i
INVESTIGATIONS OF FACTORS AFFECTING PENDULUM TEST VALUE
ON ASPHALTIC CONCRETE SURFACES
SUDESH NAIR A/L BASKARA
A project report submitted in partial fulfillment of the
requirements for the award of the degree of
Master ofCivil Engineering (Transportation and Highway)
Faculty of Civil Engineering
Universiti Teknologi Malaysia
JUNE 2009
iii
Dedicated To Highway Engineering
Relevant Parties…
iv
ACKNOWLEDGEMENT
Firstly, I am thankful to God for completing this report successfully. I would also
like to thank my supervisors, Dr Haryati Yaacob and Associate Professor Dr Mohd Rosli
Hainin for their guidance, advice and continuous support throughout the course of this
research. Their kindness and encouragement made me always active and confident.
I would also like to thank highway technicians En Suhaimi, En Rahman, En
Azman and En Ahmad for their guidance on handling and operating the laboratory
equipments.
Thank to my colleague, Dorina Anak Astana for her assistance in helping me on
my laboratory works. I also would like to express my gratitude to my parents for their
encouragement and support. Their views and tips are useful indeed.
Finally, I hope that this report will be beneficial in the future.
v
ABSTRACT
Skidding is one of the major contributions to road accidents during wet weather
condition. Therefore, a study is conducted to investigate the factors affecting Pendulum
Test Value on Asphaltic Concrete surfaces. The main objective of this study is to
determine the mix type and the crossfall percentage that best resist skid during wet
weather condition. Three different types of dense graded mixes were used in this study
which are AC10, AC14 and AC20. Those three mixes are tested using Sand Patch Test
(SPT) and are then subjected to various rainfall conditions and crossfall percentages
using Rainfall Simulator. The rainfall conditions are categorized as low rainfall, medium
rainfall and high rainfall while the crossfalls were increased 2% from 0% to 10%
crossfalls. During the event of rainfall on each mix surfaces, a Portable Skid Resistance
Tester is used on the mix to obtain the Pendulum Test Value (PTV) at different
crossfalls. Results are analyzed using analysis of variance (ANOVA) to justify the
objectives. Results from PTV shows that 4% to 10% crossfall and AC20 is the best
crossfall and surface type in resisting skid.
vi
ABSTRAK
Gelinciran merupakan salah satu penyumbang utama kepada kemalangan jalan
raya terutamanya ketika hujan. Oleh itu, kajian ini dijalankan untuk menyelidik faktorfaktor yang mempengaruhi rintangan gelinciran di atas permukaan konkrit berasfal.
Objektif utama kajian ini ialah untuk menentukan jenis campuran dan peratus sendengan
jalan yang terbaik untuk menghalang dari berlakunya gelinciran ketika keadaaan hujan.
Tiga jenis campuran konkrit berasfal digunakan dalam kajian ini iaitu AC10, AC14 dan
AC20. Ketiga-tiga campuran tersebut diuji menggunakan ujian tampalan pasir dan
kemudiannya dikenakan keadaan hujan dan peratus sendengan jalan yang berlainan
dengan menggunakan alat simulasi hujan. Keadaan hujan yang dikenakan adalah hujan
renyai, hujan sederhana dan hujan lebat manakala sendengan jalan ialah dari 0% hingga
10% dengan kenaikan 2%. Ketika hujan dikenakan ke atas permukaan campuran konkrit
berasfal. Alat rintangan gelinciran diletakkan di atas campuran konkrit berasfal untuk
mendapatkan bacaan rintangan gelinciran pada sendengan jalan berlainan. Keputusan
akan dianalisa menggunakan Analysis of Variance (ANOVA) untuk mengesahkan
objektif tersebut. Keputusan nilai rintangan gelinciran menunjukkan bahawa sendengan
jalan 4% hingga 10% dan AC20 ialah sendengan jalan dan jenis permukaan yang terbaik
untuk menghalang gelinciran.
vii
TABLE OF CONTENTS
CHAPTER
1
2
TITLE
PAGE
DECLARATION
ii
DEDICATION
iii
ACKNOWLEDGEMENTS
iv
ABSTRACT
v
ABSTRAK
vi
TABLE OF CONTENTS
vii
LIST OF TABLES
xi
LIST OF FIGURES
xiii
LIST OF SYMBOLS AND ABBREVIATIONS
xv
LIST OF APPENDICES
xvi
INTRODUCTION
1
1.1
General
1
1.2
Problem Statement
2
1.3
Objective of Study
4
1.4
Scope of Study
4
1.5
Importance of study
5
LITERATURE REVIEW
6
2.1
6
Introduction
viii
2.2
Rainfall Intensity
6
2.3
Surface Runoff
8
2.4
Hot Mix Asphalt Types
9
2.4.1 Dense-Graded Mixes
10
2.4.2 Asphalt Characteristics
11
Aggregate
11
2.5.1 Aggregate Physical Properties
12
2.5.2 Maximum Aggregate Size
13
2.5.3 Gradation
13
2.5.3.1 Gradation Design for Dense Graded Mix
15
2.5.4 Other Properties
16
2.6
Crossfall
16
2.7
Skidding
18
2.8
Pavement Surface Characteristics
20
2.9
The Skidding Factors
22
2.9.1 The Texture of The Road Surface
22
2.9.2 The Aggregate at the Road Surface
24
2.9.3 Condition of the road surface
24
2.9.4 Weather condition
25
2.10
Water Film Thickness
26
2.11
Rainfall Simulator
30
2.12
Portable Skid Tester
31
2.5
3
METHODOLOGY
32
3.1
Introduction
32
3.2
Determination of Rainfall Intensity and Flowrate
35
3.3
Material Selection
36
3.4
Sieve Analysis
36
3.4.1 Dry Sieve
37
3.5
Grade Selection
38
3.6
Blending of Stockpile Specimen
38
ix
4
3.7
Optimum Bitumen Content
39
3.8
Mixing of Specimen
39
3.9
Compaction
40
3.10
Pavement Mix
41
3.11
Rainfall Simulator Test
41
3.12
Sand Patch Method
42
3.13
Pendulum Test Value
44
3.14
Runoff Flowrate and Water Film Thickness
46
3.15
Core Specific Gravity and Degree of Compaction
46
3.16
Analysis of Variance (ANOVA)
48
DATA AND ANALYSIS
49
4.1
Introduction
49
4.2
Sieve Analysis
50
4.3
Wash Sieve
52
4.4
Penetration and Softening Point Test
52
4.5
Optimum Bitumen Content
52
4.6
Determination of Mass for Rainfall Simulator Mould
53
4.7
Determination of Rainfall Using MASMA
53
4.8
Determination of Pavement Crossfall
54
4.9
Polished Stone Value
54
4.10
Sand Patch Test
55
4.11
Pendulum Test Value
57
4.11.1 The Effect of Rainfall Intensity on PTV
59
4.11.2 The Effect of Crossfall on PTV
63
4.11.3 The Effect of Surface Texture on PTV
65
4.12
Runoff and Water Film Thickness Data
70
4.13
Degree of Compaction and Specific Gravity
71
4.14
Analysis of Variance (ANOVA) Results
73
x
5
CONCLUSION AND RECOMMENDATIONS
77
5.1
Conclusion
77
5.2
Recommendation
78
REFERENCES
80
APPENDICES
84
xi
LIST OF TABLES
NO
TITLE
PAGE
2.1
Annual Recurrence Interval For Type Of Development
7
2.2
Coefficient of Runoff (ASCE, 1960)
9
2.3
Gradation Limit For Asphaltic Concrete (JKR/SPJ/rev2005)
15
2.4
Normal Pavement Cross Slopes (AASHTO, 1990)
18
2.5
Suggested Minimum Values Of Skid Resistance Numbers
20
2.6
Correlation of Macrotexture and British Pendulum Number
22
2.7
Influence of Texture on Variables
23
3.1
Gradation Limit For Asphaltic Concrete (JKR, 2005)
38
4.1
OBC, Gmb and TMD for 100 gyrations at 4% air void
53
4.2
Polishing Stone Value Results
55
4.3
Sand Patch Test Results for AC10
56
4.4
Sand Patch Test Results for AC14
56
4.5
Sand Patch Test Results for AC20
57
4.6
Pendulum Test Value for AC10, AC14 and AC20
58
4.7
Water Film Thickness and The Runoff of AC10, AC14 and AC20 71
4.8
The Specific Gravity and the Degree of Compaction of AC10,
AC14,AC20
72
4.9
Permeability Test Result of AC20
73
4.10
ANOVA of Rainfall Intensity on Crossfall
74
xii
4.11
ANOVA of Rainfall Intensity on Surface Type
74
4.12
ANOVA of Crossfall on Rainfall Intensity
75
4.13
ANOVA of Crossfall on Surface Type
75
4.14
ANOVA of Surface Type on Rainfall Intensity
76
4.15
ANOVA of Surface Type on Crossfall
76
xiii
LIST OF FIGURES
FIGURE
TITLE
PAGE
1.1
Statistics Road Accidents in Malaysia Year 1974 to Year 2005
3
2.1
Dense Grade Mix
10
2.2
Asphaltic Concrete Surface Texture (Rebecca, 2004)
10
2.3
Separation of aggregates into different sizes
12
2.4
Typical two-lane highway with linear cross slopes
17
2.5
Microtexture and Macrotexture
21
2.6
Schematic of a Rolling Tyre on a Wet Road Surface
25
2.7
Water Film Thickness, Mean Texture Depth, and Total Flow
26
2.8
Water Film Depth vs Surface Texture Depth
27
2.9
Water Film Depth vs Slope
28
2.10
Water Film Depth vs Rainfall Intensity
28
2.11
Microtexture vs Film Thickness
29
2.12. Macrotexture vs Film Thickness
30
2.13
Rainfall Simulator
31
2.14
Portable Skid Resistance Tester
31
3.1
Flow of Laboratory Testing
33
3.2
MASMA IDF Curve for Kuala Lumpur (MASMA)
35
3.3
Hot Mixing of Aggregate and Bitumen
40
3.4
Steel Roller Compactor
40
xiv
3.5:
Pavement Mix
41
3.6
Sand Patch Test on Pavement Mixes
44
3.7
Cored Samples for Different Mixes
47
4.1
Gradation for AC10 (Elizabeth, 2006)
50
4.2
Gradation for AC14 (Elizabeth, 2006)
51
4.3
Gradation for AC20 (Elizabeth, 2006)
51
4.4
Pendulum Test Value vs Intensity at 0% crossfall
60
4.5
Pendulum Test Value vs Intensity at 2% crossfall
60
4.6
Pendulum Test Value vs Intensity at 4% crossfall
61
4.7
Pendulum Test Value vs Intensity at 6% crossfall
61
4.8
Pendulum Test Value vs Intensity at 8% crossfall
62
4.9
Pendulum Test Value vs Intensity at 10% crossfall
62
4.10
Pendulum Test Value vs Crossfall for AC10
64
4.11
Pendulum Test Value vs Crossfall for AC14
64
4.12
Pendulum Test Value vs Crossfall for AC20
65
4.13
Pendulum Test Value vs Texture Depth at 0% Crossfall
67
4.14
Pendulum Test Value vs Texture Depth at 2% Crossfall
67
4.15
Pendulum Test Value vs Texture Depth at 4% Crossfall
68
4.16
Pendulum Test Value vs Texture Depth at 6% Crossfall
68
4.17
Pendulum Test Value vs Texture Depth at 8% Crossfall
69
4.18
Pendulum Test Value vs Texture Depth at 10% Crossfall
69
xv
LIST OF SYMBOLS AND ABBREVIATIONS
AC
Asphaltic Concrete
NAPA
National Asphalt Pavement Association
JKR
Jabatan Kerja Raya
ATJ
Arahan Teknik Jalan
ASTM
American Society for Testing and Materials
AASTHO
American Association of State Highway and Transportation
Officials
UTM
Universiti Teknologi Malaysia
OBC
Optimum Bitumen Content
MRP
Malaysia Rock Product
SPT
Sand Patch Test
BPT
British Pendulum Tester
BPN
British Pendulum Number
PTV
Pendulum Test Value
SRV
Skid Resistance Value
WFT
Water Film Thickness
ANOVA
Analysis of Variance
xvi
LIST OF APPENDICES
APPENDIX
TITLE
PAGE
1
Sieve Analysis Gradation
84
2
Wash Sieve
85
3
Penetration and Softening Test
86
4
Mass for Rainfall Simulator Mould
87
5
Rainfall Design and Rational Method
91
6
Pavement Crossfall
94
7
Sand Density and Volume Calculation
95
8
Pendulum Test Value
96
9
Analysis Of Variance
103
10
Pictures of Laboratory Testing
127
1
CHAPTER I
INTRODUCTION
1.2
General
Skidding is one of the most common contributors to accidents. Skidding is more
effective to happen during wet pavement condition. Apart from that, skidding also can
happen due to the insufficient of pavement crossfall and pavement surface
characteristics. Skidding can be avoided if there is a good friction between tyre and
pavement.
Pavement surface characteristics are important for both the safety and comfort of
drivers. Pavement surfaces should provide adequate friction and maintain a good level of
ride quality and to ensure the satisfaction of the driving. Both the macrotexture and
microtexture plays an important role in the friction characteristics of pavement surfaces.
2
The frequency of rainfall based on rainfall intensity also contributes in pavement
friction. The frequency of rainfall develops water film on pavement which exists
between the tyre-pavement contact. This could eventually creates hydroplaning thus
contributes to skidding.
Crossfall is another important element in providing good friction between tyre
and pavement. Crossfall is essential in highway construction since it functions to reduce
water on pavement during rainy weather.
1.2
Problem Statement
Malaysia is currently having one of the best road systems in Asia. Eventhough
Malaysia has the best road system; the accident rate is Malaysia increased at an average
rate 9.7% yearly (Royal Malaysian Police, 2005). Figure 1.1 below shows that road
accidents had increased from 24,581 cases in 1974 to 328,264 cases in 2005.
3
Figure 1.1: Statistics Road Accidents in Malaysia Year 1974 to Year 2005
(Royal Malaysian Police, 2005).
Accidents is high in Malaysia due to many factors such as the driver
carelessness, vehicle speed, braking distance, insufficient head distance and skidding
especially on wet condition.
One major factor contributes to road accident is skidding as there is connection
between vehicle tyre and pavement. The worst skidding ever happen is during wet
pavement condition during rainy day. Besides that, a research shows that skidding
contributes to 25% of wet road accidents in United Kingdom (Kennedy et al., 1990).
As Malaysia utilizes more Asphaltic Concrete pavements and Malaysia is located
in Khatulistiwa climate region, the skidding rate on wet pavement condition has to be
determined. Therefore a lab test will be carried out to investigate the factors affecting the
Pendulum Test Value of Asphaltic Concrete surfaces.
