EVALUATION ON PROPERTIES OF TENDER MIXES
ZANARIAH BT ABD RAHMAN
A project report submitted in partial fulfillment of the
requirements for the award of the degree of Master of Engineering
(Civil – Transportation and Highway)
Faculty of Civil Engineering
Universiti Teknologi Malaysia
NOVEMBER 2006
iii
To My Parents,
Abd Rahman and Zaharah
To My Brother and Sister,
Malek Faizal and Noor Liyana
To My Fiancee,
Mohd Zaki bin Hassan
iv
ACKNOWLEDGEMENT
Alhamdulillah, with His blessing, I have successfully completed my master’s
project and I am thankful for all the people around me who has contributed to the
completion of my project.
First of all, I would like to convey my appreciation to two of my project
supervisor, PM Dr Abdul Aziz bin Chik and Dr. Mohd Rosli bin Hainin for all the
ideas, encouragement and guidance throughout the research.
My sincere appreciation also extends to En. Suhaimi, staffs of Highway and
Transportation Laboratory for the time he spends in helping and guiding me with all
the laboratory works.
Not to forget, my fellow friends Elizabeth and Norliza, thanks for the best
moments we share through hard works at the laboratory. Last but not least, I would
like to express my deepest appreciation to the loves ones, my parents, my brother and
my sister for always supporting me and to my dearest fiancée who never give up on
me. I love you all.
v
ABSTRACT
Tender mix has caused many problems to the contractor during the
construction of hot mix asphalt (HMA) pavement. The objective of this paper was to
investigate the properties of tender mixes as related to the problem of rutting. Two
mixes of ACW20 were designed in compliance to Jabatan Kerja Raya (JKR)
specification. One mix was designed with typical dense graded gradation but away
from the maximum density line (MDL) described as control mix. The other mix was
designed close to MDL to simulate tender mix. Marshall sample were prepared in
order to determine the optimum bitumen content (OBC) and volumetric properties of
compacted mixtures. Using the OBC obtained from Marshall samples, two beams
were fabricated for each mix for the wheel-tracking test. Comparisons of rut depth
between control mix and tender mix were made at 500, 1000, 2000 and 5000 passes.
Volumetric properties results indicate that ‘tender mix’ is not tender as expected due
to high voids in the mineral aggregate (VMA) compared to control mix. However,
there is a significant difference between tender mix and control mix in terms of
rutting according to the t-Test statistical analysis. Furthermore, tender mix indicated
low stability and stiffness value which show that the gradation of tender mix that was
designed close to MDL are recommended as poor gradation and show a potential
problem in mixes if the mix is used.
vi
ABSTRAK
Campuran lembut telah menimbulkan banyak masalah kepada kontraktor
jalan raya semasa proses turapan campuran berasfalt panas (HMA). Objektif bagi
kajian ini ialah untuk menilai ciri-ciri volumetrik yang pada campuran lembut dan
dikaitkan dengan masalah aluran. Dua campuran ACW20 telah direka dengan
mematuhi keperluan spesifikasi dari Jabatan Kerja Raya (JKR). Satu rekaan
campuran mempunyai gradasi gred tumpat yang tipikal tetapi menjauhi garisan
ketumpatan maksimum (MDL) dan dikenali sebagai campuran kawalan manakala
satu rekaan campuran yang lain mempunyai gradasi yang direka hampir dengan
MDL dan dikenali sebagai campuran lembut. Sampel Marshall disediakan untuk
mendapatkan kandungan bitumin optimum (OBC) dan ciri-ciri volumetrik bagi
setiap campuran. Dengan menggunakan kandungan bitumen optimum yang telah
diperolehi, dua sampel rasuk disediakan untuk campuran kawalan dan campuran
lembut sebagai sampel untuk digunakan dalam ujian jejak roda. Perbandingan bagi
kedalaman aluran antara dua campuran tersebut akan dilakukan pada 500, 1000,
2000 dan 5000 laluan. Daripada keputusan ciri-ciri volumetrik, didapati bahawa
lompang dalam agregat (VMA) bagi campuran lembut menunjukkan nilai yang tidak
dijangka iaitu nilai VMA campuran lembut lebih tinggi berbanding nilai VMA
campuran kawalan. Walaubagaimanapun, terdapat perbezaan yang ketara dalam
nilai kedalaman aluran antara campuran kawalan dan campuran lembut berdasarkan
daripada analisis statistik t-Test. Tambahan pula, campuran lembut juga
menunjukkan nilai kestabilan dan kekukuhan yang rendah dan dengan ini gradasi
bagi campuran lembut yang direka berhampiran dengan MDL dicadangkan sebagai
gradasi yang tidak sesuai digunakan kerana berpotensi untuk menimbulkan masalah
jika campuran digunakan kelak.
vii
TABLE OF CONTENTS
CHAPTER
1
2
TITLE
PAGE
DECLARATION
ii
DEDICATION
iii
ACKNOWLEDGEMENT
iv
ABSTRACT
v
ABSTRAK
vi
TABLE OF CONTENTS
vii
LIST OF TABLES
ix
LIST OF FIGURES
x
LIST OF SYMBOLS / ABBREVIATIONS
xii
LIST OF APPENDICES
xiii
INTRODUCTION
1.1
Research Background
1
1.2
Problem Statement
3
1.3
Objective
3
1.4
Scope
4
1.5
Importance of Study
4
LITERATURE REVIEW
2.1
Introduction
5
2.2
Aggregate
7
2.3
Aggregate Gradation
7
2.3.1
10
Gradation Limit of ACW20
vii
2.4
Maximum Density Line
10
2.5
Voids in the Mineral Aggregate (VMA)
13
2.5.1
15
2.6
3
Varibility in VMA
Tender Mixes
16
2.6.1 Identify Tender Mixes
17
2.6.2 Causes of Tender Mixes
17
2.6.3
Incorrect Mix Design
18
2.6.4
Smooth and Rounded Aggregates
19
2.6.5
Moisture in the Mix
20
2.7
Compaction of Tender Mixes
20
2.8
Rutting
22
2.9
Wheel Tracking Machine
24
2.9.1 Wheel Tracking Apparatus
24
RESEARCH METHODOLOGY
3.1
Introduction
26
3.2.
Gradation Design
27
3.3
Laboratory Test Procedure
29
3.3.1
Sieve Analysis of Fine and Coarse
Aggregates (ASTM C 136-84a)
3.3.2
Specific Gravity and Absorption
of Coarse Aggregate (ASTM C 127-88)
3.3.3
32
Specific Gravity and Absorption
of Fine Aggregate (ASTM C 128-88)
3.3.4
30
34
Theoretical Maximum Specific Gravity
and Density of Bituminous Paving
Mixtures (ASTM D 2041-91)
3.3.5
36
Resistance to Plastic Flow of
Bituminous Mixtures Using Marshall
Apparatus (ASTM D 1559)
3.4.
3.5
38
Mixing Specimen
40
3.4.1
41
Sample Compaction
Wheel Tracking Machine Test
42
viii
3.6.
Data Analysis
3.6.1
4
5
43
Volumetric Properties of
Compacted Mixtures
43
3.6.2
Optimum Bitumen Content
45
3.6.3
Wheel Tracking Test Result
46
3.6.4
Standard Specification
46
RESEARCH FINDINGS AND ANALYSIS
4.1
Introduction
48
4.2
Aggregate Gradation
49
4.3
Result of Volumetric Properties
52
4.4
Result of Wheel Tracking Test
53
CONCLUSIONS AND RECOMMENDATION
5.1
Introduction
56
5.2
Summary of the Findings
56
5.3
Recommendations
57
REFERENCES
Appendices A - F
59
62 - 71
ix
LIST OF TABLES
TABLE NO.
TITLE
PAGE
2.1
Gradation Limit for ACW20
10
2.2
Recommended Minimum VMA Values
14
2.3
Factor that Affect the VMA of HMA
15
3.1
Gradation Design of ACW20 Control Mix
28
3.2
Gradation Design of ACW20 Tender Mix
28
3.3
Gradation Limits for Asphaltic Concrete
47
3.4
Design Bitumen Contents
47
3.5
Test and Analysis Parameters for Asphaltic Concrete
47
4.1
Gradation of ACW20 Control Mix for Marshall Sample
49
4.2
Gradation of ACW20 Tender Mix for Marshall Sample
50
4.3
Gradation of ACW20 Control Mix for
Wheel Tracking Sample
4.4
Gradation of ACW20 Tender Mix for
Wheel Tracking Sample
4.5
51
51
Volumetric Properties of ACW20 Control Mix
and Tender Mix at OBC
52
4.6
Number of Compaction of Wheel Tracking Sample
53
4.7
Summary of Data from the
Wheel Tracking Machine Test
54
x
LIST OF FIGURES
FIGURE NO.
TITLE
PAGE
2.1
Processed Aggregates at the Quarry
7
2.2
Typical Terms Used to Identify Aggregate Gradation
8
2.3
0.45 Power Gradation Chart
11
2.4
Group of MDL Plotted on 0.45 Power Gradation Chart
11
2.5
Maximum Density Line Related to VMA
12
2.6
Illustration of VMA
13
2.7
Gradation Pattern of Tender Mix on
0.45 Power Gradation Chart
2.8
19
Stress Applied to the Sub-grade or
Base below the Asphalt Layer
23
2.9
Three Wheel Immersion Tracking Machine
25
3.1
Flow Chart of the Laboratory Works
26
3.2
Gradation Chart of Control Mix and
Tender Mix for ACW20
29
3.3
Sieves Agitated by Mechanical Apparatus
31
3.4
Aggregate Separated and Stored in
Container According to Sizes
3.5
31
The ASTM D 2041 Test Apparatus
37
o
3.6
Specimen Immerse in Water Bath at 60 C
39
3.7
Compression Testing Machine
39
3.8
Sample Place and Check for
Compaction Temperature
3.9
40
Sample after Compaction with
9 Kilograms of Steel Roller
41
xi
3.10
Well Compacted Sample
42
3.11
Maintain Water Temperature in Wheel Tracking Machine
43
3.12
Reading at Three Point for Each Sample
43
4.1
Rut Depth vs. Number of Roller Passes of
Control Mix and Tender Mix
4.2
50
Rut Depth vs. Number of Roller Passes
of Control Mix and Tender Mix
55
xii
LIST OF SYMBOLS / ABBREVIATIONS
AASHTO
-
American Association of State Highway and
Transportation Officials
ACW
-
Asphaltic Concrete Wearing Course
AI
-
Asphalt Institute
ASTM
-
American Society for Testing and Materials
FHWA
-
Federal Highway Administration
Gmb
-
Bulk Specific Gravity of Compacted Mixture
Gmm
-
Maximum Specific Gravity of Paving Mixture
Gsa
-
Apparent Specific Gravity of Aggregate
Gsb
-
Bulk Specific Gravity of Aggregate
Gse
-
Effective Specific Gravity of Aggregate
HMA
-
Hot Mix Asphalt
JKR
-
Jabatan Kerja Raya
MDL
-
Maximum Density Line
NAPA
-
National Asphalt Pavement Association
OBC
-
Optimum Bitumen Content
OPC
-
Ordinary Portland Cement
Pmm
-
total loose mixture, percent by total weight of mixture
Ps
-
percent of aggregate by total mass of mixture
Superpave
-
Superior Performing of Asphalt Pavement
UK
-
United Kingdom
US
-
United States of America
UTM
-
Universiti Teknologi Malaysia
VFA
-
Voids Filled with Asphalt
VMA
-
Voids in the Mineral Aggregate
VTM
-
Voids in Total Mix
xiii
LIST OF APPENDICES
APPENDIX
TITLE
PAGE
A
Aggregate Bulk Specific Gravity of ACW20
62
B
Maximum Specific Gravity of Loose Mixture ACW20
63
C
Calculations of Mineral Filler from Washed Sieve
Analysis Result
D
i)
Calculations of VTM in ACW20
Control Mix and Tender Mix
ii)
F
65
Calculations of VTM Before and After
Wheel Tracking Test
E
64
66
Result of Wheel Tracking Test for ACW20
Control Mix and Tender Mix
67
Photograph
69
CHAPTER 1
INTRODUCTION
1.1
Research Background
There are two types of tenderness as reported by Crawford (1989). The first
type is characterized by the asphalt mix being difficult to compact when normal
construction techniques are used. Re-compaction attempts will result in a decrease in
pavement density. The other type of tenderness is characterized by the asphalt
mixtures being slow setting after construction. This type is sensitive to turning
traffic and power steering. It may also lack resistance to critical loading, especially
during hot weather.
