<%BANNER%>

Evaluation of Rut Resistance of Superpave Fine-Graded and Coarse-Graded Mixtures


PAGE 1

EVALUATION OF RUT RESISTANCE OF SUPERPAVETM FINE-GRADED AND COARSE-GRADED MIXTURES By COLLINS BOADU DONKOR A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ENGINEERING UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Collins Boadu Donkor

PAGE 3

To John Osei-Asamoah, Chief Director of th e Ministry of Roads and Transport—Ghana

PAGE 4

iv ACKNOWLEDGMENTS My sincere thanks go to my Supervis ory Committee Chair Dr. Mang Tia for his directions, support, and encourag ement in seeing me through th is course. To the other members of my Committee, Dr Reynaldo Roque and Dr Bjorn Birgisson, I say thanks for all the things you have done for me during my 16-month stay at UF. To Dr. Drakos, George Lopp, and Tanya Riedhammer, I say thank you for the help you gave during the testing and data analysis phase for this project. To friends at the FDOT Materials Office, Greg Sholar, Howie Moseley, Steve Ross and Salil Gokhale, thanks go to you for allowi ng me to use your data and facilities and also for the help with my data analysis. To all my friends in the University and colleagues at the Materials Department, I thank you and may God bless all of you. Finally to my sister, Linda, cousin, Christina, and brot her-in-law, Peter, it would have been very difficult without your enc ouragement and support. I thank you very much.

PAGE 5

v TABLE OF CONTENTS page LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................ix ABSTRACT......................................................................................................................x ii CHAPTER 1 INTRODUCTION........................................................................................................1 1.1 Background........................................................................................................1 1.2 Research Hypothesis..........................................................................................3 1.3 Objectives..........................................................................................................4 1.4 Scope..................................................................................................................5 1.5 Research Approach............................................................................................5 2 LITERATURE REVIEW.............................................................................................7 2.0 Introduction........................................................................................................7 2.1 Characteristics of Mixture Constituents.............................................................8 2.1.1 Asphalt.....................................................................................................8 2.1.2 Aggregates...............................................................................................9 2.2 Traffic Loading................................................................................................10 2.3 Environmental Effects.....................................................................................11 2.4 Construction.....................................................................................................12 2.5 Rut Measurement.............................................................................................13 2.5.1 Non-contact Laser Height-Sensor Rut Depth Measurement.................13 2.5.2 Differential Rut Depth...........................................................................14 2.5.3 Absolute Rut Depth................................................................................14 2.6 Mixture Response Characte ristics or Parameters............................................15 2.6.1 Indirect Tension Test (IDT)...................................................................15 2.6.2 Servopac SuperPaveTM Gyratory Compaction.....................................16 2.6.3 Gyratory Testing Machine – GTMTM..................................................17 2.6.4 Asphalt Pavement Analyzer...................................................................17 2.7 Accelerated Pavement Testing.........................................................................19

PAGE 6

vi 3 MATERIALS AND METHODOLOGY....................................................................21 3.0 Introduction......................................................................................................21 3.1 Summary of Methodology...............................................................................21 3.2 Test Track Layout............................................................................................22 3.3 Pavement Structures.........................................................................................23 3.4 Asphalt Concrete Mixtures used......................................................................24 3.4.1 Aggregates.............................................................................................27 3.4.2 Asphalt...................................................................................................28 3.5 Test Tracks Construction and Instrumentation................................................28 3.6 Heavy Vehicle Simulator Test Conf iguration and Instrumentation................29 3.7 Laser Profiler...................................................................................................30 3.8 Trafficking.......................................................................................................31 3.9 Heating and Temperature Control...................................................................31 3.10 Rut Measurement.............................................................................................33 3.11 Air Void Content and Thickness Changes.......................................................34 3.12 Determination of Viscosity of Cores...............................................................35 3.13 Laboratory Testing on Plant and Laboratory Prepared Mixtures....................35 3.13.1 Asphalt Pavement Analyzer Test.........................................................36 3.13.2 Servopac Gyratory Compactor Testing...............................................36 3.13.3 SuperPaveTM Indirect Tensile Test....................................................37 3.14 Condition Survey.............................................................................................37 4 RESULTS, ANALYSIS AND DISCUSSION...........................................................39 4.0 Evaluation of Rut Profiles................................................................................39 4.1 Introduction......................................................................................................39 4.2 Rut Depth.........................................................................................................39 4.3 Transverse Rut Profile.....................................................................................42 4.4 Area Parameter Change Method Evalua tion of Transverse Rut Profiles........46 4.5 Evaluation of Core Densities...........................................................................50 4.6 Evaluation of Recovered Asphalt....................................................................53 4.7 Asphalt Pavement Analyzer Test.....................................................................54 4.8 SuperPaveTM Servopac Gyratory Compaction Results....................................57 4.9 Evaluation of Gradation of HVS Track Mixture.............................................62 4.10 Evaluation of SuperPaveTM I ndirect Tensile Test Results.............................66 4.11 Further Evaluation of the Fine-Graded Mixture..............................................68 4.12 Comparison of the Fine-graded Mixt ure of HVS Round 1 to HVS Round 3..69 4.13 Condition Survey Results................................................................................73 5 CONCLUSIONS AND RECOMMENDATIONS.....................................................75 5.1 Conclusion.......................................................................................................75 5.2 Recommendation.............................................................................................76

PAGE 7

vii APPENDIX A MIX DESIGNS...........................................................................................................77 B RECOVERED VISCOSITIES...................................................................................81 C AREA CHANGE PARAMETER...............................................................................87 LIST OF REFERENCES...................................................................................................99 BIOGRAPHICAL SKETCH...........................................................................................102

PAGE 8

viii LIST OF TABLES Table page 3.1. Job mix formula and truck samples grada tions, volumetric prope rties and air voids for fine-graded mixture bottom lift..........................................................................25 3.2. Job mix formula and truck samples grada tions, volumetric prope rties and air voids for the fine-graded mixture top lift...........................................................................25 3.3. Job mix formula and truck samples grada tions, volumetric prope rties and air voids for the coarse-graded mixture bottom lift................................................................26 3.4. Job mix formula and truck samples grada tions, volumetric prope rties and air voids for the coarse-graded mixture top lift.......................................................................26 3.5. Specification for the asphalt PG 67-22 used for the mixtures....................................28 4.1. Area change parameters of the fine -graded and coarse-graded sections....................48 4.2. Air voids level of the cores from the wheel paths and edges of the wheel paths lane 3 fine-graded mixture sections.........................................................................51 4.3. Air voids level of the cores from the wheel paths and edges of the wheel paths of lane 5, coarse mixture...............................................................................................52 4.4. Viscosity of recovered asphalt....................................................................................53 4.5. Asphalt pavement analyzer rut depth for both fine-graded and coarse-graded mixture.....................................................................................................................55 4.6. Indirect tensile test parameter versus servopac test parameters..................................67

PAGE 9

ix LIST OF FIGURES Figure page 3.1 Show the layout accelerated pa vement testing test track.............................................23 3.2. Pavement structure for the test lanes w ith fine-graded and coarse-graded mixtures..24 3.3. Job mix formula gradation curves of fi ne-graded and coarse-graded mixtures..........27 3.4. Heavy vehicle simulator te st carriage and lasers........................................................30 3.5. Longitudinal rut imprints on section 3-3B..................................................................31 3.6. Thermocouple assembly on track 3-2A......................................................................32 3.7. Heavy vehicle simulator with its insulation assembly................................................33 3.8. Location of cores to be taken af ter the HVS runs on section 3-3B.............................34 4.1. Plot of change in rut depth ve rsus number of HVS wheel passes..............................40 4.2 Maximum differential rut depths for the sections........................................................41 4.3 show the maximum absolute rut de pths for all the test sections..................................41 4.4 Differential transverse profiles at 100 passes of the fine-graded mix.........................43 4.5. Differential transverse profiles at 100 passes as compared with one of 90000 passes for the fine-graded mixture...........................................................................43 4.6. Differential rut profiles at 100 passe s of the coarse-graded mixture..........................44 4.7. Differential transverse profile at 100 pa sses as compare with one the 90000 passes for the coarse-graded mixtures.................................................................................44 4.8. Evolution of transverse profile of fine-graded mixture section 3C............................45 4.9. Evolution of transverse profile of coarse-graded mixture section 5A........................46 4.10. Initial and final surface profile of a fine-graded section 3-3C..................................47 4.11. Initial and final surface profile of a coarse-graded section 3-5A..............................47