4
1.3
Objective of Study
The objectives of this study are:
·
to investigate the effect Pendulum Test Value on various rainfall intensities,
various crossfalls and on different Asphaltic Concrete surfaces during
rainfall.
·
to recommend the crossfall and pavement type that best resist skid during wet
pavement condition.
1.4
Scope of Study
This study is carried out at Makmal Jalanraya, Universiti Teknologi Malaysia
(UTM). Rainfall Simulator is used to simulate various type of rainfall on pavement.
Three asphaltic concrete pavement type being used in this study which are AC10, AC14
and AC20. Crushed aggregates from a quarry in Ulu Choh and bitumen of 80/100 Pen is
used for the design mix. Sand Patch Test is carried out on every mix samples. Besides
that, the Pendulum Test Value is determined by using Portable Skid Tester. Data from
Manual Saliran Mesra Alam (MASMA, 2000) is used to calculate rainfall intensities and
Jabatan Kerja Raya (JKR, 2005) specification is used in the mix design for this study.
5
1.5
Importance of study
The importance of this study is mainly to propose the best crossfall and the best
asphaltic concrete mix type to resist skid by rainfall simulation before constructing road
on a particular area. Other than that, by conducting this study, the relationship of
crossfall, rainfall intensity and pavement texture can be observed and determined.
Besides that, this study can be highlighted as a proposal to JKR and local authorities.
Overall, this study is aimed to give a new idea and concept towards the road
development in Malaysia.
6
CHAPTER II
LITERATURE REVIEW
2.2
Introduction
There are four major aspect need to be taken into consideration before
conducting the study. The four main aspects are the rainfall intensity, the crossfall, the
surface type and the water film thickness. Those four aspects are interrelated between
each other in evaluating skidding by PTV during rainy weather. In order to obtain a
basic understanding on all those aspects, reviews on literature is essential.
2.2
Rainfall Intensity
The design rainfall intensity is based on the Average Recurrence Interval (ARI)
and the time of concentration. According to Manual Saliran Mesra Alam (MASMA)
7
guideline, the polynomial expression as shown in Equation 2.1 is used to calculate
rainfall intensity of less than 30min while the Intensity Duration Frequency (IDF) curve
is used to calculate tainfall intensity of more than 30min. Based on MASMA (2000), the
annual Recurrence Interval for type of development is shown in Table 2.1.
ln(It) = a + b (ln t) + c (ln t)2 + d (ln t)3
(Equation 2.1)
Table 2.1: Annual Recurrence Interval For Type Of Development (MASMA, 2000)
Average Recurrence Interval
Type of Development
Open Spaces, Parks and Agriculture
Land in urban areas
Quantity
Minor System
Major System
1
Up to 100
Quality
Residential
·
Low Density
2
Up to 100
·
Medium Density
5
Up to 100
·
High Density
10
Up to 100
5
Up to 100
10
Up to 100
Commercial, Business and
Industrial – other than CBD
Commercial, Business, Industrial in
Central Business District (CBD)
areas of Large Cities
3 month ARI
8
2.3
Surface Runoff
Surface Runoff is determined through the amount of excessive precipitation. This
includes variety of ways by which water moves across the land. The surface runoff on
pavements usually removes all water that present on pavement to drainage system.
According to MASMA, the Rational Method is the common method used to calculate
peak runoff from rainfall intensity. This method is essential to estimate the peak flowrate
at an area of less than 80 hectares. The surface runoff is calculated based on Rational
Method formula in Equation 2.2. The coefficient of runoff obtained from American
Society of Civil Engineering (ASCE, 1960) is shown in Table 2.2.
Q=
CIA
(Equation 2.2)
360
Where
Q = peak flow (m3/s)
C = coefficient of runoff
I = average rainfall intensity(mm/hr)
A = drainage area (ha)
9
Table 2.2: Coefficient of Runoff (ASCE, 1960)
Type of Drainage Area
Runoff Coefficient
Railroad yard areas
0.20 – 0.40
Unimproved areas
0.10 – 0.30
Lawns
·
Sandy soil, flat, 2%
0.05 – 0.10
·
Sandy soil, average, 2-7%
0.10 – 0.15
·
Sandy soil, steep, 7%
0.15 – 0.20
·
Heavy soil, flat, 2%
0.13 – 0.17
·
Heavy sil, average, 2-7%
0.18 – 0.22
·
Heavy soil, steep, 7%
0.25 – 0.35
·
Asphaltic
0.70 – 0.95
·
Concrete
0.80 – 0.95
·
Bricks
0.70 – 0.85
Streets
Drives and walks
0.75 – 0.85
Roofs
0.75 – 0.95
2.4
Hot Mix Asphalt Types
Hot Mix Asphalt (HMA) is the most common flexible pavement and it is
distinguished by its design and production methods which include dense-graded mixes
as well as stone matrix asphalt (SMA) and open-graded mixes. HMA is known as hot
mix, asphalt concrete (AC), asphalt or bitumen. HMA can be produced through various
types and grades. It can be determine from the total relative of:
i.
coarse aggregate (retained sieve 2.36mm)
ii.
fine aggregate (passed sieve 2.36mm)
iii.
filler (passed sieve 75 µm)
10
2.4.1 Dense-Graded Mixes
A dense-graded mix is a well-graded HMA and it is relatively impermeable.
They can be classified as fine-graded or coarse-graded. Fine-graded mixes have more
fine and sand sized particles compared to coarse-graded mixes. This mix can suit all
pavement layers and all traffic conditions. Dense graded mix involves a combination of
well graded aggregate and asphalt binder. Figure 2.1 and Figure 2.2 shows the mix
structure and the texture respectively.
Figure 2.1:
Dense Grade Mix
Figure 2.2: Asphaltic Concrete Surface Texture (Rebecca, 2004)
11
2.4.2 Asphalt Characteristics
Asphalt is a strong cementations material, sticking agent, long lasting, and high
in water repellent. Besides that, it becomes plastic when mixed with aggregate. Asphalt
has colloidal feature which is solid at normal temperature and becomes liquid on higher
temperature level.
2.5
Aggregate
Aggregate is a collective term for mineral materials such as sand, gravel and
crushed stone which is used with a binding medium such as water, bitumen, Portland
cement and lime to form compound materials such as asphalt concrete and Portland
cement concrete. (Roberts et.al., 1996)
The constituent of aggregate in HMA generally covers up to 92 to 96 percent.
Aggregates can be natural or manufactured. Natural aggregates are extracted from
larger rock formations through quarry. Manufactured aggregate is obtained from the
byproduct of manufacturing industries such as slag (byproduct of the metallurgical
processing – typically produced from processing steel, tin and copper) (Roberts
et.al.,1996)
12
2.5.1 Aggregate Physical Properties
The pavement industry typically relies on physical properties for performance
characterization. (Roberts et.al.,1996)
Commonly measured physical aggregate properties are:
·
Gradation and size
·
Toughness and abrasion resistance
·
Durability and soundness
·
Particle shape and surface texture
·
Specific gravity
Figure 2.3:
Separation of aggregates into different sizes
13
2.5.2 Maximum Aggregate Size
Maximum aggregate size can affect HMA and base courses in several ways.
Excessive small maximum sizes may form instability in HMA mix while excessive large
maximum sizes contribute to poor workability and segregation. (Roberts et.al.,1996)
Definition of the maximum aggregate size in one of two ways:
·
Maximum size. The smallest sieve through which 100 percent of the aggregate
sample particles pass
·
Nominal maximum size. The largest sieve that retains some of the aggregate
particles but generally not more than 10 percent by weight
It is important to specify whether "maximum size" or "nominal maximum size"
is being referenced.
2.5.3 Gradation
Gradation in HMA helps to determine important property including stiffness,
stability, durability, permeability, workability, fatigue resistance, frictional resistance
and resistance to moisture damage. Therefore, gradation must be a primary concern in
HMA mix design. Gradation is usually measured using a sieve analysis to determine the
14
type of gradation such as dense graded, gap graded, open graded and uniformly graded.
(Roberts et.al.,1996)
·
Fine aggregate
Defined by AASHTO M 147 as natural or crushed sand passing the No. 10 sieve
and mineral particles passing the No. 200 sieve.
·
Coarse aggregate.
Defined by AASHTO M 147 as hard, durable particles or fragments of stone,
gravel or slag retained on the No. 10 sieve. Usually coarse aggregate has a
toughness and abrasion resistance requirement.
·
Mineral filler.
Defined by the Asphalt Institute as a finely divided mineral product at least 65
percent of which will pass through a No. 200 sieve. Pulverized limestone is the
most commonly manufactured mineral filler, although other stone dust, silica,
hydrated lime, Portland cement and certain natural deposits of finely divided
mineral matter are also used.
15
2.5.3.1 Gradation Design for Dense Graded Mix (Asphaltic Concrete)
The gradation of asphaltic concrete mixes is shown in Table 2.3.
Table 2.3: Gradation Limit For Asphaltic Concrete (JKR/SPJ/rev2005)
Mix Type
Wearing Course
Wearing Course
Wearing Course
Mix Designation
AC 10
AC 14
AC 20
BS Sieve Size, mm
Percentage Passing ( by weight)
100
28.0
20.0
100
76-100
14.0
100
90-100
64-89
10.0
90-100
76-86
56-81
5.0
58-72
50-62
46-71
3.35
48-64
40-54
32-58
1.18
22-40
18-34
20-42
0.425
12-26
12-24
12-28
0.150
6-14
6-14
6-16
0.075
4-8
4-8
4-8
Note: AC20 Gradation obtained from (JKR/SPJ/1988)
16
2.5.4 Other Properties
According to Roberts et.al.(1996), other important aggregate physical properties are:
i.
Toughness and abrasion resistance. Aggregates should be hard and tough enough
to resist crushing, degradation and disintegration from activities such as
manufacturing, stockpiling, placing and compaction.
ii. Durability and soundness. Aggregates must be resistant to breakdown and
disintegration from weathering (wetting/drying) or else they may break apart and
cause premature pavement distress.
iii. Particle shape and surface texture. Particle shape and surface texture are
important for proper compaction, load resistance and workability. Generally,
cubic angular-shaped particles with a rough surface texture are best.
iv. Specific gravity. Aggregate specific gravity is useful in making weight-volume
conversions and in calculating the void content in compacted HMA
2.6
Crossfall
The American Association of State Highway and Transportation Officials
(AASHTO, 1990) recommends a cross slope of 1.5% to 2% for high type pavement,
1.5% to 3% for intermediate type pavement and 2% - 6% for low type pavements.
However, crossfalls recommended by Arahan Teknik Jalan (8/86) are 2.5% for high type
pavement, 2.5% – 3.5% for intermediate type pavement, 2.5% - 6.0% for low type
17
pavements. The cross slopes depends generally on the type of pavement used and the
rainfall intensity at the area.
AASHTO states that the high type pavement is classified as the wearing surface
that can withstand high-volume, heavy vehicles and high speed for a long period of time.
The intermediate type pavement is similar to the high type pavement except they are
constructed with less strict standard and rules. The low type pavement utilizes low cost
roads such as surface treated earth and stabilized material. Figure 2.4 illustrates the
typical two lane highway with linear cross slopes. (Garrabrant. R, 2004).
Figure 2.4: Typical two-lane highway with linear cross slopes. (Garrabrant. R, 2004)
Table 2.4 indicates an acceptable range of cross slopes. The cross slopes ranges
from steep cross slopes for drainage and relatively flat cross slopes for driver comfort
and safety. Cross slopes of 2 percent have little effect on driver effort in steering but in
areas of high rainfall intensity, a steeper cross slope can be used to facilitate drainage.
(AASHTO, 1990)
18
Table 2.4: Normal Pavement Cross Slopes (AASHTO, 1990)
Surface Type
Range in Rate Surface Slope
High-type surface
·
2 lanes
0.015 – 0.020
·
3 or more lanes, each direction
0.015 – 0.040
Intermediate surface
0.015 – 0.03
Low-type surface
0.020 – 0.060
Shoulders
·
Bituminous or concrete
·
With curbs
0.020 – 0.060
> 0.040
Equation 2.3 shows that the slope formula which is the percentage of differences
in height to differences in horizontal. (Brouwer et.al., 1985).
Slope in %
=
Height differences .
(Equation 2.3)
Horizontal differences
2.7
Skidding
Skid occurs when the available friction between the pavement surface and the
tire is insufficient to respond to the maneuver a driver is attempting to make. This
happens when the vehicle is maneuvered by the exerted forces at that stationary area
where the friction between the tire and the road surface opposes to the maneuvering
force. A dry road surface condition produces high road and tyre friction for normal
19
driving maneuvers while the wet road surfaces decreases the friction effect significantly
causing skidding accidents. (Kokkalis et.al. 2002)
Skid resistance is the force developed when a tire that is prevented from rotating
slides along the pavement surface. Skid resistance controls contact between the tire
rubber and the pavement surface.
Skid resistance is an important pavement evaluation parameter because
inadequate skid resistance will leads to higher skid related accidents besides road user
often rely on safe pavement. Skid resistance changes over time. Skid resistance is also
typically lower in wet condition and higher in dry condition. Skid resistance is generally
quantified using friction factor, skid number or British Pendulum Number. (Nor
Zolhanita, 2007).
Table 2.5 shows the minimum Skid Resistance Value measured with British
Pendulum Tester. (British Pendulum Manual, 2000)
20
Table 2.5: Suggested Min Skid Resistance Value (British Pendulum Manual, 2000)
Minimum
Category
Type of site
SRV
(Surface wet)
A
Difficult sites such as:
i.
Roundabouts
ii. Bends with radius less than 150m or unrestricted
65
iii. Gradients, 1 in 20 or steepr of length greater than 100m
iv. Approaches to traffic lights on unrestricted roads
B
Motorways, trunk and class 1 roads and heavily trafficked roads in
urban area
C
2.8
All other sites
55
45
Pavement Surface Characteristics
Pavement surface characteristics are important for both safety and comfort of
drivers. Pavement surfaces should provide adequate friction and maintain a good level of
ride quality.
The combination of good friction and low levels of roughness are
important in the design of a pavement wearing surface. (Flintsch et. al., 2001)
The surface texture of a pavement wearing surface is one of the primary
contributors to tire-pavement friction. The surface texture can be divided into two which
are macrotexture and microtexture as shown in Figure 2.5. Pavement texture can
generates energy losses in the tyre when it is skidding. In this process, some of the
energy used to deform the tyre tread during sliding on the texture is absorbed back by
the tyre to returns to its original shape. In wet conditions, as vehicle speeds increase, the
21
possibility of skidding is high. The extent to which this occurs depends upon the texture
depth. (Flintsch et. al., 2001)
Good macrotexture pavement allows for the rapid drainage of water from the
pavement that improves the contact between the tire and the pavement surface. The
macrotexture of a pavement surface results from the large aggregate particles in the
mixture. Common macrotexture measurement methods include the sand patch method,
the outflow meter, and the circular texture meter. (Flintsch et. al., 2001)
Microtexture provides direct tire pavement contact and contributes to tyre
pavement friction and it is the major contributing factor to skidding resistance at low
speeds. Microtexture is gradually polished away by heavy traffic and gradually
decreases the skidding resistance. (Flintsch et. al., 2001)
Figure 2.5: Microtexture and Macrotexture (Flintsch et. al., 2001)
22
2.9
The Skidding Factors
2.9.1 The Texture of The Road Surface
There are two texture type of road surface influencing skidding which are
macrotexture and microtexture. Macrotexture is required to the eliminate water from the
contact area between the tyre and the road surface especially at higher vehicle speeds.