The problem of compaction of tender mixes is actually has been observed for
years by United States. Tender mixtures are not stable under the roller and tend to
move laterally when rolled. This lateral movement sometimes result in hairline
crack. Hairline cracks that sometimes results when rolling tender mixes are usually
very shallow and do not cause a significant problem. However, these cracks allow
the mix to absorb moisture and may reduce the durability of the hot mix asphalt
(HMA). They may provide a weakness in the HMA pavement that may result in
crack growth and eventually premature failure. In the past year, most tender mixes
were attributed to excessive temperatures or excessive sanded mixes. There are
many other possible reasons for the tender mixes but these two causes appeared to be
mentioned most (Brown et al., 2000).
2
The complaints about tender or slow setting asphalt pavements in the United
States always arise at about the same time of year which is from about the first part
of July through the middle of September (Tarrer and Wagh, 1991). At this time of
year, ambient temperatures are high. Tender pavement rarely occurred in cool
weather therefore it seems obvious that one of the conditions that must be obtained
for this type of distress is hot weather. Furthermore, Hot Mix Asphalt Paving
Handbook (2000) shows that gradation that close to the maximum density line
(MDL) may have at times lower than desirable Voids in the Mineral Aggregates
(VMA) which will result in very little void space within to developed sufficient
asphalt thickness for durable mix. It is also recommended that such gradation to be
avoided so as not to produce mixes that are tender and difficult to compact
Brown et al. (2000) reported that in the early to mid 1990s, Superpave mixes
began to be used in the United States. For the most part, these mixes have been
coarse-graded mixes with relatively high coarse aggregate content. Experience has
shown that when these mixes are tender, they act similar to tender mixes that were
encountered in the past. Based on two surveys by National Asphalt Pavement
Association (NAPA), it appears that approximately 40 percent of coarse graded
Superpave mixes experience some tenderness (Brown et al., 2000). Therefore, as a
result of reported tenderness problems, the Federal Highway Administration
(FHWA) and NAPA held a jointly meeting in June 1998. There was a lot of
discussion about causes and cures of the tender mix problem among the attendees
which included state Department of Transportations (DOTs) and Industry
representatives (Brown et al., 2000). This shows that FHWA and NAPA are concern
about the problems created from tender mixes and is looking forward to improve the
mixes.
3
1.2
Problem Statement
Tarrer and Wagh (1991) reported that tender mixes are often difficult to
compact to the required density. Once the mix begins to move laterally, additionally
rolling results in further lateral movement and does not allow for adequate
compaction. Even though these tender mixes may not result in loss of life, they will
lower the overall pavement quality by increasing the roughness of the compacted
mixes. In general, tender mixes are difficult to roll, difficult to achieved specified
density and occasionally rut. Other than that, they will also displace under high
pressure and shove and scuff under traffic (Button et al., 1980).
A remarkable increase in traffic volume has contributed to the severe rutting
on highway and main road in Malaysia. Rutting is defined as the accumulation of
small amounts of unrecoverable strain resulting from applied wheel loads to HMA
pavement (Cooley Jr et al., 2000). Rutting in HMA will not only decrease the life of
pavement but also will create safety hazard to the public. Therefore, it is necessary
to estimate the potential of rutting on tender mixes besides investigating the
properties of tender mixes.
1.3
Objective
This study is undertaken to evaluate the properties of tender mixes as related
to rutting problem.
4
1.4
Scope
In order to accomplish the objective, this study is subjected to this following
scope and limitation:
i.
Designing two (2) ACW20 mixes using Marshall design conforming
to Jabatan Kerja Raya (JKR) specification;
a.
One mix design with typical dense graded gradation but away
from maximum density line (MDL) described as control mix.
b.
One mix design with gradation design close to MDL to
simulate tender mix.
ii.
Wheel tracking machine was used to investigate the differences in rut
depth between control mix and tender mix.
1.5
Importance of Study
From this project, the properties of tender mix that are design close to MDL
can be determined hence providing a guideline for highway engineers to produce a
high-quality pavement through well designed gradation. In relation to the properties,
the suitability of the gradation to resist rutting was also be able to determined
through analyzing the data and result from wheel tracking machine test.
CHAPTER 2
LITERATURE REVIEW
2.1
Introduction
Asphalt is of particular interest to the engineer because it is a strong cement,
readily adhesive, highly waterproof and durable. It is, moreover, highly resistant to
the action of most acids, alkalies and salts (Asphalt Institute, 1989). Asphaltic
materials are obtained from seeps or pools of natural deposits in different parts of the
world or as a product of the distillation of crude oil according to Garber and Hoel,
(2002). This shows that asphalt is mostly used for various paving purposes because
paving asphalt is waterproof and is unaffected by most acids, alkalies and salts. This
unique combination of characteristics and properties is a fundamental reason why
asphalt is an important paving material.
Hot mix asphalt (HMA) nowadays are widely use in highway constructions.
HMA design mostly involves selecting and proportioning aggregates and asphalt
binder to obtain specific construction and pavement performance properties. The
goal of the design is to find an economical blend and gradation of aggregates and
asphalt binder that give a mixture that has (Asphalt Institute, 1989):
i.
Sufficient asphalt binder to ensure a durable compacted pavement by
thoroughly coating and bonding the aggregate,
ii.
Enough workability to permit mixture placement and compaction
without aggregate segregation,
6
iii.
Enough mixture stability to withstand the repeated loading of traffic
without distortion or displacement, and
iv.
Sufficient air voids in the total compacted mix to prevent asphalt
binder from flushing, bleeding or a loss of mixture stability, yet low
enough to keep out harmful air and moisture.
Martin and Wallace, (1958) also recommended steps in the procedure for a
rational design are as follows:
i.
Select grading to be used,
ii.
Select aggregates to be employed in the mix,
iii.
Determine the specific gravity of the aggregate combination and of
the asphalt,
iv.
Determine the proportion of each aggregate required to produce the
designed grading,
v.
Make up trial specimens with varying asphalt contents,
vi.
Determine the specific gravity of each compacted specimens,
vii.
Make the stability tests on the specimens,
viii.
Calculate the percentage of voids in each paving specimen and, if the
design method in use requires it, calculate the voids in the mineral
aggregates (VMA) and the percent voids filled with asphalt, and
ix.
Select the optimum asphalt content from the data obtained.
Most commonly, HMA is divided into three different types of mix which are
dense-graded, open-graded and gap-graded. However, poor workmanship will result
in pavement deformation which eventually will lead to tender mixes. Tenderness
may appear while working with HMA and created problems during construction
especially compaction.
7
2.2
Aggregate
Aggregate is a combined term for mineral such as sand, gravel and crushed
stone. Aggregates can be natural or manufactured. Natural aggregates are generally
extracted from larger rock formations through an open excavation (quarry). Usually
the rock is blasted or dug from the quarry walls then reduced to usable sizes using a
series of screens and crushers. In Malaysia, processed aggregates from quarry is
widely use in construction of premix roads (Figure 2.1). A manufactured aggregate
is often the by-product from other manufacturing industries for example steel slags.
Figure 2.1: Processed Aggregates at the Quarry
2.3
Aggregate Gradation
About 85 percent of HMA by volume consist of mineral aggregate and one of
the most important properties of the aggregate in a HMA mix is the gradation
(Mallick et al., 1998). Gradation perhaps is the most important properties of HMA,
including stiffness, stability, durability, workability, fatigue resistance and resistance
to moisture damage. The mixtures volumetric properties including asphalt content,
VMA and VFA have been identified as important parameters for performance.
However, Prowell, Zhang and Brown (2005) indicated that VMA is considered the
8
most important parameter and is used in Superpave mixture design specifications to
eliminate the used of potentially poor-performing mixtures. Aggregate gradation is
the distribution of particles sizes expressed as a percent of the total weight.
Gradation is determined by sieve analysis by passing the material through a series of
sieves stacked with progressively smaller openings from top to bottom and weight
the material retained on each sieve. Gradation of an aggregate can be graphically
represented by a gradation curve. Figure 2.2 shows typical terms used to identify
aggregate gradation.
Figure 2.2: Typical Terms Used to Identify Aggregate Gradation
In order to produce a good quality of HMA, aggregate gradation must be
according to the specification. On the other hand, with the right and appropriate
gradation, stability and stiffness of the pavement will increase and produce a strong
premix surface to sustain loads. Gradation that produces maximum density is
believed to be the best gradation. It is involve a good particle arrangement where
smaller particles are packed together between larger particles which will eventually
reduce the void space between particles. This creates more particles to particles
contact which will increase the stability of HMA and reduce water infiltration.
Theoretically, it would seem that the best gradation for HMA is one that gives the
densest particle packing. The gradations having maximum density provides
9
increased stability through increased particle contacts and reduce void space to
permit enough asphalt to ensure durability, while leaving some air space in the
mixture to avoid bleeding or rutting. A tightly packed aggregate will have low void
in the mineral aggregates (VMA) also result in a mixture that is more sensitive to
slight changes in asphalt content (Roberts et al., 1996).
Roberts et al. (1996) also reported that Fuller and Thompson have proposed
Fuller’s curve and from their studies, Fuller and Thompson showed that a maximum
density can be obtained for an aggregate when n = 0.5. Equation 2.1 show Fuller’s
maximum density curve.
