poor ranking between different mix types. The
dynamic creep test is undergoing further development in Loading Cycles
terms of specimen dimensions or use of lateral confine-
ment to improve correlation with field performance. Figure 4.11 Typical dynamic creep curve (including slope curve)

53
 Table 4.4: Effect on Dynamic Creep r thin layers
(from Austroads, 1gg7) . surface courses and other non-structural applications.
These applications generally require a flexible mix.
A harder grade of binder will decrease fatigue life.
An increase in the fatigue life of thin asphalt layers
may be achieved by using:
lncreased bi nder content Reduce . softer binder
lncreased binder viscosity lncrease
. higher bitumen content
Grading (coarse to fine) Reduce
. polymer modified or multigrade bitumen.
In general, it has been observed that the in-service
lncreased load duration Reduce
fatigue life of the asphalt is dependent on the binder
lncreased compaction level lncrease
content and the density of the mix.
lncreased temperature Reduce The relationship between the stiffness and fatigue
behaviour of asphalt mixes, and the factors affecting
these properties are shown in Table 4.5.
There are basically two different laboratory test con- In the Austroads "Guide to the Structural Design of
ditions for fatigue tests: Road Pavements", fatigue performance is estimated in
. constant stress (constant load) terms of the formula developed by Shell based on mix
. constant strain (constant deflection). stiffness, binder volume and tensile strain in the pave-
Different applications require modification in mix ment. This relationship has been found satisfactory for
design to satisfy fatigue requiremenrs. conventional binders but is not applicable to modified
The constant stress condition is associated with binders and some special mix types. Fatigue perfor-
pavements where the asphalt is the predominant struc- mance determined by laboratory test or the Shell for-
tural layer. Such conditions are met in: mula require the application of a "shift factor" that
r thick base courses takes into account self-healing and other effects influ-
. full depth and deep strength asphalt pavements. encing the in-service behaviour of asphalt.
In these applications a harder grade of binder will
increase the stiffness of the mix, reducing deformation I

under load and increasing pavement fatigue life.

J 4.9.2 Fatigue Test
The constant strain condition is associated with An Australian apparatus has been developed for the
pavements where the asphalt is not the predominant measurement of fatigue by repeated loading of a beam
structural layer, such as: of asphalt (Figure 4.12).

Table 4.5: Relationship between Stiffness and Fatigue Life and Factors Affecting
these Properties

Binder hardness Decrease Decrease Decrease lncrease

Binder content lncrease Decrease 1
Decrease ' I ncrease
(to optimal binder (to optimal binder
content) content)
Aggregate type lncrease in Roughness lncrease I ncrease Decrease
and Angularity
Aggregate particle From open to dense lncrease lncrease Decrease
size distribution
Air void content Decrease lncrease lncrease lncrease
(i.e. increase in
mix density)
Temperature Decrease lncrease lncrease Decrease

Note:
' Stiffness and fatigue life reach optimum values at a binder content lower than the optimum binder content for maximum stability.

54

L-
 14.0

12.O 4
2
10_0 Ec
=
o c
6 I
o 3
8.0 O^
L

6.0 1

4.O 0
6.5 5.5

Bitu men Bitumen

10 20

8 18

;6 c16
!
'6
.:4 >14

2 12

0 '10
4.5 5.5 6.5
Bitumen Bitumen

2.480 100

6 2.460 90
E
e
Z 2.440
(D
80
c a L
o
?z+zo 70
a
d)
2.400 60

2.380 50
4.5 5.5 5.5
Bilumen Bitumen

Figure 4.14 Marshall test results

| ,o...
I
4.12 Modified Hubbard-Field
Stability
The Modified Hubbard-Field Method, which was the
basis of RTA NSW specifications used up until the late
1990s, is derived from the method developed by the
Corps of Engineers for airfield pavements.
Samples are compacted as described in Section Test specimen
4.5.4.
The prepared specimen is preheated then loaded
axially and forced through a ring using a compression
machine (Figure 4.15). The maximum load resisted is
the Modified Hubbard-Field stabiliry.
Interpretation of stability results and volumetric
relationships is similar to that described for the
Marshall method, above. Figure 4.15 Testing of Hubbard-Field specimen