The macro texture is determined by the size of the aggregate particles at the road
surface. (Molenaar et al, 2004)
The micro texture is determined by the roughness and angularity of the surface of
the aggregate particles. The micro texture ensures high contact pressures between the
aggregate and the tyre. (Molenaar et al, 2004)
A study by Peter Cairney et.al. (2006) on four sites along Princes Highway West
proves that a bigger macrotexture depth creates a bigger British Pendulum Number
(BPN) while a smaller mactotexture depth generates smaller British Pendulum Number.
The surface characteristics of the study shown in Table 2.6.
Table 2.6: Correlation of Macrotexture and BPN (Peter Cairney et al, 2006)
Site
Macrotexture
British Pendulum Number
Site 1
> 0.7mm
68
Site 2
< 0.4mm
64
Site 3
< 0.4mm
50
Site 4
> 0.7mm
74
23
Table 2.7 shows a study by Sandberg (1997) indicating the influence of
macrotexture and microtexture on different variables. The influence of microtexture on
friction is high and the influence of marotexture on water runoff is high.
Table 2.7: Influence of Texture on Variables (Sandberg, 1997)
Effect on Vehicle, Driver or
Road Surface Characteristics
Environment
of Importance
Friction
Macrotexture
High
Megatexture
Moderate
Microtexture
Very High
Rolling Resistance/Fuel
Macrotexture
High
Consumption/Air Pollution
Megatexture
Very High
Uneveness
High
Macrotexture
Moderate
Microtexture
Very High
Macrotexture
Very High
Megatexture
Very High
Water Runoff
Macrotexture
High
Splash and Spray
Macrotexture
High
Light Reflector
Macrotexture
High
Microtexture
Little
Macrotexture
High
Megatexture
Very High
Uneveness
High
Tire Wear
Exterior Noise
Interior Noise
Magnitude of the Influence
24
2.9.2 The Aggregate at the Road Surface
The aggregate at the road surfaces are relevant with respect to the skidding resistance:
·
the shape and the size
·
the resistance against polishing
·
the resistance against crushing
·
the bond with the binding agent (bitumen or cement).
In general crushed aggregates have sharp and rough surfaces that favor the
texture and the skidding resistance. However, polishing of the aggregates occurs under
the repeated traffic loadings. The rate of polishing is dependent on the type of aggregate
and the traffic intensity. The skidding resistance is also negatively influenced by
crushing and loss of aggregate at the road surface. (Molenaar et al, 2004)
2.9.3 Condition of the road surface
The condition of wet road surfaces is interrupted by the contact between the tyre
and the road surface and that leads to a decrease of the friction coefficient. Therefore,
the rainwater has to be removed fast from the road surface so that the friction coefficient
can be maintained at acceptable level. The applied percentage of crossfall of 2.5% is
essential to remove rainwater on pavement. Dust and water acts as a lubricant on
pavement is another factor of causing skid. Continuous rainfall can causes cleaning of
the road surface and the friction coefficient is increased. (Molenaar et al, 2004)
25
2.9.4 Weather condition
The influence of rainfall on pavement decreases the skidding resistance thus
increases the risk for hydroplaning. Hydroplaning is a phenomenon where vehicle tyre
looses contact with the road surface. Figure 2.6 schematically shows how a rubber tyre
rolls or slides over a wet road surface. Three zones are distinguished: zone 1 (no
contact), zone 2 (local contact) and zone 3 (dry contact). When zone 1 enlarges the
friction coefficient decreases, finally resulting in aquaplaning. This effect is intensified if
a standing wheel is brought to rotation very quickly; this occurs at a landing aircraft
where the phenomenon of hydroplaning was observed first. (Molenaar et al, 2004)
Figure 2.6: Schematic Representation of a Rolling Tyre on a Wet Road Surface
(Molenaar et al, 2004)
26
2.10
Water Film Thickness
The mitigation of hydroplaning and skidding depend solely on water film
thickness on the pavement. The water film thickness depends on tendency water to drain
effectively.
According to David A. Noyce et.al. (2005), the water film thickness thickness is
the thickness of the water film present on top of the pavement surface as illustrated in
Figure 2.7. The flow path of the water on pavement is determined by the pavement
surface slope. Measures to reduce water film thickness on pavement surface are
alteration of surface geometry such as crossfall, use of permeable or porous asphalt
paving mixtures, grooving in Portland cement concrete, and enhancement of surface
texture through mixture selection and design.
Figure 2.7: Water Film Thickness, Mean Texture Depth, and Total Flow
27
A research by Chesterton et.al, (2006) on aquaplaning and water film thickness
towards slope, surface texture and rainfall intensity is done. This research proves that the
water film thickness decreases when the surface texture depth increase shown in Figure
2.8. The water film thickness too decreases when the slope increase as shown in Figure
2.9. Besides that, Figure 2.10 shows when the rainfall intensity increase, the water film
thickness increase too. However, this study suggested that the Gallaway (1979) method
is better than the Road Research Laboratory in determining the water film thickness.
Gallaway (1979) recommended that a maximum WFD of 4mm should be achieved.
Figure 2.8: Water Film Depth vs Surface Texture Depth (Chesterton et.al, 2006)
28
Figure 2.9: Water Film Depth vs Slope (Chesterton et.al, 2006)
Figure 2.10: Water Film Depth vs Rainfall Intensity (Chesterton et.al, 2006)
29
Another research by Stacy (2008) proves that the microtexture depth was
negative trend which means that the increase in British Pendulum Number creates a
lower water film thickness as shown in Figure 2.11 A lower water film thickness
increase the microtexture of the mix. The same trend is seen for macrotexture. An
increase in texture depth generates lower water film thickness. The relationship of
macrotexture and water film thickness is depicted in Figure 2.12.
Figure 2.11: Microtexture vs Film Thickness (Stacy, 2008)
30
Figure 2.12.: Macrotexture vs Film Thickness (Stacy, 2008)
2.11
Rainfall Simulator
Precipitation is hard to obtain since the study is to be carried out in laboratory.
Therefore, a Rainfall Simulator is developed to simulate the rain on pavement. Rainfall
Simulator is a laboratory apparatus consists of three main components which are rain
flowmeter to simulate rain, pavement mould to place pavement and adjustable gear to
provide gradient or slope for the pavement. The purpose of Rainfall Simulator is to
simulate rain on pavements. Rainfall simulator is as shown Figure 2.13 below.
31
Figure 2.13: Rainfall Simulator
2.12
Portable Skid Tester
Portable Skid Tester is also known as British Pendulum Tester is used to measure
the skidding characteristics on pavements. Guide on using the Portable Skid Tester is
given in BS EN 13036-4. The Portable Skid Tester is as shown in Figure 2.14 below.
Figure 2.14: Portable Skid Resistance Tester
32
CHAPTER III
METHODOLOGY
3.1
Introduction
This chapter focuses on the method and process to be carried out in order to achieve
the objective of this study. Procedures discussed in this chapter are according to ASTM
and AASHTO standard. All tests will be carried out in laboratory. There are six major
parts of laboratory test to be conducted:
1)
Design of Asphaltic Concrete mix
2)
Sand Patch Test
3)
Skidding test using Portable Skid Tester and Rainfall Simulator
4)
Measurement of Water Film Thickness
5)
Collection of runoff
6)
Degree of Compaction
33
Determination of Low Rainfall,
Medium Rainfall and
High Rainfall
Sieve Analysis Test
(AC10, AC14, AC20)
2 years , 20 years and 100 years
ARI
Obtain Optimum Bitumen Content
MASMA Rainfall
Calculation & IDF Curve
Design Asphaltic Concrete Mix
(AC10,AC14 & AC20)
Designed Rainfall Intensity
(mm/hr)
Construct Pavement Mixes
(AC10, AC14, AC20)
Designed Rainfall Flow Rate
(ℓ/min)
Sand Patch Test
Continue
Continue
34
Continued
Continued
Designed Flow Rate
(ℓ/min)
Constructed Pavement Mixes
(AC10, AC14, AC20)
Rainfall Simulator
Adjustment of Crossfall
Runoff Collector
Basin
Pavement
Portable Skid Tester
Obtain Q Runoff
Obtain Water Film
Thickness
Obtain Pendulum
Test Values
Data Analysis
Conclusion and Recommendation
Figure 3.1:
Flow of Laboratory Testing
35
3.2
Determination of Rainfall Intensity and Flowrate
The rainfall intensities in Malaysia needs to be justify before conducting this
study. Since the rainfall intensities varies from one year to another, the rainfall
intensity could not be determined. Therefore, Manual Saliran Mesra Alam
(MASMA) is used as guide to determine the low, medium and high rainfall intensity.
The MASMA IDF curve is used to determine low intensity rainfall while the
polynomial expression method is used for the medium and high rainfall intensities.
The MASMA IDF Curve is shown as in Figure 3.2 below. Average Recurrence
Interval (ARI) of two years is taken for low rainfall intensity while ARI of 20 years
and 100 years is taken for medium and high rainfall intensity.
Figure 3.2: MASMA IDF Curve for Kuala Lumpur (MASMA, 2000)
A simple rainfall calculation and Rational Method calculation is done to
obtain low rainfall intensity, medium rainfall intensity and high rainfall intensity
values. The designed rainfall intensities is then converted to rainfall flowrates. The
rainfall flowrates is then used as input in conducting the Rainfall Simulator.
36
3.3
Material Selection
Aggregates, bitumen and filler are used in this study where aggregates of
various sizes are obtained from a quarry in Ulu Choh, bituminous material of 80/100
PEN and Ordinary Portland Cement (OPC) used as filler. The aggregates need to
fulfill the Polished Stone Value Test requirement in order to achieve a good
resistance level of aggregate towards polishing while the bituminous material has to
be tested with Penetration and Softening Point Test to fulfill the requirement so that
an optimum mix can be acquired.
3.4
Sieve Analysis
Sieve analysis test is used to determine the aggregate sizes from a sample
taken from quarry. Through this sieve test, the proportion of coarse aggregates, fine
aggregate and filler can be determined. The limit of the gradation used varies
according to the function and purpose of the aggregate. The best aggregates sample
sizes is the sample which best suits the ‘envelope’ through analysis. From this sieve
analysis, the gradation obtained reflects the quality and the cost of the pavement.
Standard procedures for sieve analysis are given in ASTM C 136 and AASHTO T
27.
37
3.4.1 Dry Sieve
The apparatus that will be used for dry sieve analysis are:
i.
Sieve with various sizes;
ii.
Mechanical Sieve Shaker; and
iii.
Balance with the accuracy of 0.5g.
The procedures for dry sieve analysis are as below:
i.
The sieve container is arranged in order of larger size to
smaller size of opening.
ii.
Aggregate is placed on the top sieve and shaker is turned on to
start the sieving.
iii.
Aggregate that has been sieved will be separated according to
the size.
iv.
Total aggregate from different sizes as designed will be weigh
for mixing.
38
3.5
Grade Selection
Table 3.1: Gradation Limit For Asphaltic Concrete (JKR, 2005)
Mix Type
Wearing Course
Wearing Course
Wearing Course
Mix Designation
AC 10
AC 14
AC 20
BS Sieve Size, mm
Percentage Passing ( by weight)
100
28.0
20.0
3.6
100
76-100
14.0
100
90-100
64-89
10.0
90-100
76-86
56-81
5.0
58-72
50-62
46-71
3.35
48-64
40-54
32-58
1.18
22-40
18-34
20-42
0.425
12-26
12-24
12-28
0.150
6-14
6-14
6-16
0.075
4-8
4-8
4-8
Blending of Stockpile Specimen
The blending of stockpile specimen involves a combination of blending two
or more aggregate stockpile together. This method is done by proportioning the
aggregate in order to obtain the desired aggregate that suits the gradation limits. The
gradation limits for three mix types are specified in Table 3.1. The mixes to be
prepared includes aggregates, bitumen and OPC. The stockpiles used for this study
are from 20mm, 10mm and quarry dust. The aggregate from respective stockpile is
sieved and its passing weight is calculated to obtain the percentage of passing. Once
stockpiles aggregate passing weight obtained, the proportion are made until it enters
the gradation limits.
39
3.7
Optimum Bitumen Content
The Optimum Bitumen Content of AC10, AC14 and AC20 were obtained
from the Elizabeth (2006). OBC usually ranges from 4% to 7% The procedures for
obtaining the Optimum Bitumen Content for 100 superpave gyrations shall comply
to procedures based on Elizabeth (2006).
3.8
Mixing of Specimen
Firstly, the total aggregate weight to be used is determined. Then, the
aggregates samples are divided in four trays for every mix design and are place into
the oven one day before the mixing process. This is to ensure that the aggregate is
completely dry before the mixing process can take place. The temperature of the
oven should be adjusted to 110º.
Four hour before the sample mixing, the bitumen has to be heated in the oven
for melting process. The temperature must be around 130 ºC - 145ºC. Once the
mixing process take place, the aggregates from each tray is mixed with bitumen on a
wok. Figure 3.3 shows the hot mixing process. When the mix is ready, it is filled into
the rainfall simulator mould (118cm x 61cm x 6cm). The mixing process has to be
carried out gradually between the four aggregate trays in order to maintain the heat
of mix at 170 ºC while mixing and at 150 ºC for compaction besides providing a four
layer mix into the rainfall simulator mould. The mixing process stated above is
carried out for AC10, AC14 and AC20.
40
Figure 3.3: Hot Mixing of Aggregate and Bitumen
3.9
Compaction
The compaction is carried out on each layer of mix placed into the rainfall
simulator mould at 200 times compaction using a steel roller. A total of 800 times
compaction effort is done on the mix to ensure that the 4% of air void could be
achieved. The compaction is carried out at 150 º C. Figure 3.4 shows the compaction
process using steel roller.
Figure 3.4: Steel Roller Compactor
41
3.10
Pavement Mix
Three different pavement mixes were constructed which are AC10, AC14 ad
AC20 mixtures. Each mixes is placed on a 118cm x 61cm x 5cm mould. The mixes
are left to cool in mould for one day as shown in Figure 3.5 and are ready to be
placed into the Rainfall Simulator.