PAGE 10

x 4.12. Positive A1 and negative A2 areas of a transv erse rut profile...................................49 4.13. Asphalt pavement analyzer comparisons and HVS rut depths for the fine-graded and coarse-graded mixture of the top lift.................................................................56 4.14. Aaverage gyratory shear stress versus number of gyrations for bottom lift.............57 4.15. Average gyratory shear st ress versus number of gyr ations for the top lift...............58 4.16. Gyratory shear stress versus number of gyrations of the servopac compactor for section 3C fine-graded section.................................................................................59 4.17. Gyratory shear stress versus number of gyrations for 3-3C with a change of gyration angle from 1.25 to 2.5 .............................................................................59 4.18. Gyratory slope and initial failure strain of the top lift fine-graded and coarsegraded mixtures........................................................................................................60 4.19. Pass/Fail criteria for evaluation of rut resistance......................................................61 4.20. Differential rut depths vers us gyratory slope of the co arse-graded mixture of the top lift....................................................................................................................... 61 4.21. Differential rut depths vers us gyratory slope of the fi ne-graded mixtures of the top lift....................................................................................................................... 62 4.22. Gradation of the fine-graded and coarse-gra ded plant mix mixtures of the top lift.63 4.23. Interaction unit check for lane 1 and s ection 2A of lane 2 of the fine-graded mixture top lift..........................................................................................................64 4.24. Interaction unit check for lanes 2 and 3 for the fine-graded mixture top lift............64 4.25 Porosity of the fine-graded mixture for interaction and no interaction.....................65 4.26 Unit interaction plot for the coarse-graded mixture of the top lift.............................66 4.27. Relationship between gyratory shear slope and the energy ratio of both the finegraded and coarse-graded mixtures..........................................................................68 4.28. Comparison of fine-graded and coar se-graded differential rut depths.....................69 4.29. Job mix formula HVS Round 1 and HVS Round 3 for the fine-graded mixture......70 4.30. Differential rut depth of the fine-g raded mixture HVS Round 1 and HVS Round 3.............................................................................................................................. ..70 4.31. Interacting unit check for “JMF” “HVS Round 1 and 3” and lanes 4 and 5............72

PAGE 11

xi 4.32. Porosity of job mix formula for “HVS Round 1” and “HVS Round 3” finegraded mixture..........................................................................................................72 4.33. Longitudinal crack on section 3B.............................................................................73 4.34. Strain gage induced crack on the top lift of section 3-3B fine-graded mixture........74 A-1 Gradation and sources of aggr egate for the coarse mixture.......................................77 A-2. Hot mix design data sheet for coarse-graded mixture...............................................78 A-3. Gradation and sources of aggreg ate for the fine-graded mixture..............................79 A-4. Hot mix data sheet for the fine-graded mixture.........................................................80 B-1. Viscosity of recovered as phalt section 3A bottom lift...............................................81 B-2. Viscosity of recovered as phalt section 3B top lift.....................................................82 B-3. Viscosity of recovered as phalt section 3C top lift.....................................................83 B-4. Viscosity of recovered as phalt section 5B top lift.....................................................84 B-5. Viscosity of recovered as phalt section 5A top lift.....................................................85 B-6. Viscosity of recovered as phalt section 5C bottom lift...............................................86 C-1. Area-change parameter section 2A............................................................................87 C-2. Area-change parameter section 2B............................................................................88 C-3. Area-change parameter section 2C............................................................................89 C-4. Area-change parameter section 3A............................................................................90 C-5. Area-change parameter section 3B............................................................................91 C-6. Area-change parameter section 3C............................................................................92 C-7. Area-change parameter section 4A............................................................................93 C-8. Area-change parameter section 4B............................................................................94 C-9. Area-change parameter section 4C............................................................................95 C-10. Area-change parameter section 5A..........................................................................96 C-11. Area-change parameter section 5B..........................................................................97 C-12. Area-change parameter section 5C..........................................................................98

PAGE 12

xii Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Engineering EVALUATION OF RUT RESISTANCE OF SUPERPAVETM FINE-GRADED AND COARSE-GRADED MIXTURES By Collins Boadu Donkor December 2005 Chair: Mang Tia Major Department: Civil and Coastal Engineering Two gradation specifications were develope d as part of the Superior Performing Pavement Program in 1993. They are known as coarse and fine gradations respectively. It was recommended to use the coarse grad ation in order to ac hieve better rutting resistance, however many researchers believe that either of the two gradations will do well in rutting resistance if they are properly designed and constructed. This research focuses on evaluating the rut resistance of a fine-graded mixture, as compared to a coarse-graded mixture in use in the State of Florida. Both field and laboratory evaluation me thods were used. An Accelerated Pavement Facility was constructed at the St ate Materials Office in Gainesville, Florida consisting of 6 sections of a typical fine-g raded and a typical coar se-graded mixture in use in Florida. The loads were app lied via a heavy vehicle simulator (HVS) manufactured in South Africa.

PAGE 13

xiii Asphalt Pavement Analyzer (APA) rut depths, Servopac Gyratory Compaction tests, Indirect Tensile Te st (IDT), and Bulk Density tests were performed on the specimens of the fine-graded and coarse-graded mixtures. Analyses of the results of both the differen tial and absolute rut depths show that the difference in the mean rut depths of the coarse-graded and fine-graded mixture was statistically insignificant. From the limited data collected we can not conclude that the SuperPaveTM coarse-graded mixture is either be tter or worse than the SuperPaveTM fine-graded mixture.

PAGE 14

1 CHAPTER 1 INTRODUCTION 1.1 Background Rutting is a world-wide performance pr oblem associated with Hot Mix Asphalt (HMA) mixtures. It manifests itself as a longitudinal bowl-like surface depression in the wheel paths on flexible pavements with the app lication of vehicular loads. The gradual and progressive reduction of wheel path laye r thickness leads to functional as well as structural deficiency of the pavement. Rutting can be due to instability caused by inadequate shear resistance and or de nsification in the HMA mixtures. Where pavement structure is inadequate rutting could occu r as a result of permanent deformation of the subgrade or granular base and subbase materials. Rutting creates safety, functional as well as structural problems on HMA pavements. A 10 mm rutting is likely to cause hydroplaning on pavements at 40 mph traffic speed. Development of depressi on and shoves increase the surface roughness resulting in higher vehi cle operating cost. Higher levels of rutting cause reduction in layer thickness which reduces the load spreading ability of the pavement. Large te nsile stresses could develop at the bottom or on top of the surface layer causing cracking in HMA. Low shear resistance associated with rutting of HMA mixtures have been f ound to create shear planes (Birgisson et al. 2002). Such planes of weakness have induced crack development in HMA layers. Topdown longitudinal cracking have also developed as result of high tensile stress at crowns of the shoves created by permanen t deformation of HMA layers.