57
4.13 Open Graded Asphalt The design of SMA is a combination of recipe and
measured test criteria.
There is no universal rational design system for open
Optimum performance is obtained when the fine
graded (OG) asphalt. Generally a recipe approach is
aggregate just fills the void space in rhe coarse aggre-
used based on an aggregate particle size distribution
gate particles while still retaining inter-particle contact
for texture and voids content and the type and content
of the large particles.
of binder for cohesion and durability.
The particle size distribution may be selected from
Polymer modified binders are frequently used in OG
standard grading envelopes, developed from experi-
mixes to provide greater cohesion and durability. They
ence, or using a rational design procedure that meas-
may also allow high binder contents to be used without
ures the change in density and hence dilation point
drainage of binder from hot asphalt during transport.
obtained by progressively adding increased propor-
Fibres may also be used to inhibit drainage of binder.
tions of fine material.
Generally, the durability of open graded mixes is
There is no procedure for establishing optimum
aided by the use of the highest binder contenr prac-
proportions of filler and binder and these are generally
ticable, but not so high that significant drain-down
selected from standard envelopes. Minor adjustments
and segregation of binder occurs during handling and
may be made to satisfy voids criteria in the mix.
transport.
Target design air voids are similar to other dense
AAPA Implementation Guide No.1 ("Open Graded
mixes, i.e. around 4"/o to 5"/".
Asphalt Design Guide") provides advice on the use of
Design of SMA mixes may also incorporate a check
two tests to assist in selecting the type and content of
for binder drain-off (AST 06).
binder in OG asphalt.
The Asphalt Binder Drain-Off Test (AST 06) can be
A guide to SMA mixes is provided in AAPA
Implementation Guide No.4 "Stone Mastic Asphalt
used to establish the maximum amount of binder that -
Design and Application Guide".
can be used without excessive drain-off of binder or to
Typical proportions of components of SMA are
evaluate the influence of different binder types or addi-
shown in Table 4.6.
tives. The test may also be used to establish the maxi-
mum temperature for mixing and transport for a given
binder type and binder content.
4.15 Fine Gap Graded Asphalt
The Asphalt Particle Loss Tesr (AST07) measures Fine gap graded mixes are largely designed on a recipe
the loss of aggregate particles of a sample subject to basis to match grading and binder content targets.
tumbling in a Los Angeles Abrasion Loss test apparu- Volumetric properties should, however, be checked to
tus. It gives an indication of the cohesion of the ensure appropriate air voids.
asphalt mix and hence a guide to the effectiveness of Guidelines for fine gap graded mixes are provided
binder type and content in resisting particle loss dur- in APRG Report No. 18 (see also Section 2.6.4).
ing service.

4.16 Recycled Asphalt

4.14 Stone Mastic Asphalt (SMA)
SMA is a gap-graded mix with a high coarse aggregate
content, high filler content, and high bitumen content.
Fibres are frequently used to assist in controlling
binder drain-off during handling and transport.
The high proportion of.coarse aggregate interlocks
to form a high stability skeleton (structural matrix)
with good internal friction. The voids are substantially
filled with a bitumen-filler mastic. Accordingly, SMA
retains good texture under heavy loading while having
low permeability.
The mastic has a high filler and bitumen content.
Reclaimed asphalt pavement (RAP) may be recycled as
a component in the manufacture of hot mix asphalt,
recycled using hot-in-place processes, or recycled cold
as an in-situ or plant mixed material. Hot-in-place and
cold recycling are covered separately in Chapter 8. {
Table 4.6: Typical Proportions of SMA
Components
Coarse aggregate 70-80o/o

The coarse aggregate skeleton must be able to contain (plus 4.75 mm)

all the mastic binder while maintaining the stone-to- Fine aggregate 12-18o/o
stone contact essential for rutting resistance. Too (4.75to 0.075 mm)
much mastic will result in flushing, bleeding and loss Filler 8-12o/o
of pavement shear deformation resistance. Too little (minus 0.075 mm)
mastic will result in high air voids, increased perme- Bitumen 6-7%
ability and reduced pavement durability. Fibres 0.3%