⎡d ⎤
P = 100 ⎢ ⎥
⎣D⎦
n
Equation 2.1
where,
P = % finer than the sieve
d = aggregate size being considered
D= maximum aggregate size to be used
n = parameter which adjust curve for fineness or coarseness
10
2.3.1
Gradation Limit of ACW20
Asphaltic Concrete Wearing Coarse 20 (ACW20) is the mix that is going to
be used for this study. Gradation should conform to the Standard Specification for
Road Work, JKR/SPJ/1988. Table 2.1 shows the gradation limit for Asphaltic
Concrete Clause 4.2.4.2 from Table 4.8 of JKR/SPJ/1988.
Table 2.1: Gradation Limit for ACW20 (JKR, 1988)
2.4
Mix Type
Wearing Coarse
Mix Designation
ACW20
B. S. Sieve
% Passing by Weight
28.0 mm
100
20.0 mm
76-100
14.0 mm
64-89
10.0 mm
56-81
5.0 mm
46-71
3.35 mm
32-58
1.18 mm
20-42
425 um
12-28
150 um
6-16
75 um
4-8
Maximum Density Line
Grading curves can be helpful in making necessary adjustment in mix
designs. Federal Highway Administration (FHWA) has introduced an aggregate
grading chart which is based on the Fuller gradation in early 1960s (Asphalt Institute,
1983). However, the grading chart uses a 0.45 exponent in the equation. This chart
is very convenient to determine the maximum density line (MDL) and to adjust
aggregate grading. From this chart, MDL can be obtained by drawing a straight line
from the origin at the lower left of the chart to the maximum aggregate size at the
11
upper right of the chart as shown in Figure 2.3. Furthermore, Figure 2.4 shows group
of MDL plotted on the 0.45 power gradation chart. FHWA recommends this chart to
be used as part of the HMA design process.
Figure 2.3: 0.45 Power Gradation Chart
Figure 2.4: Group of MDL Plotted on 0.45 Power Gradation Chart
Chadbourn et al., (2000) reported that Goode and Lufsey demonstrated an
aggregate having a gradation that produces a straight line on a 0.45 power gradation
chart will have the maximum achievable density, the lowest air voids content and the
12
lowest VMA in a HMA mixtures. Figure 2.5 illustrates a 0.45 power plot of an
aggregate gradation developed by FHWA and reported by Goode and Lufsey. It is
used to estimate how densely a given aggregate mixture will compact. A line drawn
from the origin of this plot through the nominal maximum aggregate size is estimated
as the maximum density line for any given aggregate. Increasing the sum of the
distances between a gradation and the MDL will tend to increase the VMA of the
compacted mixture.
Figure 2.5: Maximum Density Line Related to VMA
Gradation of maximum density may not provide sufficient voids in the
aggregate for enough asphalt to provide adequate film thickness for maximum
durability without bleeding. In such cases, deviations from the MDL are necessary
in order to increase the VMA. Minimum requirement for VMA is also necessary to
ensure that there are sufficient void in the aggregate to allow asphalt to be added to
maintain stability. From a construction standpoint, the introduction of using large
stone mixes is to minimize rutting potential of HMA. These large stone mixtures are
more resistant to rutting than the smaller aggregates size mixtures. However, the use
of a maximum aggregate size greater than 25.4 mm (1 inch) often results in harsh
mixes that tend to segregate during placement (Robert et al., 1996). Many problems
could occur caused by poor aggregate gradation and one of them is tender mixes.
13
2.5
Voids in the Mineral Aggregate (VMA)
According to Kandhal and Chakraborty, (1996), voids in the mineral
aggregate (VMA) is the sum of the air voids and the effective binder volume in the
mixture. Establishing an adequate VMA during mix design and in the field will help
establish adequate film thickness without excessive asphalt bleeding, flushing or
rutting. Roberts et al., (1996) stated that VMA describes the portion of space in a
compacted asphalt pavement or specimen which not occupied by the aggregate and
expressed as a percentage of the total volume of the mix. When aggregate particles
are coated with asphalt binder, a portion of the asphalt binder is absorbed into the
aggregate, whereas the remainder of the asphalt binder forms a film on the outside of
the individual aggregate particles. Since the aggregate particles do not consolidate to
form a solid mass, air pockets also appear within the asphalt-aggregate mixture.
Therefore, as Figure 2.6 illustrates, the four general components of HMA are:
aggregate, absorbed asphalt, asphalt not absorbed into the aggregate (effective
asphalt), and air. Air and effective asphalt, when combined, are defined as VMA
(Chadbourn et al., 2000).
Figure 2.6: Illustration of VMA
14
VMA is calculated according to the following relationship:
VMA = 100 −
Ps × Gmb
G sb
where,
Ps = Aggregate content, percent by total mass of mixture
Gsb = Bulk specific gravity of total aggregate
Gmb = Bulk specific gravity of compacted mixture
The importance of designing VMA into an HMA mix has been recognized
for many years. For many years, Asphalt Institute mix design procedures have used
minimum VMA criteria that are dependent upon maximum aggregate size. If the
VMA is too low, it can be increased by modifying the gradation, asphalt content, or
particle angularity. Table 2.2 shows typical minimum VMA values recommended by
the Asphalt Institute (Asphalt Institute, 1989).
Table 2.2: Recommended Minimum VMA Values
* Taken directly from the Asphalt Institute’s MS-4 Manual.
15
2.5.1 Variability in VMA
Previous topic has discussed about the importance of adequate VMA in an
asphalt mixture. To analyze the contribution of VMA to pavement durability, it is
important to understand the parameters of HMA that relate to the determination of
VMA. Certain characteristics of HMA mixture and its components can change the
VMA and film thickness and one of the characteristics is aggregate gradation.
Chadbourn et al., (2000) summarized the characteristics in affecting the VMA of
HMA in Table 2.3.
Table 2.3: Factors that Affect the VMA of HMA
16
2.6
Tender Mixes
Tender mixes, also known as tender pavements, comprise a major problem by
the asphalt industry in United States (US) (Marker, 1977). The occurrence of tender
pavements or slow-setting asphalt concrete mixes is not new. The behavior of tender
mixes has been described in various ways and in different circumstances. The
definition of tender pavement is defined as following (Marker, 1977):
i.
Has very low resistance to deformation by “punching” loads.
ii.
Scuffs under horizontally-applied shearing loads after compaction has
been completed.
“Punching” loads are an application of a high unit load to a very small area.
Horizontally-applied shearing loads are exemplified by those imposed by front-wheel
steering turns of stationary vehicles (Marker, 1977).
Tender pavement has been described in many ways and according to Marker
(1977), the following difficulties has been associated with tender pavement:
i.
The mix is difficult to roll.
ii.
The specified density is difficult to achieve.
iii.
The pavement ruts after construction is complete.
iv.
The pavement is soft after completion and will displace under the heel
of a shoe.
v.
The pavement “shoves” under traffic, sometimes months after
construction.
vi.
The pavement “slips” under traffic, usually fairly soon after
construction.
vii.
The pavement “scuffs” under power steering or severe braking action.
viii.
The pavement indents under a punching load.
17
2.6.1
Identify Tender Mixes
It is an advantage to identify tender mixes prior to the start of construction so
materials and design parameters may be altered. Mixtures which contain one or
more of the following characteristics which are a large portions of sand sizes, smooth
and rounded aggregates, asphalt that are highly temperature susceptible, slow setting
asphalt and high fluids content should be suspected to be tender. Button et al (1980)
favours the idea that tenderness during construction is mainly an aggregate problem
caused by using smooth, rounded aggregate and high sand and low filler percentages,
which may be aggravated by highly temperature-susceptible asphalt cement.
Selection of the proper mixing temperature for the asphalt cement may help to avoid
potential tenderness. Crawford (1989) believes that tenderness after construction
seems to relate to the slow-setting characteristics of asphalt cement, which may also
show up with a critical aggregate gradation. Two possible approaches in recognizing
tender mixes prior to placement. The first approach uses the collective field
experience of engineers to identify those materials mixtures and construction factors
which contribute to tender mixtures and the second approach is using the laboratory
tests and associated criteria for identification of mixtures that are likely to be tender
during placement (Tarrer and Wagh, 1991).
2.6.2
Causes of Tender Mixes
Identifying the specific causes of tender mixes is difficult to do. There are a
number of items that can cause mixes to be tender and any combination of these
items may result in tenderness. Works by Crawford (1989) identified that common
causes of tender mixes are any one or any combination of any of the following:
i.
Incorrect mix design,
ii.
Smooth and rounded aggregates,
iii.
Moisture in the mix,
iv.
Abnormally high ambient temperature,
18
v.
Asphalt cements characteristics,
vi.
Incorrect asphalt cement grade,
vii.
Incorrect production and construction techniques,
viii.
Inadequate bond to underlying layer, and
ix.
2.6.3
Stiffness of binder.
Incorrect Mix Design
The most important factor that can be identified with tender mix pavement is
the aggregate gradation. Tender mixes caused by poor aggregate gradation are slow
in developing sufficient stability to withstand the compaction load. If the filler is too
low, the mix might act tender due to inadequate binder stiffness since some of the
filler is needed to provide adequate binder stiffness. Generally, mixes with more
uniform aggregates sizes are more likely to be tender than a more well-graded
aggregate (Brown et al, 2000). Aggregate gradation specifications for HMA have
been developed through accumulated field experiences and in many cases, they are
established by trial and error to suite the field condition (Roberts et al., 1996).
Figure 2.7 below shows a typical gradation pattern for tender mixes (Roberts et al.,
1996). A hump is noted above the maximum density line in the curve near the
number 40 sieve (0.42 mm) and the flat slope between the number 40 (0.42 mm) and
the number 8 (2.38 mm) sieves. The most likely cause of an aggregate blend with
this shape is having an excessive use of poorly-graded of natural sands (Crawford,
1989).
19
Figure 2.7: Gradation Pattern of Tender Mix on 0.45 Power Gradation Chart
Goode and Lufsey (1962) showed that modifications to the gradation of a
tender mix to avoid such humps could produce a less critical mix with better
compaction characteristics. In aggregate blending calculations, the number 8 (2.38
mm) sieve and the number 200 (0.074 mm) sieve are often considered as key areas in
controlling the mix gradation for mix design purposes.
2.6.4 Smooth and Rounded Aggregates
Smooth and rounded aggregates have long been associated with tender mixes.
Rounded particles that may be found in sands and gravels often tend to cause the mix
to act tender during compaction. It is much easier for the rounded aggregates than
for angular aggregates to roll past adjacent aggregates resulting in lateral movement
during the compaction process (Brown et al, 2000). Rounded and polished are more
likely to produce tender mixes than angular aggregates because angular aggregate
have rough surfaces while rounded and polished aggregates lack friction and
resistance so the particles are easily slide each other under traffic loading. Friction
between rough-textured aggregate can provide resistance to deformation and at the
same time can prevent or reduce tender mix occurrence.