Figure 3.5: Pavement Mix
3.11
Rainfall Simulator Test
The sand patch test is carried out first before the rainfall simulation could
take place. Five sand patch tests is carried out for each mix. Sand Patch Method is
used to determine the texture depth of the constructed pavement mix. Once
completed, the pavement is placed into the rainfall simulator apparatus and the
Portable Skid Tester is placed on pavement to measure the Pendulum Test Value.
The PTV is measured at normal conditions first where the procedures of measuring
PTV follow the BS standard (BS13036, 2003) and no rainfall intensity is present.
Next, Then, the PTV is measured according to the low rainfall intensity, medium
42
rainfall and high intensity using adjustable valve and flowmeter. The Portable Skid
Tester is placed longitudinally with wheel path and against with the water runoff.
The procedures of conducting the rainfall simulator are as follows:
1.
Adjust the crossfall of the pavement ranging from 0% to 10% with the
increase of 2% at a time.
2.
Adjust the flowrate at the flowmeter using the valves according to the
rainfall intensity.
3.
Collect and measure the rainfall runoff collected.
4.
Measure the water film thickness.
5.
Calibrate the British Pendulum Tester
6.
Obtain the Pendulum Test Value at 5 swings
7.
Repeat Step 3 to Step 6 for different rainfall flowrate and at different
crossfall.
3.12
Sand Patch Method
Sand Patch test is used to determine texture depth of the road which is based
on BS EN 13036-1:2002. The test procedure involves spreading a known volume of
sand on a clean and dry pavement surface, measuring the area covered, and
calculating the average depth between the bottom of the pavement surface voids and
the tops of surface aggregate particles.
Firstly, make sure the pavement surface to be measured and select a dry,
homogeneous area that contains no unique, localized features such as cracks and
joints. Then, the surface is cleaned using the stiff wire brush first and subsequently
using the soft bristle brush to remove any residue, debris or loosely bonded aggregate
particles from the surface. Locate the portable windshield around the test area if
necessary.
43
Fill the cylinder of known volume with sand and gently tap the base of the
cylinder several times on a rigid surface. Add more sand to fill the cylinder to the
top, and level with a straightedge. Pour the measured volume or mass of sand on to
the surface. The material is spread carefully into a circular patch, with the disc tool,
rubber-covered side down, filling the surface voids flush with the aggregate particle
tips. Use a slight pressure on the hand, just enough to ensure that the disc will spread
out the sand so that the disc touches the surface aggregate particle tips.
The diameter of the circular area covered by the sand is measured and
recorded at a minimum of four equally spaced locations around the sample
circumference. The average of four diameters data is calculated and recorded. This
test procedure is repeated five times at the nearer location. The average of the
individual values shall be considered to be the average surface texture (macrotexture)
depth of the tested pavement surface. Figure 3.6 shows the sand patch test carried out
on pavement mixes.
The calculations involve for this test are calculation of the internal volume of
the sample cylinder and calculation of the mean texture depth.
Calculation of the internal volume of the sample cylinder:
V = πd2h / 4
Where:
V= internal cylinder volume, (mm3)
d = internal cylinder diameter, (mm)
h = cylinder height, (mm)
44
Calculation of the mean texture depth, MTD:
MTD = 4V /πD2
Where:
V = sample volume (internal cylinder volume), (mm3)
D = average diameter of the area covered by the material, (mm)
Figure 3.6: Sand Patch Test on Pavement Mixes
3.13
Pendulum Test Value
The Portable Pendulum Tester is a dynamic pendulum impact type tester
which is based on the energy loss occurring when a rubber slider edge is propelled
across a test surface. The values measured are referred to as British Pendulum
Numbers (BPN) for flat surfaces.
45
Firstly, the Portable Pendulum Tester is placed on a firm surface with the
pendulum swinging in the direction of traffic. Then, the leveling screw is adjusted to
make the pendulum support column in vertical. Then raise the axis of suspension of
the pendulum for the arm swings freely. Adjust the friction in the pointer mechanism
until the pointer arrive zero position on the test scale.
Adjust the height of the pendulum arm so that in traversing the surface the
rubber slider is in contact with it over the whole width of the slider and over the
length below. A pointer fixed to the foot of the slider assembly and a pre-marked
gauge shall be used. It must be careful not to disturb the slider from its set position.
The pointer and the pendulum are released from horizontal position by
holding the button. Catch the pendulum arm on the early portion of the return swing
and record the position of the pointer on the scale to the nearest whole number. The
pendulum and pointer returned to the release position by raising the slider using the
lifting handle. This procedure performs for five times and re-wetting the surface
again just before releasing the pendulum. Finally, the reading is recorded.
The PTV value as the mean of five swings is calculated using the formula:
PTV = ∑ ( v1 + v2 + v3 + v4 + v5 ) / 5
Where:
v1, v2,v3,v4,v5 = individual values for each swing
46
3.14
Runoff Flowrate and Water Film Thickness
The runoff flowrate and the water film thickness were obtained during the
rainfall event on the rainfall simulator. The water film thickness is measured on the
pavement where the water depth on the surface is measured using a ruler. The water
runoff from the pavement is recorded for 1 minute. The water runoff value should be
almost or closer to the flowrate of the rainfall since the air void of the asphaltic
concrete is 4% only. Water runoff from the pavement is collected into the runoff
collector basin and the volume is measured. The volume per minute is converted into
liter per minute as a comparison with the flowrate of rainfall.
3.15
Core Specific Gravity and Degree of Compaction
The specific gravity test is useful to determine the unit weight of compacted
mixes. After completion of rainfall test on the three sample mixtures, the samples are
then taken for coring. A total of five coring is done on each sample. The coring
samples are let to dry for a day before the specific gravity test could be carried out.
Bulk specific gravity is determined using the water displacement method.
The
specimens were weighed in three different conditions that is in air, in water, and
saturated surface dry.
The test is done to obtain the specific gravity of cored
samples. Degree of Compaction of the mixes is acquired by obtaining the percentage
of specific gravity of cored sample to the specific gravity of Superpave sample. The
method to conduct the specific gravity is in accordance with ASTM D 2726. The
cored samples for different mix is shown in Figure 3.7.
47
Below are the procedures for determining bulk specific gravity:
i.
Mass of specimen is immersed in water for 3 to 5 min and weigh. The
weight mass in immersion is recorded as C.
ii.
Mass of saturated surface dry specimen in air – surface dry is
recorded as B. The weight is taken after wiping the specimen surface
that has been immersed in water previously.
iii.
Mass of oven-dry specimen – compacted specimens is the sample that
is just left at room temperature. The mass is recorded as A.
The bulk specific gravity for the specimens is calculated using the following
equation:
Bulk Specific Gravity = A/ (B-C)
Where:
A = Weight of dry specimen in air
B = Weight of saturated surface dry specimen in air
C = Weight of saturated specimen in water
Figure 3.7: Cored Samples for Different Mixes
48
3.16
Analysis of Variance (ANOVA)
Analysis of variance approach shows that the value of two or more unknown
population means are likely to be different. In the case of analysis of variance, the
following assumptions must be true:
1.
The populations under the study have normal distributions.
2.
The samples are drawn randomly and each sample is independent of
the other samples.
3.
The populations from which the sample values are obtained all have
the same unknown population variance (б2). That is б12 = б22 =
б32…… бn2
The procedures of conducting the ANOVA test are:
1.
State the null and alternative hypotheses
2.
Select the Level of Significance
3.
Determine the test distribution to use
4.
Define the rejection or critical region
5.
State the decision rule
6.
Compute the test statistics
7.
Make the statistical decision
49
CHAPTER IV
DATA AND ANALYSIS
4.1
Introduction
Data is obtained from various guidelines before the results can be obtained.
Results of this study are obtained through the various tests method stated in
methodology. The results obtained are discussed in this chapter. Once the results are
obtained, the analysis is done to describe the results obtained.
Analysis of this study is focused on the relationship of three different factors
affecting skidding. The relationship of rainfall intensity, surface texture and crossfall
are obtained to evaluate the effect of those three factors on PTV. Besides that,
analysis of variance (ANOVA) is carried out to determine the best pavement type
and the best crossfall percentage to resist skid.
50
4.2
Sieve Analysis
The sieve analysis results were obtained from the Elizabeth (2006). Sieve
analysis results that are used in this study include the AC10, AC14 and AC20
gradation. The sieve analysis results are attached as in Appendix 1 and its gradation
charts are shown in Figure 4.1 to Figure 4.3.
Figure 4.1: Gradation for AC10 (Elizabeth, 2006)
51
Figure 4.2: Gradation for AC14 (Elizabeth, 2006)
Figure 4.3: Gradation for AC20 (Elizabeth, 2006)
52
4.3
Wash Sieve
The wash sieve is carried out to determine the amount of dust washed out
from the aggregates. The wash sieve results were obtained from Elizabeth (2006).
Wash sieve results of AC10, AC14 and AC20 were used in this study. The
percentage of dust is 4.5% for AC10, 3.8% for AC 14 and 3.5% for AC20. The wash
sieve results are attached in Appendix 2.
4.4
Penetration and Softening Point Test
Bitumen obtained from a quarry in Sedenak was tested for penetration and
softening point as soon as it arrives to the laboratory. Bitumen 80/100 PEN were
tested for penetration test to determine the level of sample hardness while softening
point test is done to obtain the temperature when the bitumen melts. The average
penetration of the samples is 82.8 PEN while the average softening point temperature
of the samples is 46.7 ºC. The penetration results met the JKR /SPJ/rev2005
requirement. Therefore, the bitumen can be used for mix. The results of both these
tests are attached in Appendix 3.
4.5
Optimum Bitumen Content
The Optimum Bitumen Content (OBC) was taken from Elizabeth (2006). The
OBC of 6.3% for AC10, 5.8% for AC14 and 4.2% for AC20 were obtained through
the 100 number of Superpave gyration which is compacted to 4% air void according
to NAPA method. The Obtimum Bitumen Data is shows below in Table 4.1
53
Table 4.1: OBC, Gmb and TMD for 100 gyrations at 4% air void (Elizabeth, 2006)
No of Gyration
100
4.6
Mix Type
OBC (%)
Gmb
TMD
AC10
6.3
2.278
2.391
AC14
5.8
2.297
2.400
AC20
4.2
2.352
2.443
Determination of Mass for Rainfall Simulator Mould
Three samples have to be prepared for the rainfall simulator apparatus which
are AC10, AC14 and AC20 respectively. The mass required for the mould rely on
TMD and Air Void content. A simple calculation is done and resulted to the
preparation of mass of 99.5kg for AC10, 100kg for AC14 and 101.5kg for AC20.
The mass calculation is attached in Appendix 4.
4.7
Determination of Rainfall Using MASMA
Three rainfall conditions were established for this study which are high
rainfall intensity, medium rainfall intensity and low rainfall intensity. The high
rainfall intensity of 432.40mm/hr at 100 years ARI, medium rainfall intensity of
220.50mm/hr at 20 years ARI and low rainfall intensity of 30mm/hr at 2 years ARI
were obtained through calculation as attached in Appendix 5. All those three rainfall
intensity in mm/hr are then converted into l/min so that to be used in rainfall
simulator apparatus. The conversion of mm/hr to ℓ/min is done using the Rational
Method of Q = C I A. The C value is 0.95 and the A value is 7.198 x 10-7 km2.
Therefore, the high flowrate is 4.93 ℓ/min while the medium is 2.52 ℓ/min and the
54
low flowrate is 0.34 ℓ/min. Apart from that, a normal condition represents as the
control where no rainfall intensity is present.
4.8
Determination of Pavement Crossfall
The pavement crossfall for this study is subjected from 0% crossfall to 10%
crossfall with the increment of 2% crossfall of every interval. The pavement crossfall
can be determined by adjusting the lever gear installed at the back of the rainfall
simulator apparatus. Pavement crossfall height and slope calculation is attached as in
Appendix 6.
4.9
Polished Stone Value (PSV)
The polished stone value test is conducted in order to determine the average
value when aggregates are in polished condition. The results of aggregates from Ulu
Choh quarry will provide the base value for this study which means that the value of
a new constructed pavement must be higher than the polished stone value. The
polishing stone value results are shown in Table 4.2.
55
Table 4.2: Polishing Stone Value Results
Average last
Average of four
3 readings
consecutive specimen
Specimen
Measured Value
1
52,51,50,50,50
50
2
51,50,50,50,49
50
3
53,52,50,49,48
49
4
50,50,49,49,48
49
5
52,51,50,50,49
50
6
53,52,50,49,49
49
7
51,51,50,50,49
50
8
52,51,51,50,50
50
9
52,51,50,50,50
50
10
52,51,51,50,50
50
11
50,49,49,49,49
49
12
52,51,50,50,49
50
13 (Control)
53,52,51,51,50
51
Control Value
14 (Control)
53,52,51,50,50
50
50.5
The PSV for the sample
4.10
PSV
50
50
50
50
50
50
50
Sand Patch Test
A total of five Sand Patch Tests is done for every sample. The average sand
patch test for AC10 is 0.68mm, for AC14 is 0.96mm and for AC20 is 1.26mm.The
sand patch test shows that bigger aggregate sizes has a bigger texture depth and the
smaller aggregate sizes has a smaller texture depth. This proves that the macrotexture
for AC20 is bigger and rougher than the macrotexture of AC14 followed by
macrotexture of AC10. The results for the sand patch test are shown in Table 4.3,
Table 4.4 and Table 4.5. The sand density and volume calculation is given in
Appendix 7.
56
Table 4.3: Sand Patch Test Results for AC10
DIAMETER (mm)
NO
TEXTURE
DEPTH
1
2
3
4
5
AVERAGE
1
200
220
200
200
210
206
0.73
2
230
220
210
220
215
219
0.64
3
200
200
200
210
225
207
0.72
4
210
220
225
215
225
219
0.64
5
220
210
225
210
225
218
0.65
AVERAGE
0.68
(mm)
Table 4.4: Sand Patch Test Results for AC14
DIAMETER (mm)
NO
TEXTURE
DEPTH
1
2
3
4
5
AVERAGE
1
160
165
165
155
165
162
1.17
2
170
170
170
170
170
170
1.07
3
185
195
205
190
200
195
0.81
4
180
200
195
180
200
191
0.84
5
170
185
190
180
190
191
0.92
AVERAGE
0.96
(mm)
57
Table 4.5: Sand Patch Test Results for AC20
DIAMETER (mm)
NO
4.11
TEXTURE
DEPTH
1
2
3
4
5
AVERAGE
1
160
165
150
160
160
159
1.22
2
155
150
150
150
150
151
1.35
3
150
150
155
155
150
152
1.33
4
160
165
160
170
155
162
1.17
5
165
150
150
165
165
159
1.22
AVERAGE
1.26
(mm)
Pendulum Test Value
The Pendulum Test Value is determined by three different independent
variables from the rainfall simulator. The variables are the Rainfall Intensity, the
Crossfall and the Surface Type. The Pendulum Test Value is obtained through three
different type of rainfall intensity which are the low rainfall intensity (30mmh/r),
medium rainfall intensity (220.5mm/hr) and high rainfall intensity (432.4mm/hr) and
the normal condition where no rainfall intensity is present and the procedures of
measuring PTV follow the BS standard (BS13036, 2003). Besides that, the crossfall
ranging from 0% to 10% with 2% increase plays an important role in determining the
Pendulum Test Value.