58
 Hot asphalt mixes incorporating RAP are designed
in the same manner as conventional mixes, using the
same volumetric criteria. The manufacture of recycled
asphalt may require RAP to be pre-treated by crushing
and screening as well as modifications to the asphalt
plant and mixing procedures. Manufacturing require-
ments for asphalt mixes incorporating RAP are
covered in Chapter 5.
The binder in RAP is generally stiffer than that of
new, freshly mixed asphalt. Addition of RAP to an
asphalt mix will, therefore, generally result in an
increase in mix stiffness leading to increased resilient
modulus and increased rutting resistance.
The increased stiffness may, however, result in a
decrease in fatigue life in thin surfacing applications.
Compensation for reduction in fatigue life may be
made by use of a lower viscosity binder in the new
mix or addition of a binder rejuvenating agent.
Addition of RAP in proporrions of up to about
l5To or 20% of the total mix is unlikely to signifi-
cantly change the performance characteristics of the
asphalt mix and no special requirements or adjustment
to binder grade are generally required.
Above about 20"/", and up to about 30% of RAP,
compensation for stiffening of the binder may be made
by the use of binder of one viscosity class lower than
that otherwise specified.
Adjustment of viscosity class of added binder may
be suitable for RAP proportions up ro 50% of the
total mix.
Alternative techniques for adjustment to binder
grade, or selection of binder rejuvenating agents, to
compensate for increased viscosity of RAP are des-
cribed in relation to in-situ recycling in Chapter B.
Selection of binder grade or rejuvenator for recycled
asphalt needs to consider the application. It needs to
take into account the appropriate balance between the
effect of increased binder stiffness on mix stiffness,
fatigue resistance and rutting resistance and the poten-
tial decrease in rutting resistance that may occur
through the use of softer binders or rejuvenating
agents in order to achieve lower stiffness and higher
fatigue resisrance.
Further details on recycling of asphalt may be
obtained by reference to the Austroads "Recycling
Guide" and the Austroads "Framework Specifications
for Asphalt Recycling".
4.17 Cold mixes
Cold mixes are largely designed on a recipe basis and
known field performance.
The design of cold mixes is often a compromise
between binders that remain workable at ambient
temperatures and after extended storage, and prob-
lems in service due to presence of cutter and flux oils
59
that lead to poor stability and a possibility of flushing
and bleeding. Generally, medium curing cutback bitu-
men binders provide more rapid curing and improved
field performance, but poorer workability and stock-
pile storage life, compared to cold mixes manufac-
tured with binders containing slower curing flux oils.
Bitumen emulsion binders (Section 3.1,3) are an
alternative to cutback bitumen binders in cold mixes.
Different emulsion binder types can be used depending
on mix type, aggregate condition and whether mixes
are intended for stockpiling or immediate use.
A guide to cold asphalt mixes suitable for main-
tenance patching is provided in AS 4283 "Cold mix
asphalt for maintenance patching".
A guide to the design of cold mix materials for bitu-
men stabilised base applications is provided in AAPA
- "Cold Mix Granular
Implementation Guide No.2
Materials Guide".
4.18 Mix Designs for Specialised
Applications
4.1a.1 Polymer Modified Binders
Asphalt containing polymer modified binders can be
designed in accordance with the procedures already
described for the relevanr mix type.
When designing dense graded mixes, some designers
prefer to prepare volumetric designs using conven-
tional binders and then substitute a polymer modified
binder with the desired performance attributes. This is
done, partly as conventional binders are easier to
work with in the laboratory but also as the presence
of polymers can mask subtle influences of aggregate
properties on the volumetric properties of the mix and
make the selection of optimum grading and binder
content more difficult.
In substituting a polymer modified binder into a
dense graded asphalt mix designed using conventional
binder, it is common pracice to add slightly more
binder, typically 03% by mass of the mix, to achieve
the same volumetric performance.
4.14.2 Crumb Rubber Asphalt Mixes