PAGE 15

2 Rutting being a performance measuring parame ter is associated with traffic loading as well as mixture quality. Two differentl y designed mixtures of the same pavement thickness, and loaded the same way under the same environmental conditions may behave differently in rut performance. Mixture differences could be exhibited by gradation, aggregate texture, binder grade and content, dust to effective binder ratio as well as mixture volumetrics, voids in mineral aggregates (VMA), voids filled with asphalt (VFA), and air voids (AV). The State of Florida has experienced some amount of rutting of varying extent and degrees of severity on at least two pavements constructed with SuperPaveTM coarse mixtures since its implementati on in 1995. The Coarse SuperPaveTM mixtures are the dominant HMA used in Florid a. It accounts for over 75% of all HMA construction in the state interstate, and heavy traffic highways. Properly designed SuperPaveTM coarse mixtures have been mandated by the Florida Department of Transport (FDOT) for heavy-traffic, HMA construction because of perceived better rut resistance than the fi ne-graded mixtures, complaints from both Contractors and FDOT Engineers on the difficulties associated with its use, have called for the inclusion of the fine-graded SuperPaveTM mixtures for heavy traffic and Interstate Highways. Contractors str uggle to meet minimum voids in mineral aggregate (VMA) specifications, especially when using aggregat es native of Florida (Sholar et al. 2004). The requirements of VMA in coarse-graded SuperPaveTM mixtures have even been called into question in several research projects (Nukunya et al. 2001, Anderson et al. 2001, and Kandhal et al. 1999) which indicated that more serious work remains to be done.

PAGE 16

3 Coarse-graded mixtures are difficult to co mpact, leading to large field-air voids and hence high densification and higher rut de pth, as exemplified in the Wes Track experience. It segregates easily causi ng raveling and moisture-induced damage and ultimately reduction of pavement performance. Permeability of HMA mix layers are influenced by the level of compaction. Lower level of compaction increases the permeability allowing air and moisture ingress that affects the durability of the mix. Fine-graded SuperPaveTM mixtures do not suffer such difficulties associated, with graded-coarse SuperPaveTM mixtures however, they are le ss attractive on the basis of their higher rut susceptibility. This research evaluates the longterm rut resistance of SuperPaveTM fine-graded mixtures used in Florida and makes comp arison to the predominant coarse-graded mixtures in use in the State. The H eavy Vehicle Simulator (HVS) acquired by the FDOT Materials Office and the constructe d Accelerated Pavement Testing (APT) Facility was used in this study. The HVS applied on a full-scale accelerated testing facility can be used to evaluate rut resist ance of mixtures in s ituations where knowledge of long-term performance is not available. The HVS can simulate 20 years of interstate traffic on a test pavement within a short period of time (Tia et al. 2002). 1.2 Research Hypothesis The resistance to shear deformation, which is a measure of rut resistance in HMA mixture, has been attributed mainly to inte r granular contact fric tion and interlocking of the coarse aggregate particles. Under ne ar-surface, low-confinement loading conditions, coarse-graded SuperPaveTM mixtures will out perform fine-graded SuperPaveTM mixtures in rut resistance.

PAGE 17

4 1.3 Objectives There are three main objectives in this research. Primarily, it is focused on evaluating the performance of a typical fine-graded and a typical coarse-graded SuperPaveTM mixture with respect to rutting under the Florida climate. The second objective is to quantify rut in terms of aver age rut depth using laser-captured progressive transverse profiler. This research will also evaluate performance predicting models developed at the University of Florida. Th e models include the “pass/fail criteria” to predict pavement performance based on the SuperPaveTM gyratory compaction, indirect tensile test (IDT) and asphalt pavement analyzer (APA) ch aracteristics of mixtures. Hot mix asphalt (HMA) mixtures densify upon the application of wheel loads from as built air voids contents of 7% to end of design life air void content of 4%. With further application of loads after the 4% air voids content, a good mix can mobilize enough shearing resistance to coun teract the shearing and tensile stress that causes rutting at the near-surface and low-confinement of the HMA layer. High-shear and tensile stresses at the near-surface create shear planes of changing angles (Darku and Birgisson. 2003) in rut susceptible mixtures resulting in instability rutting of the HMA layer. The HVS applied unidirectional radial tire lo ad of 9 kips with four inch wander of 90000 passes to both fine-graded and coarse-g raded test sections. A continuous progressive rut depth measurement will be captu red using a laser profiler attached to the wheel. It is expected that, the result of this study will show clearly which of the two typical (fine and coarse) Florida mi xtures have better rut resistance.

PAGE 18

5 1.4 Scope The scope of this research is to (1), ev aluate the rutting performance of a typical Florida SuperPaveTM fine-graded and co arse-graded mixture using the HVS. (2), Conduct a thorough literature review of f actors affecting rutting performance of finegraded and coarse-graded mixtures and the expe riences of some States with Florida-like climatic conditions, like Texas and Alabama wh ich have switched from coarse-graded to fine-graded SuperPaveTM mixture and (3), analyze the two mixtures, focusing on their rutting-resistance characteristic s using laboratory test results and performance predicting models developed at the University of Florida. The analyses of the test results will be focused on evaluating these performance models or index tests to accurately a nd reliably measure a mixture response characteristics or parameter that is highly correlated to the occurrence of pavement rutting over a diverse range of traffic and clim atic conditions. This will help to predict long-term performance of different HMA mi xtures without havi ng to conduct full-size long-term APT experiments, whic h are expensive to perform. 1.5 Research Approach In order to evaluate the fine-graded SuperPaveTM mixture for interstate traffic and also make comparison to the coarse-graded mixture for rut performance, an APT facility was designed and implemented at the FDOT office in Gainesville. The following activities were executed: A Heavy Vehicle Simulator (Mark IV) was used to apply a 9 kip single-wheel 115 -psi load, via a radial tire traveling at 8 mph on carefully c onstructed pavement lanes. Previous research results (Tia et al. 2002) showed that the effective way of inducing

PAGE 19

6 rutting in APT using HVS was to apply a 9 ki p wheel load in a unidirectional mode with 4 inches wander of 1 inch incremen ts via a super single radial tire. Longitudinal and transverse rut depths we re measured using two laser profilers mounted on the axis on each side of the single wheel of the HVS. Analysis of the rut profiles was performe d to evaluate rut resistance performance of the mixtures to determine which type of rutting was predominant (densification or shear instability rutting). The area-change parameter (A CP), is a physical methodology that can be used to determine whether rutti ng is primarily due to shear instability or because of densification. Cores were extracted from the wheel pa ths and along the immediate edge of the wheel paths for density and thickness measur ement. The viscosity of the asphalt recovered from cores taken from the inside a nd edge of the wheel paths were determined to evaluate the effects of grad ation, environment and traffic. The mixtures were tested and the da ta analyzed using these methodologies: (i) The University of Florida’s rutting fr amework which is a “pass/fail criteria” relating the slope and the vert ical-failure strain of the gyration curve, using the SuperPaveTM gyratory compactor. (ii) The University of Florida Energy Ratio “pass/fail” framework for cracking performance. Mixtures that exhibit greater resistance to rutting may behave poorly in cracking resistance. All th e mixtures were tested for tensile strength, and their fracture energy a nd failure strain parameters were evaluated. (iii) The APA test was used as rut distingui shing tool as a “pass/fail” framework for both mixtures.

PAGE 20

7 CHAPTER 2 LITERATURE REVIEW 2.0 Introduction The Superior Performing Pavements (SuperPaveTM) mixture design system was introduced as part of the Strategic Highway Research Program (SHRP) to replace the Marshall and other mixture design procedures 1993 as a rational mix design procedure. The main objectives of the SuperPaveTM was to provide mixtures with better resistance to rutting, fatigue, low temperature cracking a nd moisture induce damage. Two types namely “Coarse and Fine” mixtures with diffe rent gradation charac teristics are in use today. Originally a coarse-graded mixture defined as having gradation passing below the Restricted Zone (RZ) whilst the fine-gra ded mixture gradation passes above the RZ. Many researchers have evaluated the rut resi stance of mixtures. It has been held for a considerable period of time that the coarse-graded mixtures will out perform the fine-graded mixtures in rut resistance. Fo r instance, stone matrix asphalt is known to have excellent rutting resistan ce. However many researcher have found out that there is no difference in rut resistance between co arse-graded and fine-graded SuperPaveTM mixtures. (Kandhal and Cooley. 2002) did not find any significant differences in rutting resistance between coarse and fine Superpav e mixtures. (Sargand and Kim. 2003) using APA rut-depth analyses concluded that ne ither gradations passing above, through or below the restricted zone was signi ficant in affecting rut-depth. There are many factors that account for the behavior of mixtures during rutting, not just a matter of the mixture being either a coarse-graded or a fine-graded only.