20
2.6.5
Moisture in the Mix
Moisture can occur in several places that may result in the tenderness of the
mix. Excess moisture can be present in the mix when the aggregate is not properly
dried. Other than that, moisture can also exist on the existing pavement surfaces that
can cause the mix to act tender when the surface is overlaid. Moisture will increase
the liquid content of the mix and thus decreases the internal strength of the mix
during the lay down and compaction stage causing loss of cohesion by reducing the
overall viscosity of the asphalt. Hot weather promotes the conversion of internal
pore moisture in the aggregate to water vapour and softening the mix (Crawford,
1989). Moisture at this temperature will convert to steam, which greatly increases
the volume of the moisture. The steam exerts internal pressure on the mix that tends
to push the aggregates apart as the mix is being rolled. This forces result in a
decrease in internal strength when rolled causing the mix acting tender (Brown et al,
2000). As the mix cools, the moisture factor will become less critical.
2.7
Compaction of Tender Mixes
A tender mix is generally an internally unstable mix that will not properly
support the weight of the compaction equipment when hot and will move under the
applied compactive effort. The movement of the mix can take various forms (HMA
Paving Handbook, 2000):
i.
First, a bow wave may occur in front of the steel wheel on both a
vibratory and static steel wheel roller as these rollers move
longitudinally up and down the material.
ii.
Second, the material may widen out when the rollers are used to
compact the unsupported edge of lane.
iii.
Third, checking-short, transverse cracks that develop during the
compaction process may occur in the mix.
21
iv.
Fourth, longitudinal humping up and checking of the mix may occur
immediately outside of the edges of the steel wheel on the rollers.
Tenderness usually comes in one of two forms. Classical tenderness occurs
when the breakdown roller is unable to approach pavers without the mixture begins
to move (HMA Paving Handbook, 2000). When this situation occurs, the roller will
not approach the back of the pavers. They leave some distance behind the pavers to
allow the mixture to cool sufficiently. So, the main approach to handle classical
tenderness is to allow the mixture to cool. However, the best method is to re-design
because if the mixture is to be used on a high traffic route then this classical
tenderness is a sign of non-resistance to rutting.
When mid-temperature range tenderness is encountered, the roller can
approach right up to the pavers. The characteristics of mid-temperature range
tenderness normally show up under breakdown rolling if the temperature of the mix
at that point is above approximately 115oC. The mix is generally stable at higher
temperatures and when the temperature of the material drops below this level
however, the mix become unstable and tender. Here, the movement of the mix can
be seen where a bow wave may occur in front of the steel wheel rollers, checking
may occur in the surface of the material and the mix may hump up outside the edges
of the steel drums on the rollers (HMA Paving Handbook, 2000).
The mix may continue to exhibit these tenderness characteristics as the
temperature of the materials decreases to approximately 90oC or lower. If rolling is
attempted at the middle temperature range, the mix will de-compact instead of
compacting. It is not until the mix is quite cool which is less than 90oC that it
becomes stiff enough to support weight of the compaction equipment. Rolling will
often finish at temperature of 70oC or less. On the other hand, noted that the
temperature that had been mentioned before do not represent the exact values
because initial tenderness may occur at temperatures as high as 120oC or as low as
110oC, depends on the mix characteristics. The mix also may continue to show sign
of tenderness characteristics at temperature as high as 95oC or as low as 80oC.
22
Tender mixes often are difficult to compact to the required density.
According to HMA Paving Handbook (2000), a number of different techniques can
be used to compact middle temperature tender mixes to the required level of density.
First, tender mixes generally do not become tender until the mix temperature falls.
This mean a little compactive effort can be use to the mix before it becomes tender
and start to move. The roller should make as many passes as possible over the
material and as quick as possible before the mix begin to move, check or mark.
Once the movement starts, additional passes of the roller should not be made. For
most tender mixes, three to five passes of the roller can be made over each point in
the material surface before the movement or checking begins. Second, if the mix is
moving under the roller in the middle temperature range, it should be kept off the
mix until it cools to the point where it is stable enough to support the weight of the
compacter. For some mixes, several shoving of the mix may occur at the outside
edge of a steel wheel roller.
2.8
Rutting
Road network is important in Malaysia as in most of other countries. In order
to have an efficient road network, it has to be maintained to an acceptable standard.
Pavement distress in flexible pavements can be categorized into several types which
are cracks, surface deformation, potholes, patches and many more. Permanent
deformation is the distress that is characterized by a surface cross section that is no
longer in its design position. It is called “permanent” deformation because it
represents an accumulation of small amounts of deformation that occurs each time a
load is applied. This deformation cannot be recovered.
Rutting is defined as the accumulation of small amounts of unrecoverable
strain resulting from applied wheel load to HMA pavement. Permanent deformation
or rutting appears as longitudinal depressions in the wheel paths of asphalt concrete
(Wong Yee Ching, 2005). Rutting may occur at one or both wheel path of a lane.
23
Wheel path rutting is the most common form of permanent deformation. While
rutting can have many sources for example underlying HMA weakened by moisture
damage, abrasion, and traffic densification, it has two principal causes. In one case,
the rutting is caused by too much repeated stress being applied to the sub-grade or
sub-base or base below the asphalt layer as shown in Figure 2.8. Although stiffer
paving materials will partially reduce this type of rutting, it is normally considered
more of a structural problem rather than a materials problem. Essentially, there is
not enough pavement strength or thickness to reduce the applied stresses to a
tolerable level. A pavement layer that has been unexpectedly weakened by the
intrusion of moisture may also cause it. The deformation occurs in the underlying
layers rather than in the asphalt layers.
Figure 2.8: Stress Applied to the Sub-grade or Base below the Asphalt Layer
Although rutting seen to be minor road defect in Malaysia, it will not only
decrease the optimum service life of the pavement but also will creates safety hazard
to the public. Several approaches must be taken such as reconstruction of the
pavement or recycling bituminous surfacing but Jabatan Kerja Raya (JKR) suggested
reconstruction of the pavement and strengthening base or sub-base.
24
2.9
Wheel Tracking Machine
Study by Wong Yee Ching, (2005) indicated that various forms of full scale
track tests and laboratory simulated wheel tracking models have been adopted to
evaluate rutting of pavement materials. However, laboratory wheel tracking tests
remain the most practical tool to study the rutting behaviour of pavement materials
due to economical factor. The machine that is used in this study is Three Wheel
Immersion Tracking adopted by Transport and Road Research Laboratory (TRRL),
(1951) of United Kingdom (UK) (Wasage et al., 2004). The function of this machine
is to evaluate rutting on the pavement while loaded by a moving wheel to simulate
moving traffic loads.
2.9.1
Wheel Tracking Apparatus
The Three Wheel Immersion Tracking Machine used in this study consists of
a few components which are as listed below:
i.
Mainframe: A rigid welded steel fabrication mounted on four inch
freestanding anti-vibration pads.
ii.
Water tank: Heavy gauge continuously welded stainless steel, bolted
inside mainframe.
iii.
Tyre wheel: Maximum three inches No. 200 mm diameter x 45 mm
wide freely rotating Rubber Tyre wheel with hardness of 80o + 3o
IRHD each mounted on independent and interchangeable pivoted
arms to suit specimen arrangement.
iv.
Specimen mould: heavy steel welded and galvanized with loose
knockout base plate length of 407 mm, depth of 90 mm and different
width of 137 mm, 214 mm and 443 mm.
25
v.
Recorder: 150 mm diameter x 150 long x 9.6 hour rotation drum
recorder fitted with three inches No. pens aligned to record a
maximum 20 mm depth measurement at the centre point of each
specimen running surface.
Figure 2.9 show the wheel tracking machine that was used in this study that
are available at the Highway and Transportation Laboratory, UTM.
Figure 2.9: Three Wheel Immersion Tracking Machine
CHAPTER 3
METHODOLOGY
3.1
Introduction
A test plan was designed to achieve the objective as shown in Figure 3.1.
Gradation Design of ACW20
Control mix and Tender mix
Preparation of
Marshall Sample
Volumetric properties
Optimum bitumen content
Stability and flow test
Sample Preparation for Wheel
Tracking Machine Test
Determine
number of roller
passes to get 7.0
+ 1% void.
Conduct Wheel Tracking Test
Number of tires passes for each
sample (500, 1000, 2000 and
5000 passes)
Data Collection and Analysis
Rut depth vs. number of roller
Conclusions and Recommendations
Figure 3.1: Flow Chart of the Laboratory Works
27
Laboratory tests performed on the aggregates were ASTM C127 Specific
Gravity and Absorption of Coarse Aggregate, ASTM C128 Specific Gravity and
Absorption of Fine Aggregate and ASTM C136 Sieve Analysis of Fine and Coarse
Aggregate. Loose HMA sample for each mix are then tested using AASHTO T 20982 Maximum Specific Gravity of Bituminous Paving Mixtures and the effective
specific gravity of the aggregate was determined. Volumetric properties Voids in the
Total Mix (VTM), Voids in the Mineral Aggregates (VMA) and Voids Filled with
Bitumen (VFB) analysis and OBC were obtained using effective specific gravity of
aggregate. Marshall stability and flow test were conducted on compacted samples.
Samples for wheel tracking machine were prepared using OBC and compacted using
number of roller passes that could achieve voids of 7 + 1%. The samples were tested
with the machine and analysis and conclusion are based from the result of the
correlation of rut depth and number of passes.
3.2
Gradation Design
Aggregate gradation is the distribution of particles sizes expressed as a
percent of the total weight. Gradation is determined by passing the material through
a series of sieves sizes stacked with the bigger opening at the top and gradually
decreasing to smaller openings. Certain descriptive terms used in referring to
aggregate gradation are as follows:
•
Coarse aggregates is referred to all material retained on the 2.36 mm
sieve.
•
Fine aggregates is referred to all material passing the 2.36 mm sieve.
•
Mineral filler is referred to all material passing the 0.075 mm sieve.
For this study, gradation designs are shown in Table 3.1 for control mix and
Table 3.2 for tender mix. The plots of the mix gradations are shown in Figure 3.2.
The sieve size is raised to 0.45 power. One mix designed had a typical gradation
28
described as control mix conformed to JKR specification and the other mix designed
close to MDL to simulate tender mix.
Table 3.1: Gradation Design of ACW20 Control Mix
Sieve
^0.45
%
Sizes
Power Passing Bottom
Top
28
4.48
100
100
100
20
3.85
94
76
100
14
3.28
80
64
89
10
2.82
72
56
81
5
2.06
58
46
71
3.35
1.72
49
32
58
1.18
1.08
33
20
42
0.425
0.68
22
12
28
0.15
0.43
12
6
16
0.075
0.31
6
4
8
Table 3.2: Gradation Design of ACW20 Tender Mix
Sieve
^0.45
%
Sizes
Power Passing Bottom
Top
28
4.48
100
100
100
20
3.85
90
76
100
14
3.28
78
64
89
10
2.82
68
56
81
5
2.06
48
46
71
3.35
1.72
36
32
58
1.18
1.08
27
20
42
0.425
0.68
17
12
28
0.15
0.43
10
6
16
0.075
0.31
5
4
8
29
100
90
80
% passing
70
60
50
bottom
40
top
30
control
20
MDL
tender
10
0
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
sieve sizes ^0.45
Figure 3.2: Gradation Chart of Control Mix and Tender Mix for ACW20
3.3
Laboratory Test Procedure
Control mix and tender mix of ACW 20 mixes are prepared using Marshall
mix design method conformed with Standard Specification for road Work
(JKR/SPJ/1988). Laboratory tests were conducted at Highway & Transportation
Laboratory, UTM. The laboratory tests procedures followed The American Society
for Testing and Materials (ASTM) and American Association of State Highway and
Transportation Officials (AASHTO) method.