Crumb rubber blended into bitumen provides a binder
with performance characteristics similar to other elas-
tomeric PMB types.
Superficially, crumb rubber bitumen, directly substi-
tuted for conventional bitumen, will provide improved
rutting and fatigue resistance. In practice it has been
found that substitution of the same volume of binder
results in a mix that lacks adequate cohesion, and that
higher binder contents are required for equivalent per-
formance (see also Section 3.12.5 and APRG Technical
Note 9).
 It has also been found that the total proportion of
binder can be increased substantially without risk of
bleeding or instabiliry provided that the VMA is
adiusted to achieve a suitable air void content.
This feature has been exploited to manufacture
asphalt mixes with outstanding resistance to reflection
cracking and fatigue through the use of binder con-
tents (combined bitumen and crumb rubber) as high as
1,1% by mass of the total mix.
Manufacture using the "dry" process whereby the
rubber crumb is added direct to a pugmill mixer has
been found to be more effective in these mixes than
"wet" processes that pre-blend bitumen and rubber.
Binder requirements are generally established on a
recipe basis using, typically bitumen contents of 7oh
to 8''/o and crumb rubber contents of 1.8"/" to 2.5"/"
by mass of the total mix.
4.18.3 Ultra Thin Asphalt
Ultra thin asphalt comprises a range of asphalt mix
types used in asphalt layers that are typically less than
about 20 mm.
These include small nominal size dense graded, fine
gap graded, open graded and stone mastic asphalt
mixes as well as a range of mixes developed as hybrids
of those mix types, or as proprietary products.
The most common type of ultra thin asphalt is a
form of open graded mix that contains a greater
amount of graded fine aggregate to assist in shear
resistance but retains a coarse, open surface texture.
Such mixes are largely designed on a recipe basis.
Further description of this application is provided in
APRG Technical Note 8 Ultra thin asphalt surfac-
ittg.
- Duty Surfaces: The Arguments for SMA", (1998).
References
AAPA,Implementation Guide No.1, "Open Graded
Asphalt Design Guide", Melbourne, (1997).
AAPA,Implementation Guide No.2, "Cold Mix
Granular Materials Guide", Melbourne, (1997).
AAPA, Implementation Guide No.3, "Asphalt Plant
Process Control Guide", Melbourne, (1997).
AAPA, Implementation Guide No.4, "Stone Mastic
Asphalt Design and Application Guide", Melbourne,
(2000).
ARRB Transport Research, "Selection and Design of
Asphalt Mixes: Australian Provisional Guide"
Melbourne, (October 1.996).
Austroads, "Asphalt Recycling Guide", Sydney, (1997).
60
Austroads, Pavement Research Group Report 18,
"Selection and design of asphalt mixes: Australian
provisional guide", (1997).
Austroads, "Specification framework for polymer
modified binders", (2000).
Austroads, "Guide to Stabilization in Roadworks",
Sydnes (.999).
Austroads, "Framework specifications for asphalt
recycling", (2000).
Austroads, Pavement Research Group Technical Note
8 Ultra thin asphalt surfacing, (1999).
-
Austroads, Pavement Research Group Technical Note
10 The use of recycled crumb rubber in asphalt,
-
(leee).
The Asphalt Institute, "Mix Design Methods for
Asphalt Concrete and Other Hot-Mix Types",
Maryland, (1984).
Brown, E.R. and Haddock, J.E., Development of a
Mixture Design Procedure for Stone Matrix Asphalt
(SMA). Proc. Association of Asphalt Pavement
Technologists Meeting, Salt Lake Ciry (March 1,997).
Cominks, Ronald J., "The Superpave Mix Design
Manual for New Construction and Overlays, Strategic
Highway Research Program", Washington, (1,99 4).
European Asphalt Pavement Association, "Heavy
Goodrich, J.L., Asphaltic Binder Rheology, "Asphalt
Concrete Rheology and Asphalt Concrete Mix
Properties", Chevron Research and Technology
Company.
Maccarrone, S., Rebbechi, J. and Ky, A., "Evaluation
of Stone Mastic Asphalt Performance". 10th AAPA
International Flexible Pavements Conference, Perth,
(1.ee7).
Oliver, J.\f.H., "Developing an Australian Asphalt
Mix Design Method", RTA/AAPA Implementation
Workshop for New Asphalt Technology, Sydney,
(19e3).
Roberts, F,L. et al, "Hot Mix fuphalt Materials, Mixnue
Design, and Construction", NAPA Education Foundation,
Lanham, Maryland, Ed'1., (L991).
Universiry of Nottingham Department of Civil
Engineering, "Bituminous Pavements: Materials,
Design and Evaluation", (1990).