PAGE 21

8 Rutting of HMA mixtures is affected by many factors such as: (1) Characteristics of mixture constituents. (2) Traffic loading (3) Environmental effects and (4) Construction. 2.1 Characteristics of Mixture Constituents Asphalt concrete is composed of about 95% aggregate and 5% asphalt compacted at elevated temperatures to low air voids. Because the aggregates are subjected to crushing and abrasive wear during manuf acture, placing and compaction, they are generally required to be ha rd, tough, strong and durable wi th cubical shape, of low porosity, rough textured and proper gradation. To bind the aggregates together to form mixture to perform under traffic loading, the asphalt not only should be stiff enough to resist permanent deformation, should be also flexible enough to resist fracture in cold weather. 2.1.1 Asphalt Both the amount and grade of asphalt in th e mixtures influence rutting potential of a mixture. Stiffer binder and hence higher G* values increase the resistance to rutting the asphalt mixture. (Cort 2001) noted that G* varies from a ratio of 1 to 2 for the same test conditions and that the sensitivity to permanent deformation as indicated by G*/sin is definitely different from one asphalt to another. There is a limit to increasing asphalt hardness to control rutting. S tiffer binders have increase br ittleness at low temperatures and thus lower its healing capacity. Mixtures with binder conten t on either side of the op timum, impacts negatively on the permanent deformation characteristics of a mixture. Lower than optimum binder content results in increase air voids, lowe r cohesion due to lower film thickness and lower shearing resistance and higher permanen t deformation with load application.

PAGE 22

9 Excessive binder content produces in a mix cr eates pore pressures te nding to break the interlocking of the aggregate particle and results in instability and high permanent deformation. 2.1.2 Aggregates There are a whole lot of a ggregate properties that aff ect the rut resistance of mixtures. It is the aggregate properties that must provide the support to resist permanent deformation (NCAT, 1996). The aggregat e texture, angularity, nominal maximum size and gradation are critical to good rutting resistan ce of a mixture. Adhesion and cohesion of mixtures are influenced directly by aggr egate texture. Angul ar and rough textured aggregates will provide str onger frictional, bonding and inte rlocking forces to resist rutting than smooth, and rounded aggregate even if rounded aggr egates compacts better. Gradation is the most important parameter for rut susceptibility. Gradation affects the stiffness and frictional resistance of HMA mixture. The proportion and effective size of aggregates passing the 0.075 mm sieve contro ls the mass viscosity of the asphalt that surrounds the coarse aggregates. As the ma ss viscosity increases w ith increasing filler content, the mix become stiffer, increasing its ability to resist pe rmanent deformation. Increase in the nominal maximum aggregate size and the percentage of coarse aggregates increases the volume concentra tion of aggregates with more rutting resistance ability. (Stuart and Mogawer 1997) concluded that under accelerated loading conditions, increase in nominal maximum aggregat e size significantly decreased rutting susceptibility. Various DOT’s including the FDOT have limits on the percentage passing #200 sieve and the nominal maximu m size of aggregates for SuperPaveTM mixtures.

PAGE 23

10 It has been a long held view among resear chers and practitioners that generally coarse-graded SuperPaveTM mixtures have better rutti ng resistance than finer-graded SuperPaveTM mixtures. After the Wes Track experiment, this view point has been discarded. Darku (2002) observed that, it is the mixture’s ability to di late during shearing that controls its rutting potential. Increasing ability of a mixture to dilate reduces it s vertical strain thus lower rutting under loads. In fact neither coar se-graded nor fine-graded Superpave mixtures have superior rutting resistance. Rece nt studies (Kandhal et al. 2002) and (NCHRP Project 9-14) have shown that the mechanic al properties of HMA appear to be more sensitive in coarse-graded than in fine-grade d mixtures. Coarse-graded, fine-graded or mixes passing through the restricted zone may perform well under various traffic and environmental conditions. Roque (1997) poi nted out that good shearing resistance can be achieved with a broad range of aggregate structures as long as suitable gradations are used. However as noted again by Roque (2002), there is no clear-cut method of selecting an aggregate grad ation to produce good mixtures. 2.2 Traffic Loading Traffic load induced stress is the major cau se of pavement distresses apart from climatic conditions. These stresses are a functi on of wheel loads, tire pressure and type as well as thickness and stiffness of the laye rs. Collop and Cebon (1995) concluded that both dynamic vehicle loads and asphalt layer stiffness variation can have significant influence on long-term flexible pavement performance. Ullidtz and Larsen (1983) published a model which predicted the perfor mance of flexible in terms of roughness, rutting and cracking as a functi on of traffic loading and climate. With the introduction of radials tires, many States in the Unite d States and countries around the world over

PAGE 24

11 have seen significant instability rutting in HMA. The effect of tire type (radial or bias ply) could have significantly influence on how HMA ruts. Bigirsson and Roque (2002) have shown that the high transverse near-sur face stress at low confinement in the vicinity of the edges of radial tires may partly explai n the mechanism of inst ability rutting. The history of tire pressures on HMA rutting is well known. Kandhal et al (1990) traced the changes of truck tires pressures from 70 psi to 85 psi during the AASHTO Road Test and subsequent increases in States like Virginia, Florida, Texas and Illinois with averages of 96 to 110psi and maximum in the order of 155 psi. There are no lega l limits against tire pressures in the United Stat es (Kandhal et al. 1990). 2.3 Environmental Effects The effects of temperature on rutting have been investigated by many researchers. Robertson (1997) observed that the shape of the temperature profile in a pavement is dependent on the air temperature history. Summer time pavement temperatures ranged for most countries of the world around 60 to 70 C. Asphalt concrete being a thermosusceptible material will at warm enough temp eratures develop signi ficant visco-plastic strains under wheel loads. The accumulation of large visco-plastic strains results in permanent deformation of the pavement. Collop and Cebon (1983) observed that changes in the temperature of the asphalt laye rs affect the elastic and viscous properties of the asphalt. Matthews and Monismith (1992) using Hveem Stabilometer and creep test results concluded that temperature has mo re influence on rutting than the aggregate grading. As asphalt oxidizes under temperatur e and air, it looses its flexibility thus increases its stiffness with increasing rutti ng resistance but lose it cracking resistance.

PAGE 25

12 2.4 Construction Constructional inaccuracies and vari ability impact negatively on pavement performances. Villiers (2004) showed that variation in dust content and/or asphalt content that were perfectly acceptable under the current practice produced significant loss in performance or unacceptable cracking perf ormance. Hot mix asphalt construction starts with mixture design and is follow ed sequentially with production, laying and compaction. Poor mixture designs have dir ect bearing on mixture performance. For instance too well graded aggregate grada tion leads to low VMA and lower than acceptable asphalt content. R ogue (1997) stated that mixtur es that are poorly designed, produced, and constructed can result in rutting du e to plastic deformation. Mixtures with too much rounded natural sand particles have resulted in tender mixes and poor rutting performance (Buchanan and Cooley 2002). Inaccurate production or lack of proper production control of mixtures leads to paveme nts of variable performance. Nouredin (1997) concluded that, the overall pavement performance life may be significantly affected when the specific asphalt content, aggregate gradation (job mix formula) and the degree of compaction are not achieved in si tu. The success of any HMA pavement construction lies with careful and accurate laying operation. Mat thickness, segregation and moisture induced damage control results from proper and efficient laying operations. Stroup-Gardiner et al (2000) re ported that when there are higher levels of segregation, the failure mode shifts from fatigue to compre ssion i.e. rutting. A du-Osei et al (1999) concluded that construction variability can have signifi cant adverse effect on the performance of mixtures. Compaction is to reduce the air void leve l of the mixture to increase the aggregate to aggregate contact a nd interlock thus, increase mixture resistance to shearing forces. Excessive air voids due to inadequate compacti on increase rut depths