To prepare the sample for the mix, material should follow in accordance to
JKR/SPJ/1988 which is as stated below:
i.
The aggregate for asphaltic concrete shall be a mixture of coarse and
fine aggregates, and mineral filler,
ii.
The bituminous binder shall be bitumen of penetration grade (PG) 80100,
iii.
Mineral filler for this study is Ordinary Portland Cement (OPC),
30
iv.
Sieve analysis will be in conducted according to ASTM C 136 and
AASHTO T 27, and
v.
Aggregate blending is used to determine the proportion of aggregates
needed for a specified mix. The gradation limits for ACW 20 mixes
that will be prepared is as specified by JKR/SPJ/1988.
3.3.1
Sieve Analysis of Fine and Coarse Aggregates (ASTM C 136-84a)
The purpose of sieve analysis is to determine the grading of aggregate sizes
from the largest sieve sizes of 28 mm to 75 µm. Sieve analysis consists of dry sieve
is analysis and washed sieve analysis. The apparatus are:
i.
Balance,
ii.
Sieves (20 mm to 75 µm),
iii.
Mechanical Sieve Shaker, and
iv.
Oven.
The procedures for dry sieve analysis:
i.
The samples were dried at temperature of 110 + 5 0C.
ii.
Desired sizes of sieve were selected and nested in order of decreasing
sizes of openings from top to bottom. The samples were then placed
on the top sieve. Agitate the sieves by mechanical apparatus for a
sufficient period (Figure 3.3).
iii.
The quantity of material on a given sieve is limits so that all particles
have opportunity to reach sieve openings during the sieving operation.
iv.
Sieving was continued for a sufficient period until there is no remains
on any individual sieve will pass the sieve by continuous hand sieving.
v.
Each size of aggregate was weighed, separated and stored in container
(Figure 3.4).
31
Dry sieve alone is usually satisfactory for routine testing but if the aggregate
contains fine dust which may cling to the coarser aggregate particles, a washed sieve
analysis should be made. The procedures for washed sieve analysis are:
i.
Samples are prepared in the same way as for dry sieve analysis.
ii.
Two (2) sieves sizes are nested accordingly which are 600 µm and 75
µm. Samples were placed on the 600 µm sieve.
iii.
Contents of the container are agitated vigorously and wash water is
poured over the nested sieves.
iv.
Repeat the operation until wash water is clear.
v.
Returned the material retained on the sieves and the washed aggregate
dried to a constant weight.
vi.
Materials are then weighed with the loss in weight representing the
amount of material finer than 75 µm.
Figure 3.3: Sieves Agitated by Mechanical Apparatus
Figure 3.4: Aggregate Separated and Stored in Container According to Sizes
32
3.3.2
Specific Gravity and Absorption of Coarse Aggregate (ASTM C 127-88)
The method is used to determine bulk specific gravity of coarse aggregates.
The method covers the determination of specific gravity and absorption of coarse
aggregate. The apparatus are:
i.
Balance,
ii.
Sample Container,
iii.
Water Tank, and
iv.
Sieves.
The procedures for determining the specific gravities of coarse aggregates are
outlined as follows:
i.
The minimum weight of test sample used is given as below:
Nominal Maximum Minimum
Weight of Size, mm (in.)
Test Sample, kg (lb.)
12.5 (½) or less
2 (4.4)
19.0 (¼)
3 (6.6)
25.0 (1)
4 (8.8)
37.5 (1½)
5 (11)
50 (2)
8 (18)
63 (2½)
12 (26)
75 (3)
18 (40)
90 (3½)
25 (55)
100 (4)
40 (88)
112 (4½)
50 (110)
125 (5)
75 (165)
150 (6)
125 (276)
33
ii.
Test sample is oven dried to a constant weight at a temperature of 110
± 5°C, cooled in air at room temperature for 1 to 3 hours. Then, the
aggregate immersed in water for a period of 24 hours.
iii.
After 24 hours, the test sample was removed from the water and rolled
in a large absorbent cloth until all visible films of water were
removed. The larger particles were wiped individually.
iv.
Test sample then weighed in the saturated surface-dry condition.
Weight was recorded as B to the nearest 1.0 g.
v.
The saturated-surface-dry test sample immediately placed in the
sample container and the weight in water were determined as C.
Make sure that all entrapped air was removed before weighing by
shaking the container while immersed.
vi.
The test sample was dried to constant weight at a temperature of 110
± 5°C, cooled in air at room temperature 1 to 3 hours until the samples
is comfortable to handled and weighed. This weight was recorded as
A in the calculations.
vii.
The specific gravities and water absorption are then calculated as
follow:
Bulk Specific Gravity = A / (B - C)
Absorption = (B – A) 100 / A
where,
A = weight of oven-dry test sample in air, g,
B = weight of saturated-surface-dry test sample in air, g, and
C = weight of saturated test sample in water, g
The specific gravity results are reported to the nearest 0.01 and the absorption
result to the nearest 0.1%.
34
3.3.3
Specific Gravity and Absorption of Fine Aggregate (ASTM C 128-88)
The method is used to determine bulk specific gravity of fine aggregates.
The method covers the determination of specific gravity and absorption of fine
aggregate. The apparent and bulk specific gravity on the basis of weight of saturated
surface-dry aggregate determine based on aggregate after 24 hours soaking in water.
The apparatus are:
i.
Balance,
ii.
Pycnometer,
iii.
Mold, and
iv.
Tamper.
The procedures for determining the specific gravities of fine aggregates are
outlined as follows:
i.
Approximately 700 g of fine aggregate was prepared according to the
gradation of each mix.
ii.
The aggregate was dried to constant weight at a temperature of 110 ±
5°C. Then, it was cooled to a comfortable handling temperature and
6% of water was added and leaved it for 24 hours.
iii.
After 24 hours, the sample was spread and dried on a flat pan,
exposed to a slow moving fan. To secure a homogenous drying, the
sample was stirred frequently.
iv.
The cone test then runs on the sample to determine whether or not
surface moisture is present on the aggregate particles. The sample
was placed loosely in the mold and tamped with 25 light drops of the
tamper about 5 mm above the top surface of the sample.
35
v.
The mold was lifted vertically and the slumps were determined. If the
fine aggregate slumps slightly, it indicated that it has reached a
surface-dry condition. If the fine aggregate still retain the molded
shape, continue drying it and run the cone test again until the sample
slumps.
vi.
Fine aggregate in surface-dry condition was weighed and recorded as
S.
vii.
The pycnometer partially filled with water. Then, 500 ± 10 g of
saturated surface-dry fine aggregate and additional water
approximately 90% of capacity was filled into the pycnometer. The
pycnometer was rolled and agitated to eliminate all air bubbles for
about 20 minutes.
viii.
The water level in the pycnometer was brought to the calibration
capacity by added more water. The total weight of the pycnometer,
specimen and water was recorded as C.
ix.
The fine aggregate was then removed from the pycnometer and dried
to constant weight at a temperature of 110 ± 5°C. Then, it was cooled
in air at room temperature for 1 hour and weighed and recorded as A.
x.
The specific gravities are then calculated as follow:
Bulk Specific Gravity = A / (B + S - C)
where,
A = weight of oven-dry specimen in air, g,
B = weight of pycnometer filled with water, g
S = weight of the saturated surface-dry specimen, g, and
C = weight of pycnometer with specimen and water to calibration, g
36
3.3.4
Theoretical Maximum Specific Gravity and Density of Bituminous
Paving Mixtures (ASTM D 2041-91)
The method covers the determination of the theoretical maximum specific
gravity and density of uncompacted bituminous paving mixtures at 25°C. The
apparatus are:
i.
Vacuum Container (bowl),
ii.
Balance, and
iii.
Vacuum pump, capable of evacuating air from the vacuum container
to a residual pressure of 30 mm of Hg.
The procedures for determining the effective specific gravities of aggregates
are outlined as follows:
i.
Size of the sample are conformed to the following requirements:
Size of Largest Particle of
Aggregate in Mixture, mm (in.)
50 (2)
6000
37.5 (1½)
4000
25.0 (1)
2500
19.0 (¼)
2000
12.5 (½)
1500
9.5 (2/8)
1000
4.75 (No.4)
i.
Minimum Sample Size, g
500
The particles of the paving mixtures sample were separated by hand
so that the particles of fine aggregate portion are not larger than 6.3
mm (1/4 inches).
ii.
The sample were cooled to room temperature, placed in the bowl,
weighed and recorded as A.
37
iii.
Sufficient amount of water added in the bowl covered the sample at
least 1 inch above the sample.
iv.
Air trapped in the sample was removed by applying gradually
increased vacuum until the pressure manometer reads 25 mm of Hg.
This pressure was maintained for 10 minutes (Figure 3.5).
v.
During the vacuum period, the bowl and sample was agitated by a
mechanical device.
vi.
After 10 minutes, the vacuum was gently released. The bowl and
sample are placed in the water, weighed and recorded as C.
vii.
The theoretical maximum specific gravity of the sample can be
calculated as below:
Gmm = A/(A+B-C)
where,
A = weight of sample in air, g,
B = weight of bowl in water, g, and
C = weight of bowl and sample in water, g
Figure 3.5: The ASTM D 2041 Test Apparatus
38
3.3.5
Resistance to Plastic Flow of Bituminous Mixtures Using Marshall
Apparatus (ASTM D 1559)
The test covers the measurement of resistance to plastic flow of the asphalt
pavement specimens using the Marshall apparatus. Two principal features of the
Marshall method of mix design are a density-voids analysis and a stability-flow
test of the compacted test specimens.
a)
i.
Apparatus
Marshall testing machine, a compression testing device designed to
apply loads to test specimens through semi-circular testing heads at a
constant rate of strain of 51 mm (2 in.) per minute.
ii.
Water bath, at least 150 mm (6 in.) deep and thermostatically
controlled to 60°C ± 1°C.
b)
i.
Procedure
The optimum bitumen content (OBC) for a particular gradation of
aggregates was determined by preparing 15 test specimens with 75
blows for a range of different asphalt contents so that the test data
curves show a well-defined optimum value.
ii.
The specimens that have been prepared with OBC value were
immersed in water bath at 60°C for 30 to 40 minutes before test
(Figure 3.6).
iii.
When the testing apparatus is ready, the test specimen is removed
from the water bath, dried the surface and placed it in the lower
segment of the breaking head.
iv.
Then the upper segment of the breaking head was placed on the
specimen. The complete assembly was located in loading device.
v.