Three types of surface type of AC10, AC14 and AC20 were used in this study
to determine the differences surface texture and texture depth that could contribute to
the variability of Pendulum Test Value. Therefore, the effect of rainfall intensity,
crossfall and surface type on Pendulum Test Value is determined. A summary of the
Pendulum Test Value is shown as in Table 4.6. The highest Pendulum Test Value of
58
99 is recorded for AC20 at the 30mm/hr rainfall intensity with 10% crossfall while
the lowest Pendulum Test Value of 82 is recorded for AC10 at 432.4 mm/hr rainfall
intensity with 0% crossfall. The PTV is high ranged at 82 to 99 because the
constructed pavement is fresh and new where it provides high friction effect if
compared to polished value which is only 50. The full results of Pendulum Test
Value are attached in Appendix 8.
Table 4.6: Pendulum Test Value for AC10, AC14 and AC20
Intensity
Normal
30mm/hr
202.5mm/hr
432.4mm/hr
Crossfall
%
0
2
4
6
8
10
0
2
4
6
8
10
0
2
4
6
8
10
0
2
4
6
8
10
AC10
87
88
89
91
92
93
85
86
87
88
90
92
84
85
86
87
88
91
82
84
85
85
88
89
Pendulum Test Value
AC14
87
89
90
92
93
94
86
87
89
91
93
94
84
85
87
90
92
93
82
84
87
89
91
92
AC20
89
91
92
95
98
99
87
88
91
95
98
99
85
86
89
92
95
97
84
85
88
92
94
96
59
4.11.1 The Effect of Rainfall Intensity on Pendulum Test Value
Figure 4.4 to Figure 4.9 shows that the Pendulum Test Value decreases as the
rainfall intensity increase for AC10, AC14 and AC20. The Pendulum Test Value
decreases due to the excessive water on pavement surface during high rainfall
intensity compared to the low rainfall intensity where only small water film thickness
present on the pavement surface. This statement can be supported with research by
Chesterton.J et.al. (2006) indicating that increase in intensity increases the water film
thickness as shown in Figure 2.10. When there is a water film thickness on
pavement, the energy produced by the rubber slider dissipates resulting to a smaller
Pendulum Test Value.
As the crossfall increases, the Pendulum Test Value increases for all three
rainfall intensities. At 0% crossfall, the trendline for each surface type is close to
each other. However, the trendline tends to move away within the three surface type.
However, the trendline tends to move away within the three surface types when the
crossfall increase. The effect of intensity is prominent at higher crossfall because
small water film thickness is present on the surface because the macrotexture
effectively drains off the water increasing microtexture thus resulting to a higher
PTV value.
60
PTV vs Rainfall Intensity
(0% Crossfall)
y = -0.0075x + 85.372
R 2 = 0.9748
y = -0.0099x + 86.261
R 2 = 0.9991
y = -0.0074x + 87.019
R 2 = 0.952
88
87
PTV
86
AC10
85
AC14
84
AC20
83
82
81
0
100
200
300
400
500
Intensity (mm/hr)
Figure 4.4: Pendulum Test Value vs Intensity at 0% Crossfall
PTV vs Rainfall Intensity
(2% Crossfall)
89
y = -0.005x + 86.13
R 2 = 0.9991
y = -0.0074x + 87.019
R 2 = 0.952
y = -0.0074x + 88.019
R 2 = 0.952
PTV
88
87
AC10
86
AC14
85
AC20
84
83
0
100
200
300
400
Intensity (mm/hr)
Figure 4.5: Pendulum Test Value vs Intensity at 2% Crossfall
500
61
PTV
PTV vs Rainfall Intensity
(4% Crossfall)
92
91
90
89
88
87
86
85
84
y = -0.005x + 87.13
R 2 = 0.9991
y = -0.0049x + 88.777
R2 = 0.723
y = -0.0074x + 91.019
R 2 = 0.952
AC10
AC14
AC20
0
100
200
300
400
500
Intensity (mm/hr)
Figure 4.6: Pendulum Test Value vs Intensity at 4% Crossfall
PTV vs Rainfall Intensity
(6% Crossfall)
y = -0.0075x + 88.372
R 2 = 0.9748
y = -0.005x + 91.13
R2 = 0.9991
y = -0.0073x + 94.665
R2 = 0.723
96
PTV
94
92
AC10
90
AC14
88
AC20
86
84
0
100
200
300
400
Intensity (mm/hr)
Figure 4.7: Pendulum Test Value vs Intensity at 6% Crossfall
500
62
PTV vs Rainfall Intensity
(8% Crossfall)
100
y = -0.0049x + 89.777
R 2 = 0.723
y = -0.005x + 93.13
R 2 = 0.9991
y = -0.0098x + 97.907
R 2 = 0.9059
98
PTV
96
AC10
94
AC14
92
AC20
90
88
86
0
100
200
300
400
500
Intensity (mm/hr)
Figure 4.8: Pendulum Test Value vs Intensity at 8% Crossfall
PTV vs Rainfall Intensity
(10% Crossfall)
y = -0.0075x + 92.372
R 2 = 0.9748
y = -0.005x + 94.13
R2 = 0.9991
y = -0.0074x + 99.019
R 2 = 0.952
100
PTV
98
96
AC10
94
AC14
92
AC20
90
88
0
100
200
300
400
Intensity (mm/hr)
Figure 4.9: Pendulum Test Value vs Intensity at 10% Crossfall
500
63
4.11.2 The Effect of Crossfall on Pendulum Test Value
Figure 4.10 to Figure 4.12 shows that the Pendulum Test Value increase as
the crossfall increases. The 10% crossfall shows a higher Pendulum Test Value
compared to 0% crossfall. The 10% crossfall produces higher Pendulum Test Value
because it is good in runoff and water drainage. Besides that, at 10% crossfall, the
water film thickness is minimal between 0.5mm to 2.0mm compared to water film
thickness of 2.0mm up to 7.0mm at 0% crossfall where waterponding occurs on the
pavement surface. This statement can be justified by research carried out by
Chesterton.J et.al. (2006) which shows that higher percentage of flow path slope
generates smaller water film thickness as depicted in Figure 2.9.
The 10% crossfall is good in water drainage even at high rainfall intensity. At
10% crossfall, the Pendulum Test Value is getting closer to the normal condition for
different type of rainfall intensities for AC14 and AC20 indicating that the water film
thickness is approaching to zero value and the runoff is efficient. However, for the
0% crossfall, the Pendulum Test Value does not get closer to the normal condition
and has a big difference between low intensity, medium intensity and high intensity.
At high rainfall intensity at 0% crossfall, the Pendulum Test Value is small because
the water film thickness is big at 6mm to 7mm and the runoff of water from the
pavement is not efficient. This creates a big gap of Pendulum Test Value against
normal condition where there is no high water film present on pavement during
normal condition.
64
y = 0.6286x + 86.857
R2 = 0.9878
PTV vs Crossfall (AC10)
y = 0.6857x + 84.571
R2 = 0.9681
94
y = 0.6429x + 83.619
R2 = 0.9382
y = 0.6714x + 82.143
R2 = 0.942
92
PTV Value
90
Normal
30mm/hr
220.5mm/hr
88
86
432.4mm/hr
84
82
80
0
2
4
6
8
10
12
Crossfall (% )
Figure 4.10: Pendulum Test Value vs Crossfall for AC10
y = 0.7x + 87.333
R2 = 0.9847
PTV vs Crossfall (AC14)
y = 0.8571x + 85.714
R2 = 0.989
96
y = 0.9857x + 83.571
R2 = 0.9786
94
y = 1.0429x + 82.286
R2 = 0.9823
PTV Value
92
Normal
30mm/hr
220.5mm/hr
432.4mm/hr
90
88
86
84
82
80
0
2
4
6
8
10
12
Crossfall (%)
Figure 4.11: Pendulum Test Value vs Crossfall for AC14
65
y = 1.0571x + 88.714
R2 = 0.9779
PTV vs Crossfall (AC20)
y = 1.3429x + 86.286
R2 = 0.971
105
y = 1.2857x + 84.238
R2 = 0.9862
y = 1.3x + 83.333
R2 = 0.979
PTV Value
100
Normal
30mm/hr
220.5mm/hr
432.4mm/hr
95
90
85
80
0
2
4
6
8
10
12
Crossfall (%)
Figure 4.12: Pendulum Test Value vs Crossfall for AC20
4.11.3 The Effect of Surface Texture on Pendulum Test Value
Figure 4.13 to Figure 4.18 shows that increase in the texture depth resulted to
increase in the Pendulum Test Value. This is due to the influence of macrotexture
effect and microtexture effect on the aggregate surface. When the texture depth of
the pavement increases, the macrotexture plays an important role in draining off the
water. High in texture depth which is also high in macrotexture provides better
drainage which reduces the water film thickness on pavement creating higher
microtexture which increases the PTV.
Study by Chesterton et.al. (2006) as in Figure 2.8 also proves that bigger
macrotexture creates smaller water film thickness. Another study by Stacy G.
Williams (2008) on macrotexture and microtexture effect on water film thickness as
depicted in Figure 2.11 and Figure 2.12 also proves that high in British Pendulum
66
Number and macrotexture creates a smaller water film thickness. AC20 represents
bigger texture depth while the AC10 represents a smaller texture depth. On 0%
crossfall, the Pendulum Test Value is the least compared to the Pendulum Test Value
for 10% crossfall which is the highest. At 0% crossfall, water tends to fill up the
macrotexture affecting the texture depth. Therefore, the macrotexture of the smallest
texture depth of AC10 is filled with water faster than the AC20 because its texture
depth is smaller. Apart from that, the AC10 does not have a good macrotexture as the
AC20 which is important in draining water and keeping the aggregates in contact
with tyre. At 10% crossfall, water is drained out faster and more efficiently
producing a bigger Pendulum Test Value.
In conjunction to that, the trendline slope of 10% crossfall is much steeper
than the 0% crossfall. This proves that the Pendulum Test Value shows big
differences at 10% crossfall for AC10, AC14 and AC20. However, at 0% crossfall,
the trendline is not very steep and the difference of Pendulum Test Value for AC10,
AC14 and AC20 is not that significant. At 0% crossfall, since the runoff is slow and
the water film is present, the AC10, AC14 and AC20 do not show a big difference in
Pendulum Test Value. However, at 10% crossfall when the water runoff is fast and
the water film thickness is small, the AC20 tends to produce better Pendulum Test
Value since the bigger macrotexture or bigger texture depth is capable of draining the
water efficiently. Therefore, the PTV was found to be the best at the condition of
10% crossfall with high texture depth type of surface.
67
Figure 4.13: Pendulum Test Value vs Texture Depth at 0% Crossfall
Figure 4.14: Pendulum Test Value vs Texture Depth at 2% Crossfall
68
Figure 4.15: Pendulum Test Value vs Texture Depth at 4% Crossfall
Figure 4.16: Pendulum Test Value vs Texture Depth at 6% Crossfall
69
Figure 4.17: Pendulum Test Value vs Texture Depth at 8% Crossfall
Figure 4.18: Pendulum Test Value vs Texture Depth at 10% Crossfall
70
4.12
Runoff and Water Film Thickness Data
Table 4.6 shows the water film thickness and the runoff flowrate of various
surface type and at various crossfall. The water film thickness is significant from 0%
crossfall to 10% crossfall. The highest water film thickness exist at 0% crossfall
because there is waterponding exist compared to 10% crossfall where water
drainange is good. However, the water film thickness only differs a little from AC10,
AC14 to AC20. This is because the water film thickness is so small ranging from
0mm to 7mm only throughout the rainfall test. Therefore, the difference between
each surface type is small. Apart from that, the water film thickness is measured
using a ruler which is not that accurate in readings. The water runoff varies from 0%
crossfall to 10%. The higher runoff was noticed to be at 10% crossfall since it drains
the water faster and more efficient. However, the 0% crossfall has the least water
runoff value due to water accumulating on the surface and ponding. The high rainfall
intensity produces high water film thickness and high runoff. Table 4.7 shows the
water film thickness and the water runoff results.
71
Table 4.7: The Water Film Thickness and The Runoff of AC10, AC14 and AC20
Rainfall
Crossfall
Intensity
(%)
AC10
AC14
AC20
Water Film
Q
Water Film
Q
Water Film
Q
Thickness(mm)
runoff
Thickness(mm)
runoff
Thickness(mm)
runoff
(ℓ/min)
(ℓ/min)
(ℓ/min)
0
2.0
0.26
2.0
0.26
2.0
0.26
2
1.5
0.26
1.5
0.26
1.5
0.26
30
4
1.5
0.26
1.5
0.26
1.5
0.26
mm/hr
6
1.0
0.52
1.0
0.52
1.0
0.52
8
1.0
0.52
1.0
0.52
1.0
0.52
10
0.5
0.52
0.5
0.52
0.5
0.52
0
4.0
2.10
3.0
2.10
3.0
2.10
2
3.0
2.10
2.5
2.10
2.5
2.10
202.5
4
2.5
2.10
2.5
2.10
2.5
2.10
mm/hr
6
2.0
2.36
2.0
2.36
2.0
2.36
8
1.5
2.36
1.5
2.36
1.5
2.36
10
1.0
2.62
1.0
2.62
1.0
2.62
0
7.0
4.72
6.0
4.72
6.0
4.72
2
6.0
4.72
5.0
4.72
5.0
4.72
432.4
4
5.0
4.98
5.0
4.98
4.0
4.98
mm/hr
6
4.0
4.98
4.0
4.98
3.0
4.98
8
3.0
5.24
3.0
5.24
3.0
5.24
10
2.0
5.24
2.0
5.24
2.0
5.24
4.13
Degree of Compaction and Specific Gravity
The degree of compaction is carried to assure that the 4% air void is
maintained. The compaction of the samples is done at 800 roll using the compaction
roller. Degree of compaction is obtained through the specific gravity of cored sample
72
and specific gravity superpave. Table 4.8 shows the average degree of compaction of
each mix type.The degree of compaction for AC10 meets the specification of 98%100% degree of compaction for wearing course. However, the degree of compaction
for AC20 did not meet the specification.
Therefore, permeability test is done on AC20 cored samples to investigate the
possibility of water infiltrating through the samples which could decrease the degree
of compaction value. The permeability test results are shown in Table 4.9.Based on
permeability result from Table 4.9, the average k value of AC20 mix which is 433 x
10-5 cm/s is bigger than 125 x 10-5 cm/s which proves that the AC20 is permeable.