PAGE 26

13 on application of traffic loads. It should be noted that, for adequa te rutting resistance, stone-to-stone contact is critical. HMA laye rs consolidate more w ith higher initial air voids content. Harvey and Popescu (2000) showed that good construction compaction helps to reduce the amount of densification th at occurs under traffi cking and reduces the amount of rutting caused by densification. Pe terson et al (2004) observed that achieving proper compaction of asphalt pavement is crucial to its longev ity and acceptable performance. 2.5 Rut Measurement Accurate rut measurement is important sa fety consideration on our roads ways. Rut depth of over 10 mm provides considerable ri sk to motorist in either rainy or icy weather. Hydroplaning caused by pond or slic kly ice sheet in longitudinal rut depression jeopardizes the safety of motoring public. Rutting has been known to contribute insignificantly to longitudinal r oughness in pavement. In fact the contribution of rutting to serviceability is in order of -1.3(RD)2 where RD is rut depth in inches and very small compared to surface roughness in IRI. However rut depth of 0.5 inches could have considerable safety as well as structural im plications. It is important therefore to measure and quantify rutting accurately in orde r to select an appropr iate repair strategy. 2.5.1 Non-contact Laser Height-Sensor Rut Depth Measurement Advancement in technology has provided engineers with a fa st, efficient and automated rut depth measurement using laser non-contact height-sen sors called laser profilers. Traditionally rut depth measuremen ts using strings and or straight edges and wedges have been dangerous and also time cons uming. It is expensive when used for either project or network leve l evaluations and usually imprac tical to measure at regular intervals. Gokail (2003) observed the impracticality of measuring ruts conventionally

PAGE 27

14 when using the HVS under APT applications The space restrictions underneath the HVS loading system will not allow accurate rut data collection with conventional methods. The non-contact laser measuring systems have rendered these manual methods obsolete in many countries. The non-c ontact method consists of measuring the transverse profile by digitizing the pavement surface and then analyzing the data to calculate the rut depth using a simulated stra ight-edge at normal traffic speeds. The distance between the simulated straight-edge and the lowest point along the transverse profile is calculated as the rut depth. Ti a (2002) showed that tw o very different rut measurements can be calculated using the tr ansverse surface profile out from the laser profiler. The “differential rut depth” and “absolute rut depth” can be calculated from transverse surface profile. 2.5.2 Differential Rut Depth When a straight line is drawn to touch the two highest point of the differential surface profile obtained by subtracting the initi al surface profile from the surface profile after some trafficking, the greatest distan ce between the straight line and differential surface profile is the “differen tial rut depth”. The functi on of this parameter is to incorporate the instability charac teristics of the material into the rutting prediction Drakos (2002) and includes the dilation portion of the deformed material into the rut measurement. 2.5.3 Absolute Rut Depth Absolute rut depth is the difference betw een the lowest points on the initial surface profile and the surface profile af ter some period of trafficki ng. Both the initial and as trafficked rut depths are measured the same wa y. A straight line is drawn to touch the highest points on the surface profile and the gr eatest distance to th e lowest point on the

PAGE 28

15 surface profile measured as the rut depth. V illiers (2004) concluded that absolute rutting cannot be used as a measure of mixture perf ormance. One must evaluate each section carefully to assess the contri bution of different layers. 2.6 Mixture Response Charac teristics or Parameters Several mixture response characteristics or parameters exist for both laboratory as well as field evaluation of HMA. Perman ent deformation and fatigue response have been measured in the laboratory using the Servopac SuperPaveTM Gyratory Compaction (SGC), Asphalt Pavement Analyser (APA), SuperPaveTM Indirect Tension Test (IDT) characteristics and the Gyratory Testing M achine (GTM). The slope and the vertical failure strain parameters from the Servop ac SGC as proposed by Bi rgisson et al (2002) and the APA rut depth are tools for rut depth evaluation. Low m-value and high Dissipated Creep Strain Energy (DCSE) have been known to associate higher fracture resistance as calculated from the stress-stra in curve of HMA under st ated temperature and time of loading (Roque and Butler 1997). Nukunya (2002) evaluated mixture rutting resistance using IDT creep results. The ratio of initial to final gyration angle as measured with the GTM has been used by many researchers to evaluate mixture instability. 2.6.1 Indirect Tension Test (IDT) Top – down cracking is the primary mode of pavement distress in Florida. Approximately 80% of the State’s deficient highways are due to top – down cracking Sholar (2004). Mixtures that exhibit high resistance to rutting may exhibit high propensity to fracture because of its excessive stiffness and low healing potential at low and intermediate temperatures. Such mixtur es may exhibit high creeping compliance. Kim (2003) showed that there is a direct relationship between the rate of creep and the

PAGE 29

16 rate of micro-damage accumulation while inve stigating the effect of Styrene butadiene styrene (SBS) on coarse SuperPaveTM fractur e potential. The SuperPaveTM IDT tests can be used to measure the diametric Res ilient Modulus (MR), Creep Compliance (CC) and Tensile Strength (TS) of HMA. (R oque et al. 1999) developed a framework for evaluating the fracture properties of HMA us ing Energy Ratios (ER) from calculated Dissipated Creep Stain Energy (DCSE), Fracture Energy Limit (FE), and MR. 2.6.2 Servopac SuperPaveTM Gyratory Compaction The Servopac Super pave Gyratory Compact or (SGC) has excell ent versatility in what parameters it could measure and also th e ease which control parameters could be changed to simulate field conditions. Measurement during compaction such height, density, air voids and shear strength can be obtained with this compactor. The ultimate field and initial densities of a mi x can be simulated using the SGC. (Birgisson and Darku 2002) proposed a “Pa ss – Failure” framework for evaluating mixtures rutting potential based on the Sl ope of the gyratory shear against number gyrations and the vertical failure strain. The minimum acceptable gyratory shear slope is 15 kPa and the optimum range for good rutti ng resistance is betw een 1.4% and 2.0% for vertical failure strain. Mixt ure with failure strain less th an 1.4% are considered brittle whilst that greater than 2.0% are considered plastic. Instability rutting is manifested in a r earrangement of the aggregate structure (Birgisson and Darku 2002). This rearra ngement is produced during compaction by changing the gyration angle of the Servop ac SGC from 1.25 to 2.5 at 7% air voids content of the mixture. The straining of the mixture during this change of gyration angle and the degree of rearrangement afterwards gi ves an indication of the mixtures rutting resistance in the field.

PAGE 30

17 The Vertical failure Strain is the ratio of the difference in specimen height at maximum shear stress to the height of speci men at the first minimum shear stress after the change of angle of gyration from 1.25 to 2.5 2.6.3 Gyratory Testing Machine – GTMTM The Gyratory Testing Machine (GTMTM) was developed by the Army Corp of Engineers in the late 1950 at the Waterways Experimental Station. It is a compactor as well as a tester. The GMTTM combines vertical pressure and shear displacement to simulate field roller compaction and future traffic densification thus, mixture properties measured during compaction and densification with the GMTTM could be used to evaluate mixture performance. The measur e of mixture’s stability is the Gyratory Stability Index (GSI) and is related the mixture’s resistance to rutting. The maximum gyration angle divided by the in itial gyration angle is the GSI. GSI values of more than 1.1 indicate instability whilst va lues close to 1 shows the mixt ure is stable. The GTM is the only device capable of monitoring change s in mixture response with densification. Moseley (1999) concluded that the GTMTM give a good indication of mixture performance by measuring shear resistance. 2.6.4 Asphalt Pavement Analyzer The Asphalt Pavement Analyzer (APA) was a development from the Georgia Load Wheel Tester to evaluate rutting and moisture susceptibility of HMA mixtures in the mid 1990s. Asphalt mixtures compacted mainly in the SGC are placed in mold and subjected to 8000 cycles of loaded 100 psi pressure hos e in an environmental chamber maintained at 64 C.