The flow was recorded before the specimen being loaded.
39
vi.
Next, the load was applied to the specimen by constant movement of
50.8 mm minimum until the maximum load is reached and recorded
as Marshall stability (Figure 3.7).
vii.
The flow after the failure was recorded. The last flow value recorded
was deducted to the earliest flow value. This will indicates as a flow
value in mm.
viii.
Noted that the entire procedure for stability and flow tests shall be
completed within 30 s.
Figure 3.6: Specimen Immerse in Water Bath at 60oC
Figure 3.7: Compression Testing Machine
40
3.4
Mixing Specimens
In this process, the mixing of specimen is according to the specification.
Mixing temperature is 150oC to 160oC and for the compaction, the temperature is
between 130oC to 140oC. The OBC is taken from the Marshall test result. The
procedures for samples mixing should be followed accordingly in order to produce
good quality of sample. The procedures are:
The aggregates were oven dried at 160oC in the oven 24 hours before
i.
the mixing process.
ii.
Bitumen was placed in the oven at 130oC 2 hours before mixing
sample.
iii.
Mixing bowl heated first before placing the sample into it.
iv.
Bitumen was added when the aggregate in the bowl reached 160oC
and mixing process begun.
v.
Mixing process continued until all the aggregates were covered with
bitumen. The mixed samples were then placed into the steel mould.
vi.
Temperature is checked again in order to get a suitable temperature
for compaction which is 130oC to 140OC. Once it reached the
compaction temperature, the mix is ready to be compact (Figure 3.8).
Figure 3.8: Sample Place and Check for Compaction Temperature
41
3.4.1
Sample Compaction
Good compaction will increases the resistance of pavement to deformation
and improves the durability. Two samples for each mix were prepared for wheel
tracking test. Before conducting the test, each sample should be compacted using a 9
kilograms of roller to determine the number of roller passes and obtain voids of 7 +
1% on each samples (Figure 3.9).
Figure 3.9: Sample after Compaction with 9 Kilograms of Steel Roller
Trial and error method was carried out in order to determine the number of
roller passes for compaction purposes. For this study, the initial number of roller
passes was 250 passes. The test procedures for determining number of roller passes
are:
i.
With the appropriate compaction temperature, the sample was
compacted starting with 250 passes up until the VTM of 7 + 1%.
ii.
The compacted samples were cooled to room temperature and
extracted from the mould to conduct Bulk Specific Gravity test of
Compacted Bituminous Mixtures Using Saturated Surface-Dry
Specimens (ASTM D 2726).
iii.
VTM was calculated and plotted against number of roller passes.
iv.
Determined the number of roller passes on the graph at VTM of 7 +
1%.
42
3.5
Wheel Tracking Machine Test
After achieving the compaction of 7 + 1% air voids, the samples will be test
with wheel tracking machine. The test procedures are as follows:
i.
Before conducting the test, the well compacted samples are placed in
the oven for 3 to 5 hours at temperature of 60oC (Figure 3.10).
ii.
Meanwhile, maintained the temperature of water in wheel tracking
machine at temperature of 60oC. The speed of the wheel calibrate to
40 passes per minute (Figure 3.11).
iii.
After all parameter is ready, samples were removed from the oven and
immersed into the wheel tracking machine.
iv.
Before running the test, an initial reading should be taken and after
each of 500, 1000, 2000 and 5000 passes. Reading is taken at three
points on each sample to get the average reading.
v.
After the 5000 passes reading is taken, plots the graph of rut depth vs.
number of roller passes for each mix to summarize the result (Figure
3.12).
Figure 3.10: Well Compacted Sample
43
Figure 3.11: Maintain Water Temperature in Wheel Tracking Machine
Figure 3.12: Reading at Three Point for Each Sample
3.6
Data Analysis
3.6.1
Volumetric Properties of Compacted Mixtures
There are there (3) volumetric properties most commonly measured to
evaluate the physical characteristic of HMA which are voids in total mix (VTM),
voids in the mineral aggregate (VMA) and voids filled with bitumen (VFB) (Huner
and Brown, 2001).
The air voids, VTM in the total compacted asphalt mixture consists of small
spaces of air between the coated aggregate particles throughout a compacted paving
mixture, expressed as percent of the total volume of the compacted paving mixture.
The VTM was determined by equation:
44
⎡ G − Gmb ⎤
VTM = 100 × ⎢ mm
⎥
⎣ Gmm ⎦
Equation 3.1
where,
VTM = air voids in compacted mixture, percent of total volume
Gmm = maximum specific gravity of paving mixture (determined
directly for a paving mixture by ASTM D 2041/AASHTO T
209)
Gmb
= bulk specific gravity of compacted mixture (ASTM D 1188
or D 2726/AASHTO T 166)
The voids in the mineral aggregate, VMA, are defined as the intergranular
void space between the aggregate particles in a compacted paving mixture that
includes the air voids and the effective asphalt content, expressed as a percent of the
total volume. The VMA is calculated on the basis of the effective or bulk specific
gravity of the aggregate and is expressed as a percentage of the bulk volume of the
compacted paving mixture. Therefore, the VMA can be calculated by subtracting the
volume of the aggregate determined by its bulk specific gravity from the bulk
volume of the compacted paving mixture. If the mix composition is determined as
percent by mass of total mixture, VMA is determined by equation:
⎡G P ⎤
VMA = 100 − ⎢ mb s ⎥
⎣ G sb / e ⎦
Equation 3.2
where,
VMA = voids in mineral aggregate (percent of bulk volume)
Gsb/e = bulk or effective specific gravity of total aggregate
Gmb
= bulk specific gravity of compacted mixture (ASTM D 1188
or D 2726/AASHTO T 166)
Ps
= aggregate content, percent by total mass of mixture
45
Voids Filled with Bitumen, VFB is the percentage portion of the volume of
intergranular void space between the aggregate particles that is occupied by the
effective asphalt. In other words, the VFB is the percentage of the VMA that are
filled with asphalt binder. VFB is determined by equation:
⎡VMA − VTM ⎤
VFB = 100 − ⎢
⎥⎦
VMA
⎣
Equation 3.3
where,
VFB = voids filled with bitumen, percent of VMA
VMA = voids in mineral aggregate, percent of bulk volume
VTM = air voids in compacted mixture, percent of total volume
3.6.2
Optimum Bitumen Content
The optimum bitumen content (OBC) for HMA can be determined from the
data obtained. From Foster, (1982) view, Marshall mix design procedures are the
best tools available for determining optimum asphalt content for a given aggregate
blend, optimum being the maximum amount that can be put into the mix without
having too much. There are two methods to determine optimum asphalt content that
is Asphalt Institute (AI) method and National Asphalt Pavement Association
(NAPA) method. The optimum asphalt content in this study was determined using
AI method. In this method, the optimum bitumen content of the mix is determined
from data obtained range from different bitumen content. Series of curves are
plotted to get the density, percent of VFB, percent of VTM, Marshall stability and
flow value. The OBC are then determined by the average of selected four series of
curve which are VTM at 4%, VFB at 80%, peak curves of density and maximum
Marshall stability.
46
3.6.3
Wheel Tracking Test Result
Samples for wheel tracking machine were prepared using OBC and
compacted using number of roller passes that could achieve voids of 7 + 1%. The
samples were then tested with the wheel tracking machine for 500, 1000, 2000 and
5000 passes. Analysis and conclusion will be base from the result of the relationship
between rut depth and number of passes. Analysis using paired t-Test will identify if
any significance difference exist between control mix and tender mix.
3.6.4
Standard Specification
All Marshall samples and wheel tracking samples were prepared according to
the JKR/SPJ/1988 as a guideline. Tables 3.3 show the appropriate envelopes for
ACW20 gradation of control mix and tender mix that used in this study. The design
bitumen contents used in the mix design were shown in Table 3.4 for asphaltic
concrete mix. Parameter resulted and analysed from this study will be compared to
the JKR/SPJ/1988 requirements given in Table 3.5 for further interpretation
47
Table 3.3: Gradation Limits for Asphaltic Concrete
Mix Type
Wearing Course
Mix Designation
ACW 20
B.S. Test Sieve
% Passing by Weight
28.0 mm
100
20.0 mm
76 - 100
14.0 mm
64 – 89
10.0 mm
56 – 81
5.0 mm
46 – 71
3.35 mm
32 – 58
1.18 mm
20 - 42
425 μm
12 – 28
150 μm
6 - 16
75 μm
4–8
Table 3.4: Design Bitumen Contents
Mix Type
ACW 20 - Wearing Course
Bitumen Content (%)
4.5 - 6.5
Table 3.5: Test and Analysis Parameters for Asphaltic Concrete
Parameter
Wearing Course
Stability, S
> 500 kg
Flow, F
> 2.0 mm
Stiffness, S/F
> 250 kg/mm
Air voids in mix, VTM
3.0% - 5.0%
Voids in aggregate filled with bitumen, VFB
75% - 85%
CHAPTER 4
RESULT AND ANALYSIS
4.1
Introduction
The findings in this study were obtained from the laboratory test and were
presented in this chapter. The findings were also compared to the JKR/SPJ/1988
specification. Through the laboratory work, Marshall samples were prepared to
determine the optimum bitumen content (OBC), volumetric properties and Marshall
stability and flow. After that, samples for wheel tracking test were prepared and test
to determine number of roller passes required to compact the samples. The
compaction of the samples should conformed 7 + 1% air voids. Specific gravity for
each samples that had been compacted were then determined before conducting the
test and after conducting the wheel tracking test. With number of roller passes
determined, two samples for each mix were prepared and run simultaneously with
wheel tracking machine. Initial readings were taken before the test run and after
each 500, 1000, 2000 and 5000 passes. The relationship between rut depth and
number of roller passes of control mix and tender mix were determined from the
graph plotted.
49
4.2
Aggregate Gradation
Two gradations were design for ACW20 conformed to JKR/SPJ/1988
specification. One gradation consist a typical gradation of dense graded design away
from maximum density line (MDL) described as control mix. The other gradation
was design close to MDL to simulate tender mix described as tender mix. The
aggregate gradation for control mix and tender mix are shown in Table 4.1 and Table
4.2 respectively. The calculations for mineral filler are calculated according to the
result of washed sieve analysis. The calculations can be referred in Appendix C.
Figure 4.1 show the plotted graph for both gradations. The aggregates blending for
wheel tracking test sample for control mix and tender mix are shown in Table 4.3
and Table 4.4 respectively. All sieve size is raised to 0.45 power and total weight of
Marshall sample prepared was 1,200 grams and 11,000 grams for wheel tracking
sample.