Australian Standards
AS 2150 Hot mix asphalt

AS 2891 Methods of sampling and testing asphalt

AS 2891.1 Sampling of asphalt
AS 2891.2.1 Sample preparation Mixing, quarter-
-
ing and conditioning of asphalt in the laboratory
AS 2891.2.2 Sample preparation Compaction of
-
asphalt specimens using a Wratory compactor
AS 2891,.3 Bitumen content and, aggregate grading
AS 2891.5 Determination of stability and flow
Marshall procedure
-
AS 2891.6 Determination of stability and flow by the
Modifi ed Hubbard-Field procedure
AS 2891..7 Determination of the maximum density of
asphalt
AS 289L.8 Voids and density relationships for com-
pacted mixes
AS 2891.9 Determination of bulk density of com-
pacted asphalt
AS 2891.12.1 Determination of the permanent com-
pressive strain characteristics of asphalt
Dynamic creep test
-
AS 2891.13.1 Determination of resilient modulus of
asphalt Indirect tensile test
-
AS 4283 Cold mix asphalt for maintenance patching.

Austroads Manual of Test Methods

AST01 Deformation resistance of asphalt mixtures by
the wheel tracking test
AST02 Stripping potential of asphalt Tensile
strength ratio
-
AST03 Fatigue life of compacted bituminous mixes
subject to repeated flexural bending
AST04 Asphalt binder content (Ignition oven
method)
AST05 Sample preparation
- compaction of asphalt
slabs suitable for characterisation
A5T06 Asphalt binder drain-off
AST07 Asphalt particle loss

61
Appendix A4.1
Combined Aggregate Particle Size
Distribution (Rothfuchs Method)
This is a graphical method for proportioning rwo or
more aggregates to obtain a required particle size dis-
tribution.
The steps are as follows:
1. Determine the required (target) particle size distri-
bution.
2. Plot the cumulative percentage passing values of the
required particle size distribution.
Use a linear ordinate scale for the percentage pass-
ing, and a varying scale of sieve size such that the
particle size distribution plots as a straight line. This
is readily done by drawing an inclined straight line
of convenient length and marking off the sieve size
along the base at points vertically below the inter-
section of the straight line and the specified percent-
age passing the ordinate corresponding to the sieve
sizes (in the specification).
3. The particle size distributions of the component
materials are plotted on the chart to the same
scales. In general, it will be found that they do not
plot as straight lines.
4. For each of the component materials, draw a
straight line such that the areas between the distri-
bution curve and the straight line are equal on
either side of the line.
The line for the coarsest material should pass
through 100% passing point for the maximum
aggregate size, and for the finest material, the line
should pass through the origin of the chart.
5. Join the 0o/" point of the line for the coarsest mater-
ial to the 100% point of the second coarsest mater-
ial and so on.
The percentages for mixing can be read from the
points where these joining lines cross the straight
Iine of the particle size distribution of the required
mix.
(Reference: NAASRA, "Guide to Stabilization in
Roadworks", Sydne5 1986.)

-I
I

o
o)
c
a
I Eg
ro I 69
16
I
o
60 B3F
c) I

o)
(d I

6+o
o *-e 6
q)
rol
I oo) CO
(I
coluo
NI P

EL:
0.075 0.300 1.18 2.36 4.75 6.70 9.50 13.2 19.0 26.5
0.150 0.600
AS Sieve Sizes (mm)
(Scale proportionate to percentage passing sieve)

Figure A4.1 Rothfuchs method for combining aggregates

62
 & Appendix A4.2 Adjustment of Mix Designs
i
:i

iF.
tit
it1
1,

Possible remedies Possible remedies

. Reduce percent filler, or bitumen, or both . lncrease percent filler, and percent of bitumen
. lncrease percent of coarse aggregate

Possible remedies Possible remedies

. Reduce percent fillel or bitumen, or both. . lncrease percent filler
o Porous aggregates, such as some slag or limestones, . Change proportions of coarse and fine aggregates
require maximum specified percent of bitumen. to produce a lower VMA
. Change proportions of coarse and fine aggregates . lncrease proportion of crushed materials in mix
to produce a lower VMA.

Possible remedies:
. lf percentage of bitumen is near high limit, try increasing percentage of coarse aggregate and reducing
percentage of bitumen
. lf the percentage of bitumen is near low limit, the mineral aggregate may be deficient. Aggregate particles with
rougher textures and less rounded surfaces will provide greater stability while maintaining or increasing the void
content. lf the coarse aggregate is gravel, this may be the cause of low stability.
. Before rejecting either fine or coarse aggregates, trial mixes should be prepared and tested, using both the
maximum and minimum percentages of c.oarse aggregate permitted by the specification.

63
 Table A4.3: lnteraction of Variables Showing Effect on the Voids Filled with
Binder and Any SecondarY Effects

Aggregate absorption Binder % Air voids decreased

Aggregate absorption Air voids % Binder % (Would have

to be increased to
compensate for particle
size distribution change)

Aggregate absorption Decreased Air voids increased

(Loss of effective

Air voids % Aggregate absorption Binder % lncreased VMA unaltered (Would

(change in aggregate need to be increased
quality and type) to compensate)
Aggregate absorption VMA (Due to change in Decreased Binder % (Would need
aggregate particle size to be increased to
distribution) compensate)

Binder % Aggregate absorption Decreased VMA unaltered (Need

(Loss of effective to increase to
compensate)

Aggregate Air voids % Binder % lncreased VMA unaltered (Need

absorption to increase to
compensate)

Binder % Voids % Decreased VMA would increase

Binder % VMA Decreased Air voids increased

Binder Content Aggregate absorption VMA (Change in Decreased Air voids increased
(%) aggregate particle
size distribution)

Aggregate absorption Air voids % (Change Decreased VMA increased (Result

in aggregate particle of change in aggregate
size distribution) particle size distribution)

Air voids % Aggregate absorPtion Decreased VMA unaltered

(Loss of effective (Provided aggregate
particle size distribution
unaltered)

64