PAGE 31

18 In Florida, mixtures that exhibits rut depth of over 8 mm after the 8000 cycles are considered unacceptable for field rutting performance. Rut depth is measured as the difference between lowest point before and after the 8000 cycle application. Various researchers have used the APA to characterize HMA mixtures. Research at NCAT showed the APA was sensitive to mixtures with different asphalt binder and varying gradation. Thus coarse-graded and fine-graded mixtures would show different results. This observation is not shared by the FWHA at Turner-Fairbanks which compared APT and Load Wheel Test (LWT) re sults for rutting. They concluded that none of the LWT could distinguish between good and poor performing mixtures which, was clearly distinguished by the APT. Sargand and Kim (2003) using APA rut-depth analyses concluded that neith er gradations passing above, through or below the restricted zone was significant in affecting rut-depth. Drakos (2002) showed the APA hose does not capture the critical lateral stresses found to be detrimental to rutting as well as cracking of HMA pavement and modified the loading mechanism from a hose to a strip know as the New APA to simulating stress distribution in a single radial tire rib. Drakos (2002) agai n showed that neither the New APA nor the original APA could distinguish between coarse-graded and fine-graded SuperPaveTM mixture performance at 4% air voids and at 64 C. They reported significant difference in performance at 70 C using the New APA at the same air void. Rut depths were measured using Absolute Rut Depth Method.

PAGE 32

19 2.7 Accelerated Pavement Testing The history of Accelerated Pavement Testi ng (APT) dates back to late 1900s when Michigan implemented the first APT facility with the view to determine pavement response and performance under a controlle d, accelerated accumulation of damage. Accelerated pavement testing offers e normous potential for studying pavement distress mechanisms and for evaluating perfor mance of asphalt mixtures and pavement in relatively short period of time, (Roque and Tia 2005). (Lea and Heath 1997) have observed that, APT has been used primarily for longterm plan to systematically test and analy ze existing materials, using a defined testing matrix over a large number of test sections. The Fe deral Highway Administration conducted test on mixtures of different gradatio ns and asphalt contents at the Accelerated Loading Facility at Turner Fairbanks Highway Center. (Romero and Stuart 1998) using the data from the ALF at Turner Fairbanks Highway Center showed significant field performance of mixtures with different gr adation but of the same asphalt content. The Texas Accelerated Pavement Test Center (TxAPT) has a new fixed facility that can yield information for accelerated pavement test. The Mobile Load Simulator (MLS) and new instrumentation has been used to de monstrate the effect of heavy loads on thin load-zone pavements and comparison of this pavement with good and superior quality crushed stone bases. The MLS use two wheels instead of a single wheel which is used in Florida. (Hudson et al. 2004) concluded th at fixed APT centers provide more output from the equipment and thus better payout on the investment over the short term than using in-place highways. One role of APT, which was often neglected but have gained popul ar recently, is in evaluating innovative materials. In South Af rica, the Council of Scie ntific and Industrial

PAGE 33

20 Research (CSIR) used the HVS and the AP T facility to evaluate what they called “inverted pavement”. It was found that the presence of a crushed stone base course over a stabilized subbase course in a pavement has a bridging effect at tr ansferring cracks in the stabilized subbase and inhibits the rate of crack propagation to the surface”. Evaluation of new materials, in order to cut down on the cost of experimental evaluation and bring the produc ts to the markets quicker has been done well using APT facility as shown in the evaluati on of SBS modifiers in Florida. Styrene Butadine Styrene (SBS) modifier’s effect on rutting resistance of Superpave fine mixture was evaluated using the APT facility at the Florida’s Department Transport Material Office in Gainesville.

PAGE 34

21 CHAPTER 3 MATERIALS AND METHODOLOGY 3.0 Introduction To be able to accomplish the objectives of this research, experiments were designed that enables us to apply accelerated wheel loads using the South African Heavy Vehicle Simulator (HVS) Mark IV on test tracks cons tructed with the fine -graded mixture and coarse-graded SuperPaveTM mixture. This research has combined accelerated pavement testing with laboratory testi ng using various equipment a nd data analysis tool to characterize and to predict the behavior of the mixtures using theoretical models, materials response characteristics as well as pavement performance models. 3.1 Summary of Methodology The Test Tracks and the HVS were equippe d with appropriate instrumentation to measure pavement temperatures, rut depths, strains and pressures. Rut data were analyzed procedures to predict mixture pe rformance. The laboratory tests were conducted to measure the volumetric and perfor mance parameters of the mixtures. These tests measured the Air Voids, VMA, VFA, APA, Gyratory Slope and Vertical Failure Strain and IDT Creep, Resilient Modulus a nd Indirect Tensile Strength. Mixture Stability calculated as the Gyratory Stability Index, which is related to rutting potential of the mixtures were measure. Analysis of th e gradations of the Job Mix Formula and Plant mixtures were performed using the Flor ida HMA Gradation Guidelines (FHMAGG) recently developed at the University of Florid a. The mixtures Dominant Aggregate Size Range (DASR), Porosity, Interaction Char acteristics and the Interstitial Volume

PAGE 35

22 properties were evaluated to characterize th e mixtures. Visual condition survey was undertaken to observe the sever ity and extent of pavement di stresses that have occurred so as to measure pavement performance of the mixtures under real load conditions. 3.2 Test Track Layout The construction of the test track was star ted in January and was completed in the same month in 2005. The test track is locate d within the FDOT State Materials Office in Gainesville, Florida. The construction of the asphalt concrete surface was done on an existing limesrock base and granul ar subbase. Each lane has two lifts of asphalt concrete from an asphalt plant located in the City of Gainesville. The accelerated pavement test tracks consisted of three (3) lanes paved with fine-graded and two (2) lanes paved with coarse-graded SuperPaveTM mixtures. Each lane is 132 feet long and is divided into three (3) sections of A, B and C each 44 feet. All the lanes are of equal width of 12 feet and were constructed by conventional plants mixing, placement and compaction processes that exemplify the real world situat ion of construction and material variability including mixture’s asphalt cont ent and gradation variations. Each of the sections of 34 feet long had 20 feet of test area and 7 feet at each end fo r acceleration and deceleration of the wheel. The south end of the tracks was reserved for maneuvering the HVS. It was important that no preloading of the track s were introduced by the HVS during the maneuvering because the 40 ton weight of machine could cause damage to the pavement. The testing sequence was arranged such the effect of time on each lane could average out. Figure 3.1 shows the layout of the test tracks and the instrumentation plan and locations.

PAGE 36

23 Figure 3.1 Show the layout accelerated pavement testing test track. 3.3 Pavement Structures The pavement structure consists of a pr epared sandy subgrade, a layer of over 10.5 in (265 mm) of limerock base course and 12 in (305 mm) of granular subbase. Lanes 1 through to lane 3 were constructed with two lifts of 2 in (50 mm) fine-graded SuperPaveTM mixture surfacing whilst Lanes 3 and 4 we re finished with two lifts of 2 in (50 mm) coarse-graded SuperPaveTM mixture. Figure 3.2 shows the pavement structure for the APT test track.

PAGE 37

24 Fine-graded mixture tracks Coarse-graded mixture track Figure 3.2. Pavement structure for the test lanes with fine-graded and coarse-graded mixtures 3.4 Asphalt Concrete Mixtures used Two mixtures were used for this study. Both are 12.5 mm nominal size mixtures and vary only on definition of being a SuperPaveTM fine-graded or a coarse-graded mixture. They were made using Georgia granite aggregates, natural sand from a local sand pit at Gainesville and PG 67-22 asphalt. The mixes were designed by the FDOT using traffic correspond to class D (10-30 millions) ESAL’s over the design period of 20 years. The mix was produced by a batch plan t located in Gainesville. Even though it was planned not to have significant gradati on variations within each mix type, samples from delivery trucks used for the construc tion of the test tracks show considerable variations. This may resu lt in significant aggregate structural differences and performance variations. Tables 3.1, 3.2 3.3 and 3.4 show the Job Mix Formula gradations and grading of samples from the trucks used for the construction of the bottom and top lifts lanes for both coarse-graded and fine -graded mixtures. Mixture volumetric properties and asphalt contents are also shown. The optim um asphalts chosen for the fine-graded and coarse-grade mixt ures are 4.6 and 4.5% respectively. 50 mm AC Toplift 50 mm AC Bottomlift 50 mm AC Toplift 50 mm AC Bottomlift 265 mm Limerock base 305 mm Granular subbase 305 mm Granular subbase 265 mm Limerock base