The AC20 samples is permeable due to insufficient compaction effort creating bigger
air voids which eventually led to the smaller percentage in degree of compaction.
Table 4.8:The Specific Gravity and the Degree of Compaction of AC10, AC14,AC20
Mix Type
AC10
AC14
AC20
Air (g)
SSD (g)
Water (g)
SG Core
SG Superpave
1007.9
1008.8
560.1
2.246
99
1107.7
1108.4
614.4
2.242
98
1100.1
1101.1
610.0
2.240
1078.7
1079.9
595.9
2.229
98
1074.0
1074.8
598.0
2.253
99
1079.1
1083.6
597.5
2.220
97
942.2
949.0
525.0
2.222
97
1131.4
1135.5
630.3
2.240
994.1
997.1
556.5
2.256
98
1129.4
1133.5
630.0
2.243
98
1029.7
1037.9
580.8
2.253
96
1116.6
1118.7
625.3
2.263
96
1158.7
1162.7
655.7
2.285
1135.1
1141.9
636.6
2.246
96
1081.4
1086.6
606.0
2.250
96
2.278
2.297
2.352
DOC(%) Average
98
97
97
98.4
97.4
96.2
73
Table 4.9: Permeability Test Result of AC20
A (cm2)
L (cm)
A (cm2)
T (s)
h1 (cm)
h2 (cm)
K (cm/s)
7.070
6.400
78.550
11.000
67.000
62.000
0.00406
7.070
6.800
78.550
11.000
67.000
62.000
0.00432
7.070
7.000
78.550
12.000
67.000
62.000
0.00407
7.070
7.000
78.550
10.000
67.000
62.000
0.00489
7.070
6.800
78.550
11.000
67.000
62.000
0.00432
Average
0.00433
4.14
Analysis of Variance (ANOVA) Results
ANOVA is carried out to determine the factors affecting PTV differs or
otherwise according to the variables. The summary of ANOVA results on each factor
is shown in Table 4.10 to Table 4.15 below. The complete analysis of ANOVA is
attached in Appendix 9.
The table 4.10 shows that at 0% and 2% crossfall, there is a significant
difference between the three types of intensities condition. This proves that at 0%
crossfall and at 2% crossfall, the differences in intensities affect the Pendulum Test
Value. However, at 4%, 6%, 8% and 10% crossfall, there is no significant
differences can be noticed between the three rainfall intensities. This means that the
Pendulum Test Values does not vary much between the three rainfall intensities at
4% to 10% crossfall. Besides that, insignificant PTV value between low rainfall
intensity to high rainfall intensity proves that the water even at high rainfall intensity
could be drained efficiently reducing the water film thickness on the pavement.
74
Based on the ANOVA analysis, the range from 4% to 10% crossfall have
proven to be efficient in draining water. However, the 10% crossfall could not be
implemented on roads construction because it affects the driver’s comfort and the
driving maneuver. Therefore, the 4% crossfall can be suggested to be implemented
for future road construction since it also has a good drainage effect to resist skid.
Table 4.10: ANOVA of Rainfall Intensity on Crossfall
Variables
Pvalue
Pcrit
Condition
Significant
Remarks
0% crossfall
0.014247
0.05
P < Pcrit
Yes
Reject Ho, Accept H1
2% crossfall
0.012875
0.05
P < Pcrit
Yes
Reject Ho, Accept H1
4% crossfall
0.296296
0.05
P > Pcrit
No
Accept Ho, Reject H1
6% crossfall
0.615448
0.05
P > Pcrit
No
Accept Ho, Reject H1
8% crossfall
0.651732
0.05
P > Pcrit
No
Accept Ho, Reject H1
10% crossfall
0.650963
0.05
P > Pcrit
No
Accept Ho, Reject H1
Table 4.11 shows no significant differences were noticed on PTV between
rainfall intensities for AC10, AC14 and AC20. However, AC20 has the highest
Pvalue towards rainfall intensities which proves that the AC20 is the least significant
towards the three rainfall intensities because the water film thickness of AC20 is
lesser due to the high texture depth it has to drain water efficiently.
Table 4.11: ANOVA of Rainfall Intensity on Surface Type
Variables
Pvalue
Pcrit
Condition
Significant
Remarks
AC10
0.269631
0.05
P > Pcrit
No
Accept Ho, Reject H1
AC14
0.504632
0.05
P > Pcrit
No
Accept Ho, Reject H1
AC20
0.531881
0.05
P > Pcrit
No
Accept Ho, Reject H1
Table 4.12 shows PTV of 30mm/hr, 202.5mm/hr and 432.4mm/hr rainfall
intensities has significant difference between crossfall. 30mm/hr recorded the least
75
significant since 30mm/hr produces small water film thickness on pavement giving
the almost similar PTV readings.
Table 4.12: ANOVA of Crossfall on Rainfall Intensity
Variables
Pvalue
Pcrit
Condition
Significant
Remarks
30mm/hr
0.010865
0.05
P < Pcrit
Yes
Reject Ho, Accept H1
202.5mm/hr
0.001834
0.05
P < Pcrit
Yes
Reject Ho, Accept H1
432.4mm/hr
0.003194
0.05
P < Pcrit
Yes
Reject Ho, Accept H1
Table 4.13 shows that the Pendulum Test Value of all three AC10, AC14 and
AC20 has significant difference towards increase percentage of the crossfall. This
proves that as the crossfall increases, there is increase on the Pendulum Test Value
indicating good skidding resistance. This means that the AC10, AC14 and AC20 is
good in resisting skid when the crossfall gets bigger. However, AC20 is the best
surface type to drain water due to the higher macrotexture it has. AC20 proves to be
the best surface type since the Pvalue of AC20 recorded the least value proving that
AC20 is the significantly high towards crossfall from the ANOVA analysis.
Table 4.13: ANOVA of Crossfall on Surface Type
Variables
Pvalue
Pcrit
Condition
Significant
Remarks
AC10
0.001384
0.05
P < Pcrit
Yes
Reject Ho, Accept H1
AC14
2.36E-06
0.05
P < Pcrit
Yes
Reject Ho, Accept H1
AC20
5.24E-07
0.05
P < Pcrit
Yes
Reject Ho, Accept H1
Table 4.14 shows PTV of 30mm/hr, 202mm/hr and 432.4 mm/hr rainfall
intensities has no significant between surface type. However, the 30mm/hr rainfall
intensity shows a slightly higher differences between AC10, AC14 and AC20
because it produce only a small water film thickness and the surface texture plays a
76
more important role. However, with the largest rainfall intensity, coarser surface
texture covered by water.
Table 4.14: ANOVA of Surface Type on Rainfall Intensity
Variables
Pvalue
Pcrit
Condition
Significant
Remarks
30mm/hr
0.104239
0.05
P > Pcrit
No
Accept Ho, Reject H1
202.5mm/hr
0.248989
0.05
P > Pcrit
No
Accept Ho, Reject H1
432.4mm/hr
0.19509
0.05
P > Pcrit
No
Accept Ho, Reject H1
Table 4.15 shows that 0% and 2% crossfall has no significant differences on
PTV between surface type. This proves that 4% to 10% crossfall drains water
efficiently providing different PTV according to the surface type where the surface
texture plays an important role in determining PTV.
Table 4.15: ANOVA of Surface Type on Crossfall
Variables
Pvalue
Pcrit
Condition
Significant
Remarks
0% crossfall
0.489078
0.05
P > Pcrit
No
Accept Ho, Reject H1
2% crossfall
0.506023
0.05
P > Pcrit
No
Accept Ho, Reject H1
4% crossfall
0.046258
0.05
P < Pcrit
Yes
Reject Ho, Accept H1
6% crossfall
0.005248
0.05
P < Pcrit
Yes
Reject Ho, Accept H1
8% crossfall
0.003613
0.05
P < Pcrit
Yes
Reject Ho, Accept H1
10% crossfall
0.002843
0.05
P < Pcrit
Yes
Reject Ho, Accept H1
77
CHAPTER V
CONCLUSION AND RECOMMENDATIONS
5.1
Conclusion
Based on the results above, the following conclusions can be made:
The Pendulum Test Value on 432.4mm/hr intensity is smaller than the
Pendulum Test Value of 220.5mm/hr and 30mm/hr. The 30mm/hr Pendulum Test
Value has the highest Pendulum Test Value because of the existence of small water
film thickness on the pavement.
The Pendulum Test Value for 0% crossfall is smaller than the 10% crossfall.
The 10% crossfall has a greater Pendulum Test Value because it is efficient in water
drainage creating smaller water film thickness thus resulting to a bigger PTV.
However, the 10% crossfall could not be implemented since it affects the driver’s
comfort. Therefore, this study recommend 4% crossfall since the ANOVA results
proved that crossfall ranging from 4% to 10% is efficient in water drainage from the
pavement.
78
The Pendulum Test Value for AC20 is bigger than the AC14 and AC10. This
is because the AC20 surface texture is rough and good in water drainage since it has
a high macrotexture depth.
5.2
Recommendation
It is recommended that this study is continued with a broader scope since this
study can provide a good knowledge on skidding. The recommendations to further
this regarding the factors affecting Pendulum Test Value are as follows:
1.
Stone Mastic Asphalt mixes and Porous Asphalt mixes are recommended to
further investigate the factors affecting Pendulum Test Value since this study
is only limited to Asphaltic Concrete mixes only. Besides that, it is suggested
to expand the scope of this study by introducing the rigid pavement mixes
and surface dressing mixes. These will provide a good comparison of
different types of mixes besides obtaining the best mix type that resist
skidding.
2.
This study only uses three different rainfall conditions which are low rainfall
intensity, medium rainfall intensity and high rainfall intensity. It is
recommended that the intensities could be varied to at least five rainfall
conditions to provide a good relationship against Pendulum Test Value.
3.
Dynamic load or tyre is suggested to be located into the rainfall simulator in
order to obtain speed that could cause hydroplaning and skidding since speed
is one of the major contributors to skidding during rainy climate.
79
4.
A better measurement tool is suggested to replace the use of ruler to measure
the water film thickness and the runoff depth to produce an accurate reading
during the event of rainfall.
REFERENCES
American Association of State Highway and Transportation Officials, “A Policy on
Geometric Design of Highways and Streets” 1990
ASCE, Design Manual for Storm Drainage, American Society of Civil Engineers,
New York, 1960.
British Pendulum Manual, Operation manual of the British Pendulum SKID
resistance tester. United Kingdom:Wessex Engineering Ltd. 2000.
British Standards Institution. BS EN 13036-4. Method for measurement of
slip/skid resistance of a surface – The Pendulum Test. London: British Standards
Institution. 2003
Brouwer et.al (1985), Introduction to Irrigation, Food And Agriculture Organization
Of The United Nations (FAO), Rome, Italy
Chesterton.J et.al. (2006). The Use of the Gallaway Formula for Aquaplaning
Evaluation in New Zealand, Transportation and the Pursuit of Excellence,
NZIHT & Transit NZ 8th Annual Conference.
81
David A. Noyce et.al. (2005), Incorporating Road Safety Into Pavement
Management: Maximizing Asphalt Pavement Surface Friction For Road
Safety Improvements, Midwest Regional University Transportation Center,
Traffic Operations and Safety (TOPS) Laboratory.
Elizabeth Chong Eu Mee (2006). The Effects Of Nominal Maximum Aggregate Size
On The Properties of HMA Using Gyrotory Compactor. Universiti
Teknologi Malaysia.Master Project Report.
Flintsch G.W et al. Effect of HMA Properties on Pavement Surface Characteristics.
Department of Civil and Environmental Engineering, Virginia Polytechnic
Institute and State University.
Gallaway, B. M., et. al., (1979). Pavement and Geometric Design Criteria for
Minimizing Hydroplaning, Federal Highway Administration, Report No.
FHWA-RD-79-31.
Garrabrant.R, (2004), Highway Engineering, Standard Handbook for Civil
Engineering, McGraw Hill.
Jabatan Kerja Raya (JKR), A Guide on Geometric Design of Roads, Arahan
Teknik Jalan (8/86), Public Works Department Malaysia. 1986
Jabatan Kerja Raya. Standard Specification for Road Works. Kuala Lumpur.
JKR/SPJ. 1998-rev 2005
82
Kennedy, C.K., Young, A.E., Butler, I. C. (1990). Measurement of Skidding
Resistance and Surface Texture and the Use of Results in the United
Kingdom. ASTM STP 1031, Philadelphia, U.S.A.
Kenneth Young et al. (2004), Hydraulic Design of Drainage for Highways,
Hydraulic Design Handbook, McGraw Hill.
Kokkalis. A.G. Tsohos. G.L, Panagoul.O.K, (2002), Consideration of Fractals
Potential in Pavement Skid Resistance Evaluation. Journal of Transportation
Engineering.
Manual Saliran Mesra Alam, Design Rainfall, Urban Stromwater Management
Manual, Kuala Lumpur, Malaysia: JPS Malaysia Chapter 13.Pg 13-4. 2000
Molenaar et.al, (2004). Geometric and structural design of roads and railways:
Structural design of roads. CT3041. United Kingdom
Nor Zolhanita Binti Ahmad Tarmidi (2007). Penilaian Terhadap Campuran Halus
Gred Terbuka. Universiti Teknologi Malaysia.Degree Project Report.
Peter Cairney et.al. (2006), A pilot study of the effects of macrotexture on stopping
distance, Atsb Transport Safety REPORT CR 226, Australian Transport
Safety Bureau, Australia.
83
Rebecca S. Mc Daniel. (2004). Field Evaluation of Porous Asphalt Pavement. United
States. North Central Superpave Center
Roberts,F.L., P.S., Brown, E.R., Lee, Y.D.,Kennedy, T.W, (1996), ‘Hot-Mix
Asphalt Materials, Mixture Design and Construction’, Napa Research and
Education Foundation.
Royal Malaysia Police, Road Accident Statistical Report, Road Traffic Branch, Bukit
Aman. 2005.
Sandberg U. (1997). Influence on Road Surface Texture on Traffic Characteristics
Related to Environment, Economy, and Safety. A State-of-the-art Study
Regarding Measures and Measuring Methods.
Swedish National Road and
Transport Research Institute.