Table 4.1: Gradation of ACW20 Control Mix for Marshall Sample
Sieve
^0.45
%
Sizes
Power Passing Bottom
% Retain
Mass Retain Each
Top
Each Size
Size, g
28
4.48
100
100
100
0
0
20
3.85
94
76
100
6
72
14
3.28
80
64
89
14
168
10
2.82
72
56
81
8
96
5
2.06
58
46
71
14
168
3.35
1.72
49
32
58
9
108
1.18
1.08
33
20
42
16
192
0.425
0.68
22
12
28
11
132
0.15
0.43
12
6
16
10
120
0.075
0.31
6
4
8
6
72
PAN
6
Filler
8.8
2% OPC
24
50
Table 4.2: Gradation of ACW20 Tender Mix for Marshall Sample
Sieve
^0.45
%
Sizes
Power Passing Bottom
% Retain
Mass Retain Each
Top
Each Size
Size, g
28
4.48
100
100
100
0
0
20
3.85
90
76
100
10
120
14
3.28
78
64
89
13
144
10
2.82
68
56
81
10
120
5
2.06
48
46
71
19
240
3.35
1.72
36
32
58
12
144
1.18
1.08
27
20
42
9
108
0.425
0.68
17
12
28
10
120
0.15
0.43
10
6
16
7
84
0.075
0.31
5
4
8
5
60
PAN
5
Filler
0
2% OPC
24
100
bottom
90
top
control
80
MDL
70
tender
% passing
60
50
40
30
20
10
0
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
sieve sizes ^0.45
Figure 4.1: Gradation Chart of Control Mix and Tender Mix for ACW20
5
51
Table 4.3: Gradation of ACW20 Control Mix for Wheel Tracking Sample
Sieve
^0.45
%
Sizes
Power Passing Bottom
% Retain
Mass Retain Each
Top
Each Size
Size, g
28
4.48
100
100
100
0
0
20
3.85
94
76
100
6
660
14
3.28
80
64
89
14
1540
10
2.82
72
56
81
8
880
5
2.06
58
46
71
14
1540
3.35
1.72
49
32
58
9
990
1.18
1.08
33
20
42
16
1760
0.425
0.68
22
12
28
11
1210
0.15
0.43
12
6
16
10
1100
0.075
0.31
6
4
8
6
660
PAN
6
Filler
80.2
2% OPC
220
Table 4.4: Gradation of ACW20 Tender Mix for Wheel Tracking Sample
Sieve
^0.45
%
Sizes
Power Passing Bottom
% Retain
Mass Retain Each
Top
Each Size
Size, g
28
4.48
100
100
100
0
0
20
3.85
90
76
100
10
1100
14
3.28
78
64
89
13
1430
10
2.82
68
56
81
10
1100
5
2.06
48
46
71
19
2090
3.35
1.72
36
32
58
12
1320
1.18
1.08
27
20
42
9
990
0.425
0.68
17
12
28
10
1100
0.15
0.43
10
6
16
7
770
0.075
0.31
5
4
8
5
550
PAN
5
Filler
0
2% OPC
220
52
4.3
Result of Volumetric Properties
The volumetric properties consist of VTM, VMA and VFA for ACW 20
control mix and tender mix were calculated using effective specific gravity of the
aggregate obtained from Maximum Specific Gravity of Bituminous Paving Mixtures
(AASHTO T 209-82) test. The optimum bitumen content (OBC) of the mix is
determined from data obtained range of different bitumen content. Series of curves
are plotted to get the density, percent of VFB, percent of VTM, Marshall stability
and flow value. The OBC are then determined by Asphalt Institute (AI) method
which is by the average of selected four series of curve. The curves are VTM at 4%,
VFB at 80%, peak curves of density and maximum Marshall stability. The results of
volumetric properties at OBC are shown in Table 4.5.
Table 4.5: Volumetric Properties of ACW20 Control Mix and Tender Mix at OBC
Volumetric
ACW20
Properties
Control mix
Tender Mix
JKR Specification
OBC (%)
5.1
5.4
4.5-6.5
Stability, S (kg)
1485
1112
> 500
Flow, F (mm)
5.66
7.39
> 2.00
VMA (%)
14.80
15.00
-
VFB (%)
78.8
82.5
75-85
Stiffness (kg/mm)
262.6
150.5
>250
Based from the result, the VMA for tender mix is unexpectedly higher than
control mix which results in higher OBC. The OBC determined for control mix and
tender mix was performed using AI method. The result of OBC for tender mix was
5.4% higher than control mix which is 5.1%. Resulting from a higher OBC of tender
mix, the VMA for tender mix also had become slightly higher than control mix.
VMA for tender mix is 15.0% and control mix is 0.2% less than tender mix which is
14.8%. The flow for tender mix also higher than control mix which is 7.39 mm for
tender mix and 5.66 mm for control mix. This shows that although the gradation of
tender mix was design close to MDL, the VMA is still higher than gradation design
53
away from MDL and it could be due to other factor. It is important to note that the
determination of OBC in this study was performed using AI methods. If it would
have been National Asphalt Pavement Association (NAPA) method, the OBC would
be 4.7% compared to 5.4% determine by AI method. Interestingly for stability and
stiffness, tender mix indicated lower value than control mix. Stability for tender mix
is 1112 kg compared to control mix which is 1485 kg. Same goes for stiffness where
tender mix also indicated lower value than control mix where stiffness for tender mix
is 150.5 kg/mm and control mix is 262.6 kg/mm. High stability and stiffness for
control mix suggested that the further the gradation from MDL, the stronger the mix.
However for stiffness value, tender mix failed the specification requirement.
4.4
Result of Wheel Tracking Test
Before conducting the test, samples were prepared by compacting the
samples using 9 kilograms of steel roller and determine the number of roller passes
so that the VTM of the samples should be 7 + 1% air voids. Wheel tracking test can
be conducted after desired VTM for control mix and tender mix are achieved. Table
4.6 shows the result of the test conducted to determine the number of compaction to
get 7 + 1% air voids and also the result of VTM before and after samples were run
with wheel tracking machine.
Table 4.6: Number of Compaction of Wheel Tracking Sample
Number of
Voids in Total Mix, VTM (%)
ACW20
Compaction
Before
After
Control Mix
700
7.51
7.18
Tender Mix
450
7.41
5.54
Result from Table 4.6 shows that the number of roller passes required to
compact control mix is 700 which are higher than tender mix where 450 roller passes
is sufficient for tender mix to achieve 7 + 1% of VTM. The VTM of the samples are
then determined before and after wheel tracking test and from the result indicated
54
that total amount of VTM decreasing for control mix is 0.33% from 7.51% to 7.18%
compared to tender mix where tender mix shows higher percentage of decreasing
which is 1.87% from 7.41% to 5.54% of VTM. Even though tender mix shows less
number of compaction required, the surface of the samples after compaction was not
smooth and uniform. The surface of the samples also shows signs of bleeding. The
calculations of this result are attached in Appendix D.
Two samples were prepared according to the number of compaction and run
simultaneously during the wheel tracking test. Data were taken before the test run as
initial reading and taken after each of 500, 1000, 2000 and 5000 passes. Reading is
mark at three points on each sample to get the average reading. The summaries of
the results from the test are shown in Table 4.7 and plotted in Figure 4.2 while the
full data collected during the test are attached in Appendix E.
Table 4.7: Summary of Data from the Wheel Tracking Machine Test
Rut Depth (mm)
Roller Passes
Control Mix
Tender Mix
Initial Reading
0
0
500
3.84
4.07
1000
5.29
5.60
2000
6.26
8.05
5000
9.75
11.75
55
14.00
12.00
rut depth, mm
10.00
8.00
6.00
tender mix
control mix
Power (tender mix)
Power (control mix)
4.00
2.00
0.00
0
1000
2000
3000
4000
5000
6000
roller passes
Figure 4.2: Rut Depth vs. Number of Roller Passes of Control Mix and Tender Mix
Based from Table 4.7 and graph plotted in Figure 4.2, the result indicated that
tender mix has slightly higher rut depth as compared to control mix. The rut depth
increase as the number of passes increase. Rut depth between control mix and tender
mix started to show differences at 2000 passes and after 5000 passes, the rut depth
are 9.75mm and 11.75mm for control and tender mix respectively. Other analysis
was also carried out which are the paired t-Test statistical analysis. Based from the
paired t-Test analysis result, the p-value was 0.0239 which is less than 0.05 (p<0.05).
This is meaning that there are significant different between rut depth of control mix
and rut depth of tender mix.
CHAPTER 5
CONCLUSIONS AND RECOMMENDATIONS
5.1
Introduction
In this chapter, the conclusion and recommendation are based on the
evaluation on the properties of tender mix that had been conducted on the laboratory.
In this study, one types of HMA mixes chosen for the purpose study which is
ACW20. Two gradations were designed described as control mix and tender mix.
The volumetric properties were analysed as well as rutting behaviour using wheel
tracking machine.
5.2
Summary of the Findings
Generally, based from the result it was found that even though the gradation
of tender mix was designed close to MDL but the VMA of tender mix was
unexpectedly higher than control mix where gradation of control mix was designed
away from MDL. This is because the OBC determined for tender mix was higher
than control mix and therefore, this affected the VMA. The OBC of tender mix was
5.4% and control mix was 5.1%. The VMA of tender mix was also higher than
control mix which was 15.0% and control mix was 14.8%. Both stability and
stiffness for tender mix however was lower than control mix. Stability for tender
mix was 1112 kg while control mix was 1485 kg. In the other hand, stiffness for
tender mix was 150.5 kg/mm and control mix was 262.6 kg/mm. The JKR
57
requirement for stiffness value should be higher than 250 kg/mm. For tender mix,
this condition appeared to be critical because tender mix gradation design was
designed within the JKR/SPJ/1988 grading limitation yet the stiffness failed the
requirement. This means gradation that design close to MDL is not suitable even
though the VMA value is higher than gradation design away from MDL. This
indicated that plotting the gradation on the 0.45 power graph is important because of
the MDL.
For wheel tracking test, rut depth after 5000 passes for control mix was 9.75
mm while tender mix was 11.75 mm. From statistical analysis, control mix and
tender mix statistically show significance different in rut depth but no significance
differences practically based from the data collected because the differences in rut
depth between control mix and tender mix was small which is 2.00 mm. Despite all
that, the gradation for tender mix has lower stability and stiffness. Moreover, the
stiffness for tender mix failed the requirement from JKR. Therefore, it is suggested
that the gradation for tender mix is not suitable and using the gradation could expose
to rutting and other pavement deformation such as shoving or cracking.
5.3
Recommendations
The study shows that OBC and VMA of tender mix is higher than control
mix while the stability and stiffness for tender mix is lower than control mix. Note
that the OBC for this study was determined using AI method rather than NAPA
method where if it would be NAPA method, the OBC for tender mix will be lower
than control mix. The method for AI is by taking the average value from four curves
which are VTM at 4%, VFB at 80%, peak curve of density and maximum Marshall
stability. Unlike AI, NAPA method is taking bitumen content corresponds to the
VTM at 4% as the OBC for the mix. Therefore, it is recommended to use NAPA
method in determining the OBC for future study in evaluating the properties of
tender mix. Also recommended that other types of mix such as ACW10, ACW14,
Gap Graded, Stone Mastic Asphalt (SMA) or Open Graded to be used in future study
58
and designed close to MDL to simulate tender mix. As for rutting using wheel
tracking machine, it is recommended to run the test for more than 5000 passes up
until the samples fail.