PAGE 38

25 Table 3.1. Job mix formula and truck samples gradations, volumetric properties and air voids for fine-graded mixture bottom lift. HVS Round 3 (Fine Gradation) Summary Data Lane111222333 SectionABCABCABC TruckJMF112233445 Gmm2.5792.5852.5852.5782.5782.5722.5722.6072.6072.609 Gmb2.4752.4872.4872.4932.4932.4972.4972.5042.5042.484 AC content 4.64.14.14.64.64.34.34.14.14.3 Air Voids 4.03.83.83.33.32.92.94.04.04.8 VMA 14.713.913.914.114.113.713.713.313.314.1 VFA 73737377777979707066 Pbe4.54.14.14.54.54.34.33.83.83.9 Dust Ratio 1.11.01.01.01.01.11.11.11.11.2 19 100.0100.0100.0100.0100.0100.0100.0100.0100.0100.0 12.5 98.097.697.698.698.695.995.997.297.296.8 9.5 90.086.886.890.090.086.786.784.884.887.7 4.75 68.054.754.763.463.460.560.556.756.760.3 2.36 48.038.338.344.744.742.842.839.939.943.2 1.28 34.029.529.532.932.932.132.130.530.532.8 0.425 25.023.623.626.026.025.325.324.624.626.4 0.3 16.015.315.316.216.217.017.015.815.817.6 0.15 8.06.86.87.67.67.97.97.57.58.0 0.075 4.94.14.14.74.74.74.74.44.44.7 Density 93.093.192.893.393.394.293.792.592.191.9 Fine Graded Mix Bottom Lift Table 3.2. Job mix formula and truck samples gradations, volumetric properties and air voids for the fine-graded mixture top lift. HVS Round 3 (Fine Gradation) Summary Data Lane111222333 SectionABCABCABC TruckJMF77 & 8855 & 66445 Gmm2.5792.5942.5942.5902.5912.5912.5982.6022.6022.591 Gmb2.4752.4932.4932.4872.5062.5062.5042.4912.4912.506 AC content 4.64.44.44.34.24.24.24.14.14.2 Air Voids 4.03.93.94.03.33.33.64.34.33.3 VMA 14.714.214.214.313.613.613.714.014.013.6 VFA 73737372767674696976 Pbe4.54.34.34.34.24.24.24.14.14.2 Dust Ratio 1.11.21.21.01.21.21.21.31.31.2 19 100.0100.0100.0100.0100.0100.0100.0100.0100.0100.0 12.5 98.097.797.796.897.097.097.497.597.597.0 9.5 90.088.588.585.485.285.285.988.288.285.2 4.75 68.061.661.658.959.159.157.361.661.659.1 2.36 48.044.544.542.543.043.041.844.644.643.0 1.18 34.034.134.132.732.832.832.233.733.732.8 0.425 25.027.327.326.226.426.426.026.926.926.4 0.3 16.017.517.516.517.217.217.017.217.217.2 0.15 8.08.28.27.68.28.28.08.28.28.2 0.075 4.94.94.94.45.15.14.95.25.25.1 Densit y 93.092.892.192.693.793.493.992.791.492.3 Fine Graded Mix Top Lift

PAGE 39

26 Table 3.3. Job mix formula and truck samples gradations, volumetric properties and air voids for the coarse-graded mixture bottom lift. Lane444555 SectionABCABC TruckJMF667778 Gmm2.5892.5732.5732.5722.5722.5722.568 Gmb2.4852.4612.4612.5482.5482.5482.451 AC content 4.54.84.84.64.64.64.6 Air Voids 4.04.34.34.44.44.44.6 VMA 14.615.715.715.615.615.615.8 VFA 73727272727271 Pbe4.44.74.74.64.64.64.6 Dust Ratio 1.00.90.90.80.80.80.8 19 100.0100.0100.0100.0100.0100.0100.0 12.5 98.098.298.297.297.297.296.9 9.5 90.089.589.586.486.486.485.9 4.75 54.048.548.544.444.444.443.0 2.36 32.029.829.828.128.128.127.0 1.28 23.022.522.521.821.821.821.3 0.425 17.018.218.217.917.917.917.6 0.3 11.011.611.611.911.911.911.6 0.15 5.06.16.15.95.95.95.8 0.075 4.54.04.03.83.83.83.7 Density 94.593.993.694.293.392.894.4 Coarse Graded Mix Bottomlift HVS Round 3 (Coarse ) Summary Data Table 3.4. Job mix formula and truck samples gradations, volumetric properties and air voids for the coarse-graded mixture top lift Lane444555 SectionABCABC TruckJMF223112 Gmm2.5892.5792.5792.5832.5732.5732.579 Gmb2.4852.4682.4682.5142.4572.4572.468 AC content 4.54.64.64.94.54.54.6 Air Voids 4.04.34.32.74.54.54.3 VMA 14.615.215.214.015.615.615.2 VFA 73727281717172 Pbe4.44.64.64.64.54.54.6 Dust Ratio 1.00.80.80.90.80.80.8 19 100.0100.0100.0100.0100.0100.0100.0 12.5 98.096.696.697.896.896.896.6 9.5 90.085.885.889.288.088.085.8 4.75 54.047.847.853.747.347.347.8 2.36 32.030.130.133.729.529.530.1 1.28 23.023.123.125.822.822.823.1 0.425 17.019.019.021.018.718.719.0 0.15 11.012.612.613.912.112.112.6 0.3 5.05.95.96.55.75.75.9 0.075 4.53.83.84.13.63.63.8 Density 94.592.792.692.193.493.793.4 HVS Round 3 (Coarse ) Summary Data Coarse Graded Mix Toplift

PAGE 40

27 3.4.1 Aggregates The coarse mix was made up of 27% of #78 stone with Georgia code GA-553 and Florida code 43, 33% #89 GA-553 code 51, 32% W-10 Screening GA-553 code 20 and 8% natural from Starvation Hill Pit. The pr oportions are by weight of total aggregates. The fine-graded mixture was made from the same aggregate source with the following mix proportions, 27%, 10%, 53% and 10% of Florida code 43, 51, 20 and Starvation Hill sand respectively. These gradations are typical for asphalt pavement construction in Florida. The same aggregate was used for the produc tion of both mixtures but vary only on the gradation requirements. Figure 3-3 shows the gradation of the coarse-graded and fine-graded mixtures. 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0 0.000.501.001.502.002.503.003.504.00 Sieve size (mm) ^0.45Percentage Passing (%) JMF FINE JMF COARSE Figure 3.3. Job mix formula gradation curves of fine-graded and coarse-graded mixtures.

PAGE 41

28 3.4.2 Asphalt The asphalt used for the design and producti on of the two mixtures was a grade PG 67-22 (AC-30) which is typically used fo r high traffic volume asphalt pavement construction in Florida. The AC has consider able stiffness at in service temperature of Florida’s climate and influen ces rutting performance. In this research no asphalt modifier or additive was used in the mixtures. Tabl e 3.5 shows the test requirements and specification for the asphalt used for produc tion of both fine-graded and coarse-graded mixtures. Table 3.5. Specification for the asphalt PG 67-22 used for the mixtures. TestSpecificationTemperatureC ViscosityMax. 3 Pas135 G*/Sin Min. 1.0 kPa64 TestSpecificationTemperatureC G*/Sin Min. 2.2 kPa64 TestSpecificationTemperatureC G*Sin Max. 5000 Pas25 Creep Stiffness0.300 CS 300-12 @ 60 sec Original Asphalt RTOF PAV The optimum asphalt content for the fine-g raded and coarse-graded mixtures were 4.6% and 4.5% respectively. The optimum asphalt content was determined at Ndes of 100 gyrations with the Pine Gyratory Compactor and at 4% air voids content. Appendix A shows the Mix designs for both the fine -graded and coarse-graded mixtures 3.5 Test Tracks Construction and Instrumentation All five test lanes, which were made up of three (3) lanes of fine-graded mixture and two (2) lanes of coarse-graded mixture, were constructed at the APT site at the FDOT State Materials Office in Gainesville. Lane 1 was for aging the studies, lanes 2