Stacy G. Williams (2008). Surface Friction Measurements of Fine-Graded Asphalt
Mixtures, Mack-Blackwell Transportation Center 2066, Arkansas. Final Report
84
APPENDIX 1
SIEVE ANALYSIS GRADATION
Percentage Passing (by weight)
Mix Design
AC10
AC14
AC20
BS Sieve
^0.45
LL
Gradation
UL
LL
Gradation
UL
LL
Gradation
UL
28
4.479
-
-
-
-
-
-
100
100
100
20
3.85
-
-
-
100
100
100
76
94
100
14
3.279
100
100
100
90
93
100
64
80
89
10
2.818
90
95
100
76
79
86
56
72
81
5
2.063
58
65
72
50
56
62
46
58
71
3.35
1.723
48
56
64
40
47
54
32
49
58
1.18
1.077
22
27
40
18
23
34
20
33
42
0.425
0.68
12
15
26
12
14
24
12
22
28
0.15
0.426
6
10
14
6
10
14
6
12
16
0.075
0.312
4
6
8
4
6
8
4
6
8
LL – Lower Limit
UL – Upper Limit
85
APPENDIX 2
WASH SIEVE
Mix
AC10
Sample
AC14
AC20
I
II
I
II
I
II
Mass Before Washing
g
1128.0
1128.0
1128.0
1128.0
1128.0
1128.0
Mass After Washing
g
1079.3
1076.0
1084.3
1086.1
1087.5
1090.0
Mass of Dust
g
48.7
52.0
43.7
41.9
40.5
38.0
Average
g
50.4
42.8
39.3
Percentage
%
4.5
3.8
3.5
205.3
174.5
160.1
Weight of Dust for 4600g g
86
APPENDIX 3
PENETRATION AND SOFTENING TEST
Penetration Test
1
2
3
4
5
Average
Sample 1
82
83
82
84
83
82.8
Sample 2
84
82
83
83
83
83.0
Sample 3
82
84
82
82
83
82.6
Average Penetration
82.8 PEN
Softening Point Test
Ball A
Ball B
Average Temperature
Sample 1
47.0
47.5
47.3
Sample 2
46.0
46.0
46.0
Average Temperature
46.7 ºC
87
APPENDIX 4
MASS FOR RAINFALL SIMULATOR MOULD
Mass/Volume = (100 – Air void)/100 x TMD
Volume
= 118cm x 61cm x 6cm
= 43188cm3
ACW 14
Mass/43188
= (100 – 4)/100 x 2.400
= 100.0kg
ACW 20
Mass/43188
= (100 – 4)/100 x 2.443
= 101.5kg
88
ACW 10
Mass/43188
= (100 – 4)/100 x 2.391
= 99.5kg
ACW 10
Aggregate
Bitumen
(Total Mix Mass = 99.5kg)
93.700
x 99.5
=
93.232 kg
6.300
x 99.5
=
6.269 kg
Mass
Sieve
%
Size
Passing
% Retained
14.000
100.000
-
10.000
95.000
5.000
4.662 kg
5.000
65.000
30.000
27.969 kg
3.350
56.000
9.000
8.391 kg
1.180
27.000
29.000
27.037 kg
0.425
15.000
12.000
11.188 kg
0.150
10.000
5.000
4.662 kg
0.075
6.000
4.000
3.729 kg
Pan
-
6.000
5.594 kg
100.000
93.232 kg
Total
Required
- kg
89
ACW 14
Aggregate
Bitumen
(Total Mix Mass = 100.0kg)
94.200
x 100
=
94.200 kg
5.800
x 100
=
5.800 kg
Sieve
%
Size
Passing
% Retained
20.000
100.000
-
14.000
93.000
7.000
6.594 kg
10.000
79.000
14.000
13.188 kg
5.000
56.000
23.000
21.666 kg
3.350
47.000
9.000
8.478 kg
1.180
23.000
24.000
22.608 kg
0.425
14.000
9.000
8.478 kg
0.150
10.000
4.000
3.768 kg
0.075
6.000
4.000
3.768 kg
Pan
-
6.000
5.652 kg
100.000
94.200 kg
Total
Mass
Required
- kg
90
ACW 20
Aggregate
Bitumen
(Total Mix Mass = 101.5kg)
95.800
x 101.5
=
97.237 kg
4.200
x 101.5
=
4.263 kg
Sieve
%
Size
Passing
% Retained
28.000
100.000
-
20.000
94.000
6.000
5.834 kg
14.000
80.000
14.000
13.613 kg
10.000
72.000
8.000
7.779 kg
5.000
58.000
14.000
13.613 kg
3.350
49.000
9.000
8.751 kg
1.180
33.000
16.000
15.558 kg
0.425
22.000
11.000
10.696 kg
0.150
12.000
10.000
9.724 kg
0.075
6.000
6.000
5.834 kg
Pan
-
6.000
5.834 kg
100.000
97.237 kg
Total
Mass
Required
- kg
91
APPENDIX 5
RAINFALL DESIGN AND RATIONAL METHOD
Rainfall Design using MASMA
Design values are taken for Kuala Lumpur at duration time of 5 minutes.
For ARI =100 Years Design Storm
ln(It) = a + b (ln t) + c (ln t)2 + d (ln t)3
a
= 5.0064
b
= 0.8709
c
= -0.3070
d
= 0.0186
ln I30 = 5.1490
I30 = 172.24 mm/hr
P30 = 86.12 mm
ln I60 = 4.702
I60 = 110.20 mm/hr
P60 = 110.20 mm
FD = 2.08 (Table 13.3)
PD = P30 – FD (P60 – P30)
100
P5
= 86.12 – 2.08 (110.20 – 86.12) = 36.03 mm
100
I5
= 36.03 / (1/12)
= 432.40 mm/hr
92
Design values are taken for Kuala Lumpur at duration time of 15 minutes.
For ARI = 20 Years Design Storm
ln(It) = a + b (ln t) + c (ln t)2 + d (ln t)3
a
= 4.9781
b
= 0.7533
c
= -0.2796
d
= 0.0166
ln I30 = 4.959
I30 = 142.4 mm/hr
P30 = 71.2 mm
ln I60 = 4.515
I60 = 91.3 mm/hr
P60 = 91.3 mm
FD = 0.8 (Table 13.3)
PD
= P30 – FD (P60 – P30)
20
P15
= 71.20 – 0.8 (91.30 – 71.20) = 55.12 mm
20
I15
= 55.12 / (1/4)
= 220.48 mm/hr
Data taken from IDF Curve (MASMA) for duration 30 min and above.
For ARI = 2 Years Design Storm
Intensity100
= 30 mm/hr
Rational Method Calculation
Q = 0.278 C I A
C = 0.95
93
High Intensity100
Medium Intensity20
= 432.40 mm/hr
= 220.48 mm/hr
Low Intensity2
Area
= 30 mm/hr
= 1.18m x 0.61m
= 0.00118km x 0.00061km
=7.2 x 10-7 km2
High Intensity Flowrate
100
Q5 = 0.278 x 0.95 x 432.40 x (7.2 x 10-7)
= 8.222 x 10-5 m3/s
= 0.082 ℓ /s
= 4.93 ℓ/min
Medium Intensity Flowrate
20
Q5
= 0.278 x 0.95 x 220.48 x (7.2 x 10-7)
= 4.192 x 10-5 m3/s
= 0.042 ℓ /s
= 2.52 ℓ/min
Low Intensity Flowrate
2
Q100 = 0.278 x 0.95 x 30 x (7.2 x 10-7)
= 5.704 x 10-6 m3/s
= 5.704 x 10-3 ℓ /s
= 0.34 ℓ/min
Therefore;
The acceptable range for design rainfall flow rate = 0.3ℓ/min to 5ℓ/min
94
APPENDIX 6
PAVEMENT CROSSFALL
2% Crossfall
2 / 100 = X / 0.61
X = 0.0122m = 1.22cm
X
0.61
4% Crossfall
8% Crossfall
4 / 100 = X / 0.61
8 / 100 = X / 0.61
X = 0.0244m = 2.44cm
X = 0.0488m = 4.88cm
6% Crossfall
10% Crossfall
6 / 100 = X / 0.61
10 / 100 = X / 0.61
X = 0.0366m = 3.66cm
X = 0.061m = 6.10cm
95
APPENDIX 7
SAND DENSITY AND VOLUME CALCULATION
The procedure of calculating sand volume and density are as follows:
Mass of Mould
= 35.2g
Mass of Mould + Sand
= 65.5g
Mass of Sand
= 30.3g
Mass of Mould + Water
= 59.4g
Mass of Water
= 24.2g
Assumption: Volume of water = Volume of sand
Volume of water
= Mass of water / Density of water
= 24.2 g / 1000 kg/m3
= 0.0000242 m3
Volume of sand
= Volume of water
= 0.0000242 m3
Density of sand
= Mass of sand / Volume of sand
= 30.3 g / 0.0000242 m3
= 1252kg/ m3
APPENDIX 8
PENDULUM TEST VALUE
Normal Condition (AC10)
Crossfall (%)
PTV
Temperature Correction Corrected
1
2
3
4
5
Average
ºC
Factor
PTV
0
86
86
85
85
84
85
33
+2
87
2
87
87
86
86
86
86
33
+2
88
4
88
88
87
87
87
87
33
+2
89
6
89
89
89
88
88
89
33
+2
91
8
91
91
90
90
90
90
33
+2
92
10
91
91
91
91
91
91
33
+2
93
97
30mm/hr Rainfall Intensity Condition (AC10)
Crossfall (%)
PTV
Temperature Correction Corrected
1
2
3
4
5
Average
ºC
Factor
PTV
0
83
83
83
83
83
83
32
+2
85
2
84
84
84
83
83
84
32
+2
86
4
85
85
85
85
84
85
32
+2
87
6
87
86
86
86
86
86
32
+2
88
8
88
88
88
88
88
88
32
+2
90
10
90
90
90
89
89
90
32
+2
92
220.5mm/hr Rainfall Intensity Condition (AC10)
Crossfall (%)
PTV
Temperature Correction Corrected
1
2
3
4
5
Average
ºC
Factor
PTV
0
82
82
82
82
81
82
32
+2
84
2
84
84
83
83
83
83
32
+2
85
4
85
85
85
84
83
84
32
+2
86
6
86
85
85
85
85
85
32
+2
87
8
87
87
86
86
86
86
32
+2
88
10
89
89
89
88
88
89
32
+2
91
98
432.4mm/hr Rainfall Intensity Condition (AC10)
Crossfall (%)
PTV
Temperature Correction Corrected
1
2
3
4
5
Average
ºC
Factor
PTV
0
81
80
80
80
80
80
31
+2
82
2
83
82
81
81
81
82
31
+2
84
4
83
83
83
82
82
83
31
+2
85
6
84
83
83
83
83
83
31
+2
85
8
87
86
85
85
85
86
31
+2
88
10
88
88
88
87
86
87
31
+2
89
Normal Condition (AC14)
Crossfall (%)
PTV
1
Temperature Correction Corrected
2
3
4
5
Average
ºC
Factor
PTV
0
85 85
85
85
85
85
29
+2
87
2
87 87
87
86
86
87
29
+2
89
4
88 88
88
88
87
88
29
+2
90
6
90 90
90
89
89
90
29
+2
92
8
92 92
91
91
91
91
29
+2
93
10
93 93
92
92
92
92
29
+2
94
99
30mm/hr Rainfall Intensity Condition (AC14)
Crossfall (%)
PTV
1
Temperature Correction Corrected
2
3
4
5
Average
ºC
Factor
PTV
0
84 84
84
83
83
84
29
+2
86
2
85 85
85
85
85
85
29
+2
87
4
87 87
87
86
86
87
29
+2
89
6
90 90
89
89
89
89
29
+2
91
8
92 92
91
91
90
91
29
+2
93
10
93 93
92
92
92
92
29
+2
94
220.5mm/hr Rainfall Intensity Condition (AC14)
Crossfall (%)
PTV
1
Temperature Correction Corrected
2
3
4
5
Average
ºC
Factor
PTV
0
82 82
82
82
82
82
29
+2
84
2
84 84
83
83
83
83
29
+2
85
4
86 86
85
85
85
85
29
+2
87
6
89 89
88
88
88
88
29
+2
90
8
90 90
90
89
89
90
29
+2
92
10
91 91
91
90
90
91
29
+2
93
100
432.4mm/hr Rainfall Intensity Condition (AC14)
Crossfall (%)
PTV
1
Temperature Correction Corrected
2
3
4
5
Average
ºC
Factor
PTV
0
81 80
80
80
80
80
29
+2
82
2
82 82
82
81
81
82
29
+2
84
4
86 86
86
85
84
85
29
+2
87
6
88 88
87
87
87
87
29
+2
89
8
90 90
89
89
89
89
29
+2
91
10
91 91
90
90
90
90
29
+2
92
Normal Condition (AC 20)
Crossfall (%)
PTV
Temperature Correction Corrected
1
2
3
4
5
Average
ºC
Factor
PTV
0
88
87
87
87
87
87
30
+2
89
2
90
90
90
89
88
89
30
+2
91
4
90
90
90
90
90
90
30
+2
92
6
93
93
93
93
93
93
30
+2
95
8
96
96
96
96
95
96
30
+2
98
10
97
97
97
97
97
97
30
+2
99
101
30mm/hr Rainfall Intensity Condition (AC 20)
Crossfall (%)
PTV
Temperature Correction Corrected
1
2
3
4
5
Average
ºC
Factor
PTV
0
85
85
85
85
85
85
30
+2
87
2
86
86
86
86
85
86
30
+2
88
4
90
89
89
89
89
89
30
+2
91
6
93
93
93
93
93
93
30
+2
95
8
97
96
96
96
95
96
30
+2
98
10
97
97
97
96
96
97
30
+2
99
220.5mm/hr Rainfall Intensity Condition (AC 20)
Crossfall (%)
PTV
Temperature Correction Corrected
1
2
3
4
5
Average
ºC
Factor
PTV
0
83
83
83
83
83
83
29
+2
85
2
85
85
84
84
84
84
29
+2
86
4
87
87
87
87
86
87
29
+2
89
6
90
90
90
90
90
90
29
+2
92
8
93
93
93
92
92
93
29
+2
95
10
96
96
96
95
94
95
29
+2
97
102
432.4mm/hr Rainfall Intensity Condition (AC 20)
Crossfall (%)
PTV
Temperature Correction Corrected
1
2
3
4
5
Average
ºC
Factor
PTV
0
82
82
82
82
81
82
29
+2
84
2
84
83
83
83
83
83
29
+2
85
4
86
86
86
86
86
86
29
+2
88
6
90
90
90
89
89
90
29
+2
92
8
93
92
92
92
91
92
29
+2
94
10
95
94
94
93
93
94
29
+2
96
103
APPENDIX 9
ANALYSIS OF VARIANCE
Effect of Rainfall Intensity on Crossfall
ANOVA of Rainfall Intensity on 0% Crossfall
30 mm/hr
85
86
87
202.5mm/hr
84
84
85
432.4 mm/hr
82
82
84
SUMMARY
Groups
Column 1
Column 2
Column 3
Count
3
3
3
Sum
258
253
248
Average
86
84.33333333
82.66666667
Variance
1
0.333333
1.333333
ANOVA
Source of Variation
Between Groups
Within Groups
SS
16.66666667
5.333333333
df
2
6
MS
8.333333333
0.888888889
F
9.375
Total
22
8
AC10
AC14
AC20
Anova: Single Factor
P-value
F crit
0.014247 5.143253
104
ANOVA of Rainfall Intensity on 2% crossfall
30 mm/hr
86
87
88
202.5mm/hr
85
85
86
432.4 mm/hr
84
84
85
SUMMARY
Groups
Column 1
Column 2
Column 3
Count
3
3
3
Sum
261
256
253
Average
87
85.33333333
84.33333333
Variance
1
0.333333
0.333333
ANOVA
Source of Variation
Between Groups
Within Groups
SS
10.88888889
3.333333333
df
2
6
MS
5.444444444
0.555555556
F
9.8
Total
14.22222222
8
AC10
AC14
AC20
Anova: Single Factor
P-value
F crit
0.012875 5.143253
105
ANOVA of Rainfall Intensity on 4% crossfall
AC10
AC14
AC20
30 mm/hr
87
89
91
202.5mm/hr
86
87
89
432.4 mm/hr
85
87
88
Sum
Average
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