59
REFERENCES
Asphalt Institute. (1983). Principles of Construction of Hot Mix Asphalt Pavements.
Manual Series No. 22.
Asphalt Institute. (1989). The Asphalt Handbook. 1989 Edition. Lexington: Asphalt
Institute.
Brown, E. R., Lord, B., Decker, D. and Newcomb, D.. (2000). Hot Mix Asphalt
Tender Zone. NCAT Report 00-02.
Button, J. W., Epps, J. A., Little, D. N., and Gallaway, B. M.. (1980). Influence of
Asphalt Temperature Susceptibility on Pavement Construction and
Performance. NCHRP Report. (268).
Chadbourn, B. A., Skok, E. L., Newcomb, D. E., Crow, B. L., and Spindler, S..
(2000). The Effect of Voids in Mineral Aggregate (VMA) on Hot Mix Asphalt
Pavement. MN/RC–2000–13. Minnesota Department of Transportation.
Crawford, C.. (1989). Tender Mixes: Probable Cause, Possible Remedies. NAPA.
(108/86)
Foster, C. R..(1982). Development of Marshall Procedures for Designing Asphalt
Paving Mixtures. NAPA Information Series 84.
60
Garber, N. J. and Hoel, L. A..(2002). Traffic and Highway Engineering. 3rd Edition.
Brooks/Cole.
Goode, J. F. and Lufsey, L. A.. (1962). A New Graphical Chart for Evaluating
Aggregate Gradation. AAPT.
Huner, M. H. and Brown, E. R..(2001). Effect of Re-heating and Compaction
Temperature on Hot Mix Asphalt Volumetrics. NCAT Report01-04.
Jabatan Kerja Raya. (1998). Standard Specification for Road Works (JKR/SPJ/1998).
Kuala Lumpur: JKR.
Kandhal, P. S. and Chakraborty, S..(1996). Evaluation of Voids in the Mineral
Aggregate for HMA Paving Mixtures. NCAT Report No. 96-4.
Mallick, R. B., Buchanan, S., Brown, E. R., and Huner, M..(1998). An Evaluation of
Superpave Gyratory Compaction of Hot mix Asphalt. NCAT Report No. 98-5.
Manual of Three Wheel Immersion Tracking Machine. Wessex Engineering and
Metalcraft Co.
Marker, V.. (1977). Tender Mixes: The Causes and Prevention. Asphalt Institute. No.
168 (IS-168).
Martin, J. R. and Wallace, H. A..(1958). Design and Construction of Asphalt
Pavement. McGraw-Hill Book Co.
Prowell, B. D., Zhang, J., and Brown, E. R.. (2005). Aggregate Properties and the
Performance of Superpave-Designed Hot Mix Asphalt. NCHRP Report (539).
61
Roberts, F. L., Kandhal, P. S., Brown, E. R., Lee, D. Y., and Kennedy, T. W. (1996).
Hot Mix Asphalt Material, Mixture Design, and Construction. 2nd Edition.
Maryland: NAPA Research and Education Foundation.
Tarrer, A. R. and Wagh, V.. (1991). Factor Influencing Mix Setting Characteristic
and Test to Predict Mix Setting Characteristic. SHRP-A/ UWP-91-508.
US Army Corps of Engineers. (2000). Hot Mix Asphalt Paving Handbook. 2000
Edition. United States of America: US Army Corps of Engineers.
Wasage, T. L. J., Ong, G. P., Fwa, T. F., and Tan, S. A. (2004). Laboratory
Evaluation of Rutting Resistance of Geosynthetics Reinforced Asphalt
Pavement. Vol. 44 Issue 2. Singapore: Centre of Transportation Research,
Department of Civil Engineering, National University of Singapore.
Wong Yee Ching. (2005). Evaluation of Rutting on Different Types of Hot Mix
Asphalt Gradation. Universiti Teknologi Malaysia: Degree Project.
62
APPENDIX A
Aggregate Bulk Specific Gravity of ACW20
Table B2.1: Bulk Specific Gravity for Coarse Aggregate
ACW 20
Sample 1
Sample 2
Sample 3
Weight of saturated test sample in water, C
650.7
651.7
650.0
1056.9
1054.2
1053.7
1045.3
1045.6
1044.5
2.57
2.60
2.59
(g)
Weight of saturated-surface-dry test sample
in air, B (g)
Weight of oven-dry test sample in air, A (g)
Bulk specific gravity = A/ (B-C)
Average bulk specific gravity
2.587
Table B2.2: Bulk Specific Gravity for Fine Aggregate
ACW 20
Sample 1
Sample 2
Sample 3
surface-dry
500.2
500.4
501.1
Weight of pycnometer filled with water, B
875.5
877.1
878.4
1173.2
1184.7
1189.1
Weight of oven-dry specimen in air, A (g)
493.3
491.2
493.6
Bulk specific gravity = A / (B + S - C)
2.43
2.55
2.59
Weight
of
the
saturated
specimen, S (g)
(g)
Weight of pycnometer with specimen and
water to calibration, C (g)
Average bulk specific gravity
i.
Aggregate Specific Gravity Blend (control mix) =
ii.
Aggregate Specific Gravity Blend (tender mix) =
2.522
100
42
58
+
2.587 2.522
100
52
48
+
2.587 2.522
= 2.548
= 2.555
63
APPENDIX B
Maximum Specific Gravity of Loose Mixture ACW20
Maximum Specific Gravity for ACW 20 Control Mix
ACW 20
Sample 1
Sample 2
Weight of sample in air, A (g)
1928.7
1928.7
Weight of bowl in water, B (g)
1392.9
1392.8
Weight of bowl and sample in water, C (g)
2533.0
2533..1
Maximum specific gravity = A/(A+B-C)
2.446
2.446
Average maximum specific gravity
2.446
Average effective specific gravity of aggregate at 4.5% AC
2.619
Maximum Specific Gravity for ACW20 Tender Mix
ACW20
Sample 1
Sample 2
Weight of sample in air, A (g)
1929.3
1929.3
Weight of bowl in water, B (g)
1392.9
1392.8
Weight of bowl and sample in water, C (g)
2522.8
2522.8
Maximum specific gravity = A/(A+B-C)
2.413
2.414
Average maximum specific gravity
2.414
Average effective specific gravity of aggregate at 5.5% AC
2.615
64
APPENDIX C
Calculations of Mineral Filler from Washed Sieve Analysis Result
Control Mix:
Mineral Filler = Before Washed Sieve – After Washed Sieve
Before Wash (mass retain from sieve 28mm to 75µm)
= 1128g
Average from two sample of washed sieve
= 1088.8g
Dust coated on the aggregates (1128g – 1088.8g)
= 39.2g
2% of OPC
= 2% х 1200g = 24g of OPC.
4% of Pan
= 4% х 1200g = 48g – 39.2g = 8.8 g of Pan.
Total weight of sample for control mix = 1128 + 24 +8.8 = 1160.8g.
Tender Mix:
Mineral Filler = Before Washed Sieve – After Washed Sieve
Before Wash (mass retain from sieve 28mm to 75µm)
= 1140g
Average from two sample of washed sieve
= 1101.9g
Dust coated on the aggregates (1128g – 1088.8g)
= 38.1g
2% of OPC
= 2% х 1200g = 24g of OPC.
3% of Pan
= 3% х 1200g = 36g (no Pan is added)
Total weight of sample for tender mix = 1140 + 24 = 1164g.
65
APPENDIX D
i.
Calculations of VTM in ACW20 Control Mix and Tender Mix
Control Mix:
TMD =
100
= 2.424
94.9
5.1
+
2.615 1.03
700 roller passes,
Bulk Specific Gravity =
8900.9
= 2.242
9003.2 − 5033.6
⎡ 2.242 ⎤
VTM = ⎢1 −
× 100 = 7.51%
⎣ 2.424 ⎥⎦
Tender Mix:
TMD =
100
= 2.418
94.6
5.4
+
2.619 1.03
450 roller passes,
Bulk Specific Gravity =
8943.7
= 2.239
9031.5 − 5036.7
⎡ 2.239 ⎤
VTM = ⎢1 −
× 100 = 7.41%
⎣ 2.418 ⎥⎦
66
ii.
Calculations of VTM Before and After Wheel Tracking Test
Control Mix:
VTM Before = 7.51%
Average Bulk Specific Gravity after Test = 2.250
⎡ 2.250 ⎤
VTM After = ⎢1 −
× 100 = 7.18%
⎣ 2.424 ⎥⎦
Therefore, the decreasing of VTM is 0.33%.
Tender Mix:
VTM Before = 7.41%
⎡ 2.284 ⎤
VTM After = ⎢1 −
× 100 = 5.54%
⎣ 2.418 ⎥⎦
Therefore, the decreasing of VTM is 1.87%.
67
APPENDIX E
Result of Wheel Tracking Test for ACW20 Control Mix and Tender Mix
Control Mix:
No.
Data
Passes
Point
Sample 1
0
1
4148
4355
2
4371
4395
3
4348
4393
Avg
4289
1
3889
3984
2
4052
3980
3
3984
3818
Avg
3975
1
3741
3710
2
3908
3851
3
3851
3777
Avg
3833
1
3592
3618
2
3812
3771
3
3741
3721
Avg
3715
1
3325
3248
2
3531
3394
3
3400
3260
Avg
3419
500
1000
2000
5000
x
0.01mm
42.89
39.75
38.33
37.15
34.19
x
Sample 2
4381
3927
3779
3703
3301
Rut Depth
0.01mm
Average
(mm)
43.81
43.35
0.00
39.27
39.51
3.84
37.79
38.06
5.29
37.03
37.09
6.26
33.01
33.60
9.75
68
Tender Mix:
No.
Data
Passes
Point
0
1
4514
4490
2
4840
4710
3
4645
4539
Avg
4666
1
4136
4149
2
4422
4311
3
4161
4115
Avg
4240
1
3997
3979
2
4226
4162
3
4042
3973
Avg
4088
1
3822
3883
2
3785
3824
3
3831
3758
Avg
3813
1
3388
3496
2
3416
3552
3
3430
3403
Avg
3411
500
1000
2000
5000
x
Sample 1 0.01mm
46.66
42.40
40.88
38.13
34.11
x
Sample 2
4580
4192
4038
3822
3484
Rut Depth
0.01mm Average
(mm)
45.8
46.23
0.00
41.92
42.16
4.07
40.38
40.63
5.60
38.22
38.18
8.05
34.84
34.48
11.75
69
APPENDIX F
Mixing Specimens for Marshall Test
Prepared Marshall Compacted Sample
Compacted Samples Ready for Stability and Flow Test
70
The Bowl and Sample Placed in the Water for TMD Test
Mixing Sample for Wheel Tracking Test
Wheel Tracking Sample Place into the Steel Mould
71
Wheel Tracking Sample after Test
Rut Depth after 5000 Passes on Wheel Tracking Machine