ANOVA
Source of Variation
Between Groups
Within Groups
Total
Count
3
3
3
SS
8.666666667
17.33333333
26
267
262
260
df
2
6
8
Variance
89
4
87.33333333 2.333333
86.66666667 2.333333
MS
4.333333333
2.888888889
F
P-value
F crit
1.5 0.296296 5.143253
106
ANOVA of Rainfall Intensity on 6% crossfall
30 mm/hr
88
91
95
202.5mm/hr
87
90
92
432.4 mm/hr
85
89
92
SUMMARY
Groups
Column 1
Column 2
Column 3
Count
3
3
3
Sum
274
269
266
Average
91.33333333
89.66666667
88.66666667
Variance
12.33333
6.333333
12.33333
ANOVA
Source of Variation
Between Groups
Within Groups
SS
10.88888889
62
df
2
6
MS
5.444444444
10.33333333
F
P-value
F crit
0.526882 0.615448 5.143253
Total
72.88888889
8
AC10
AC14
AC20
Anova: Single Factor
107
ANOVA of Rainfall Intensity on 8% crossfall
30 mm/hr
90
93
98
202.5mm/hr
88
92
95
432.4 mm/hr
88
91
94
SUMMARY
Groups
Column 1
Column 2
Column 3
Count
3
3
3
Sum
281
275
273
Average
93.66666667
91.66666667
91
Variance
16.33333
12.33333
9
ANOVA
Source of Variation
Between Groups
Within Groups
SS
11.55555556
75.33333333
df
2
6
MS
5.777777778
12.55555556
F
P-value
F crit
0.460177 0.651732 5.143253
Total
86.88888889
8
AC10
AC14
AC20
Anova: Single Factor
108
ANOVA of Rainfall Intensity on 10% crossfall
30 mm/hr
92
94
99
202.5mm/hr
91
93
97
432.4 mm/hr
89
92
96
SUMMARY
Groups
Column 1
Column 2
Column 3
Count
3
3
3
Sum
285
281
277
Average
95
93.66666667
92.33333333
Variance
13
9.333333
12.33333
ANOVA
Source of Variation
Between Groups
Within Groups
SS
10.66666667
69.33333333
df
2
6
MS
5.333333333
11.55555556
F
P-value
F crit
0.461538 0.650963 5.143253
Total
80
8
AC10
AC14
AC20
Anova: Single Factor
109
Effect of Rainfall Intensity on Surface Type
ANOVA of Rainfall Intensity on AC10
30mm/hr
202.5mm/hr
0%
85
84
2%
86
85
4%
87
86
6%
88
87
8%
90
88
10%
92
91
432.4mm/hr
82
84
85
85
88
89
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
Count
Sum
6
6
6
ANOVA
Source of Variation
Between Groups
Within Groups
SS
18.77777778
98.33333333
Total
117.1111111
df
Average
Variance
528
88
6.8
521 86.83333333 6.166667
513
85.5
6.7
MS
F
P-value
F crit
2 9.388888889 1.432203 0.269631 3.68232
15 6.555555556
17
110
ANOVA of Rainfall Intensity on AC14
30mm/hr
202.5mm/hr
0%
86
84
2%
87
85
4%
89
87
6%
91
90
8%
93
92
10%
94
93
432.4mm/hr
82
84
87
89
91
92
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
ANOVA
Source of Variation
Between Groups
Within Groups
Total
Count
Sum
6
6
6
SS
540
531
525
df
Average
Variance
90
10.4
88.5
13.9
87.5
15.5
19
199
MS
2
9.5
15 13.26666667
218
17
F
P-value
F crit
0.71608 0.504632 3.68232
111
ANOVA of Rainfall Intensity on AC20
30mm/hr
202.5mm/hr
0%
87
85
2%
88
86
4%
91
89
6%
95
92
8%
98
95
10%
99
97
432.4mm/hr
84
85
88
92
94
96
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
ANOVA
Source of Variation
Between Groups
Within Groups
Total
Count
Sum
6
6
6
SS
32.33333333
368.1666667
400.5
df
Average
Variance
558
93
26
544 90.66666667 23.46667
539 89.83333333 24.16667
MS
F
P-value
F crit
2 16.16666667 0.658669 0.531881 3.68232
15 24.54444444
17
112
The Effect of Crossfall on Rainfall Intensity
ANOVA of Crossfall on 30mm/hr Intensity
0%
2%
4%
AC10
85
86
87
AC14
86
87
89
AC20
87
88
91
6%
88
91
95
8%
90
93
98
10%
92
94
99
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
Column 4
Column 5
Column 6
ANOVA
Source of Variation
Between Groups
Within Groups
Total
Count
3
3
3
3
3
3
SS
196.6667
95.33333
292
Sum
Average Variance
258
86
1
261
87
1
267
89
4
274 91.33333 12.33333
281 93.66667 16.33333
285
95
13
df
MS
F
P-value
F crit
5 39.33333 4.951049 0.010865 3.105875
12 7.944444
17
113
ANOVA of Crossfall on 202.5mm/hr Intensity
0%
2%
AC10
84
85
AC14
84
85
AC20
85
86
4%
86
87
89
6%
87
90
92
Average
84.33333
85.33333
87.33333
89.66667
91.66667
93.66667
Variance
0.333333
0.333333
2.333333
6.333333
12.33333
9.333333
8%
88
92
95
10%
91
93
97
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
Column 4
Column 5
Column 6
ANOVA
Source of Variation
Between Groups
Within Groups
Total
Count
3
3
3
3
3
3
SS
200
62
262
Sum
253
256
262
269
275
281
df
MS
F
P-value
F crit
5
40 7.741935 0.001834 3.105875
12 5.166667
17
114
ANOVA of Crossfall on 432.4mm/hr Intensity
0%
2%
AC10
82
84
AC14
82
84
AC20
84
85
4%
85
87
88
6%
85
89
92
Average
82.66667
84.33333
86.66667
88.66667
91
92.33333
Variance
1.333333
0.333333
2.333333
12.33333
9
12.33333
8%
88
91
94
10%
89
92
96
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
Column 4
Column 5
Column 6
Count
3
3
3
3
3
3
ANOVA
Source of Variation
Between Groups
Within Groups
SS
212.9444
75.33333
Total
288.2778
Sum
248
253
260
266
273
277
df
MS
F
P-value
F crit
5 42.58889 6.784071 0.003194 3.105875
12 6.277778
17
115
The Effect of Crossfall on Surface Type
ANOVA of Crossfall on AC10
0%
None
87
30mm/hr
85
202.5mm/hr
84
432.4mm/hr
82
2%
88
86
85
84
4%
89
87
86
85
6%
91
88
87
85
Sum
338
343
347
351
358
365
Average
84.5
85.75
86.75
87.75
89.5
91.25
Variance
4.333333
2.916667
2.916667
6.25
3.666667
2.916667
8%
92
90
88
88
10%
93
92
91
89
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
Column 4
Column 5
Column 6
Count
4
4
4
4
4
4
ANOVA
Source of Variation
Between Groups
Within Groups
SS
122.8333
69
df
MS
F
P-value
F crit
5 24.56667 6.408696 0.001384 2.772853
18 3.833333
Total
191.8333
23
116
ANOVA of Crossfall on AC14
None
30mm/hr
202.5mm/hr
432.4mm/hr
0%
87
86
84
82
2%
89
87
85
84
4%
90
89
87
87
6%
92
91
90
89
Sum
339
345
353
362
369
373
Average
84.75
86.25
88.25
90.5
92.25
93.25
Variance
4.916667
4.916667
2.25
1.666667
0.916667
0.916667
8%
93
93
92
91
10%
94
94
93
92
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
Column 4
Column 5
Column 6
Count
4
4
4
4
4
4
ANOVA
Source of Variation
Between Groups
Within Groups
SS
227.2083
46.75
df
MS
F
P-value
F crit
5 45.44167 17.49626 2.36E-06 2.772853
18 2.597222
Total
273.9583
23
117
ANOVA of Crossfall on AC20
0%
None
89
30mm/hr
87
202.5mm/hr
85
432.4mm/hr
84
2%
91
88
86
85
4%
92
91
89
88
6%
95
95
92
92
8%
98
98
95
94
10%
99
99
97
96
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
Column 4
Column 5
Column 6
Count
4
4
4
4
4
4
Sum
345
350
360
374
385
391
ANOVA
Source of Variation
Between Groups
Within Groups
SS
442.375
74.25
df
5
18
Total
516.625
23
Average Variance
86.25 4.916667
87.5
7
90 3.333333
93.5
3
96.25
4.25
97.75
2.25
MS
F
P-value
F crit
88.475 21.44848 5.24E-07 2.772853
4.125
118
The effect of Surface Type on Rainfall Intensity
ANOVA of Surface Type on 30mm/hr Intensity
AC10
AC14
AC20
0%
85
86
87
2%
86
87
88
4%
87
89
91
6%
88
91
95
8%
90
93
98
10%
92
94
99
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
ANOVA
Source of Variation
Between Groups
Within Groups
Total
Count
6
6
6
SS
Sum
528
540
558
df
Average
Variance
88
6.8
90
10.4
93
26
MS
76
216
2
15
292
17
F
P-value
F crit
38 2.638889 0.104239 3.68232
14.4
119
ANOVA of Surface Type on 202.5mm/hr Intensity
AC10
AC14
AC20
0%
84
84
85
2%
85
85
86
4%
86
87
89
6%
87
90
92
8%
88
92
95
10%
91
93
97
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
ANOVA
Source of Variation
Between Groups
Within Groups
Total
Count
6
6
6
SS
44.33333333
217.6666667
262
Sum
Average
Variance
521 86.83333333 6.166667
531
88.5
13.9
544 90.66666667 23.46667
df
MS
F
P-value
F crit
2 22.16666667 1.527565 0.248989 3.68232
15 14.51111111
17
120
ANOVA of Surface Type on 432.4mm/hr Intensity
AC10
AC14
AC20
0%
82
82
84
2%
84
84
85
4%
85
87
88
6%
85
89
92
8%
88
91
94
10%
89
92
96
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
Count
6
6
6
ANOVA
Source of Variation
Between Groups
Within Groups
SS
56.44444444
231.8333333
Total
288.2777778
Sum
Average
Variance
513
85.5
6.7
525
87.5
15.5
539 89.83333333 24.16667
df
MS
F
2 28.22222222 1.826024
15 15.45555556
17
P-value
F crit
0.19509 3.68232
121
Effect of Surface Type on Crossfall
ANOVA of Surface Type on 0% Crossfall
30mm/hr
202.5mm/hr
432.4mm/hr
AC10
85
84
82
AC14
86
84
82
AC20
87
85
84
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
Count
3
3
3
Sum
251
252
256
ANOVA
Source of
Variation
Between Groups
Within Groups
SS
4.666667
17.33333
df
2
6
Total
22
8
Average Variance
83.66667 2.333333
84
4
85.33333 2.333333
MS
F
P-value
F crit
2.333333 0.807692 0.489078 5.143253
2.888889
122
ANOVA of Surface Type on 2% Crossfall
AC10
86
85
84
30mm/hr
202.5mm/hr
432.4mm/hr
AC14
87
85
84
AC20
88
86
85
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
Count
3
3
3
ANOVA
Source of
Variation
Between Groups
Within Groups
SS
2.888889
11.33333
Total
14.22222
Sum
Average Variance
255
85
1
256 85.33333 2.333333
259 86.33333 2.333333
df
MS
F
P-value
F crit
2 1.444444 0.764706 0.506023 5.143253
6 1.888889
8
123
ANOVA of Surface Type on 4% Crossfall
AC10
87
86
85
30mm/hr
202.5mm/hr
432.4mm/hr
AC14
89
87
87
AC20
91
89
88
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
ANOVA
Source of
Variation
Between Groups
Within Groups
Total
Count
3
3
3
SS
16.66667
9.333333
26
Sum
Average Variance
258
86
1
263 87.66667 1.333333
268 89.33333 2.333333
df
MS
F
P-value
F crit
2 8.333333 5.357143 0.046258 5.143253
6 1.555556
8
124
ANOVA of Surface Type on 6% Crossfall
30mm/hr
202.5mm/hr
432.4mm/hr
AC10
88
87
85
AC14
91
90
89
AC20
95
92
92
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
Count
3
3
3
Sum
260
270
279
ANOVA
Source of
Variation
Between Groups
Within Groups
SS
60.22222
12.66667
df
2
6
Total
72.88889
8
Average Variance
86.66667 2.333333
90
1
93
3
MS
F
P-value
F crit
30.11111 14.26316 0.005248 5.143253
2.111111
125
ANOVA of Surface Type on 8% Crossfall
30mm/hr
202.5mm/hr
432.4mm/hr
AC10
90
88
88
AC14
93
92
91
AC20
98
95
94
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
Count
3
3
3
Sum
266
276
287
ANOVA
Source of
Variation
Between Groups
Within Groups
SS
73.55556
13.33333
df
2
6
Total
86.88889
8
Average Variance
88.66667 1.333333
92
1
95.66667 4.333333
MS
36.77778
2.222222
F
16.55
P-value
F crit
0.003613 5.143253
126
ANOVA of Surface Type on 10% Crossfall
30mm/hr
202.5mm/hr
432.4mm/hr
AC10
92
91
89
AC14
94
93
92
AC20
99
97
96
Anova: Single Factor
SUMMARY
Groups
Column 1
Column 2
Column 3
Count
3
3
3
Sum
272
279
292
ANOVA
Source of
Variation
Between Groups
Within Groups
SS
68.66667
11.33333
df
2
6
Total
80
8
Average Variance
90.66667 2.333333
93
1
97.33333 2.333333
MS
F
P-value
F crit
34.33333 18.17647 0.002843 5.143253
1.888889
127
APPENDIX 10
Penetration Test
Sand Patch Test
Pendulum Calibration
Polished Stone Value Test
Rainfall Simulator
Sample Coring