PAGE 42

29 and 3 were for fine-graded mixtures whilst lane 4 and 5 were co arse-graded mixture rutting studies. Each lane has two lifts of 50 mm asphalt concrete surface. The lanes were compacted to 7 1% field air voids at the optimum asphalt content. Samples of the mixtures were taken from the trucks and te sted for its asphalt content, gradation and theoretical maximum specific gravity Gmm. After compaction, cores were taken to determine air void content and lift thickness of each section. These tests were carried out by the FDOT State Materials Office. The Tables in Appendix B shows the thicknesses and air voids test results of each lane/section of the test tracks. Figure 3.1 shows the instrumentation of the test truck. A total of nine (9) pressure cells were installed at the bottom layer. Seventy (70) Strain ga ges were installed at various locations and at th e top of the top lift and top of the bottom lift during construction. Six (6) K-type thermocoupl es were installed at the beginning of application of wheel loading for each s ection. The thermocouples measured temperatures at a depth of 50 mm from the surface of the pavement. 3.6 Heavy Vehicle Simulator Test Configuration and Instrumentation The HVS was manufactured in of South Afri ca. It is self propelled, 40 foot long and weighs 40 tons. The HVS is capable of applying different loads and could be equipped with different tires in either singe or dual type. In this research, the HVS was equipped with a test carriage having a 16in wi de super single radial tire of 18000 lbs of load at 115 psi pressure. Th e test carriage is mounted on a test beam that allows both longitudinal and transverse movement. The lo ad via the radial tire makes a 4 inch of transverse movement called wander in one in ch increment for 30 feet Uni-Directional

PAGE 43

30 longitudinal traverse. Tia et al (2002) found that this is the HVS loading configuration that simulates actual field situation. 3.7 Laser Profiler The Laser Profiler, model SLS 5000TM manufactured by LMI Selcom was mounted on a test carriage, one each side and 30 inches apart. Each pass of the laser’s records 58 data points. This process is repeated until each laser covers the lateral distance between them, making a total of 61 sw eeps. There is a coincidence of laser points at the 61st point for the two lasers. The last sweep of the right laser overlaps the first sweep of the left laser. The total lateral distance the lasers make sums up to be 60 inches. Data captured by computers conn ected the HVS was used to calculate the longitudinal profile and the tr ansverse profiles using a comp uter software written by Tom Byron of the FDOT. The figure below shows the HVS Test Carriage and Lasers. Figure 3.4. Heavy vehicle simulato r test carriage and lasers. lasers

PAGE 44

31 3.8 Trafficking A total of 90,000 passes of a 9000 lb, via a 12 inches wide radial tire at 115 psi was applied on each section of the Test Track at speed 8 mph in one direction only (unidirectional mode). Fourteen thousand (14000) passes were applied each day running continuously and only stopped for repairs and servicing of the HVS or if any of the pavement heaters malfunctioned. Testing of the tracks was terminated after a total of 90000 passes were applied to each section of the lanes. Florida States requires some for maintenance interventions when the average rut depth exceeds 12.5mm of routes carriage heavy traffic. The figure below shows the longitudinal rut profiles made after the HVS run with 90,000 passes for the sections. Figure 3.5. Longitudinal rut im prints on section 3-3B. 3.9 Heating and Temperature Control The effects of temperature on rutting have been investigated by many researchers. Robertson (1997) observed that the shape of the temperature profile in a pavement is

PAGE 45

32 dependent on the air temperature history. There could be several inflection points, depending on air temperatur e history of the location. Thus to simulate Florida climatic co nditions, the pavement was heated to 50 C. The heating, measuring and temperature cont rol system of the test track pavement consisted of three pairs of Watlow Raymax 1525 radiant heaters capab le of heating the space enclosed by the HVS and insulators to a temperature of 50 C. Pavement temperature was measured by 6 K-type ther mocouple inserted into the pavement one each side of the wheel path at 3 pair locatio ns. The average temperature was measured by the thermocouples at a depth of 2 inches of the pavement and was recorded by the monitoring computer. To avoid the situ ation of varying air temperature of the Gainesville area causing variations in the pa vement temperatures, the track was shielded from the effects of the environment using 3in thick Styrofoam board covered with an 0.08in thick aluminum sheeting. Figure 3.6 and 3.7 shows the thermocouple layout on a test section and the HVS and insulation assembly. Figure 3.6. Thermocouple assembly on track 3-2A.

PAGE 46

33 Figure 3.7. Heavy vehicle simulator with its insulation assembly. 3.10 Rut Measurement The two non-contact lasers mounted on the Test Carriage measures the vertical distance between the surface of the pavement a nd the laser position. The initial runs of the HVS was set to apply no load as it move s longitudinal with the specified maximum wander of 4 inches in 1 inch increments. Such runs set the baseline reference for subsequent runs with load application. The LMI Silicon Lasers have a 0.025% resolution and collect and output real time data every 4 inch es taking 58 transverse data per longitudinal pass. The two lasers, left and right as they are designated make a “straight” and diagonal longitudinal moveme nt interchangeably to complete the maximum wander of 4 inches. The process would be repeated 30 times until each laser would sweep over a lateral distance of 30 inches. Analys is of the profiler data was done using a Computer Software written by Tom Byron of the FDOT. The initial surface profile before the test is subtracted from the measured surf ace after the test to obtain the differential surface profile.

PAGE 47

34 3.11 Air Void Content and Thickness Changes The layer thickness and air void content cha nge with the applica tion of wheel loads in and just outside the wheel path as a result of rutting of the HMA mi xture. The initial as constructed layer thicknesses and air voids were determined by taking 2 cores each per section per layer. Their thicknesses were meas ured with a caliper. The bulk specific gravity was determined and air voids calculated. After the HVS runs 4 cores each were taken from each section. Two (2) of those cores were from inside the wheel paths and the 2 others from just the edge of the wheel path in the humps created by the shearing of the mixtures. The thickness was measured tran sverse to the directi on of travel of the wheel along the side of the core after which the top and bottom lift where separated using a diamond saw. The figure below shows the core locations compar ing center of wheel path to edge of wheel path Figure 3.8. Location of cores to be take n after the HVS runs on section 3-3B.

PAGE 48

35 3.12 Determination of Viscosity of Cores Two cores were taken to determine the viscos ity of the asphalt shortly after laying. The asphalt was extracted according to the Reflux Asphalt Extraction procedure, ASTM 2171-95 and recovered from Tric hloroethylene solvent usi ng ASTM 5404-97 from cores taken from the section before and after the HVS runs. Two cores each were taken from the sections before the runs. After the runs, cores were taken from lanes 3 and 5 only. Lane 3 has the fine graded mixture while la ne 5 has the coarse-graded mixture. In lane 3, two specimens were taken the bottom lift of section 3A, two taken from wheel path top lift of section 3B a nd two specimens taken from the edge (hump) top lift of section 3C. In lane 5, two specimens were ta ken from the hump of section 5A in the top lift, two from the top lift wheel path of section 5B whilst two cores were taken from the bottom lift of section 5C. The viscosity at 60 C of the recovered asphalt was measured using the Brookfield Viscometer DV III+ according to ASTM D 4402. All asphalt samples were tested in triplicates. Three shea r rates were used in the test. An initial shear rate of 15 % torque was varied to 50 % and then to 85 % at 1.5 minutes intervals. The average viscosity was calculated as the vi scosity at the different rate of shear at 60 C. 3.13 Laboratory Testing on Plant and Laboratory Prepared Mixtures A series of laboratory tests were perf ormed to determine the mixture response characteristics and performance prediction pa rameters of the two mixtures taken from delivery trucks and stored during the constr uction phase of the te st tracks. Some specimens were also batched in the labor atory and tested for similar parameters.

PAGE 49

36 3.13.1 Asphalt Pavement Analyzer Test Rut susceptibility has been correlated to APA rut depths in many researches and a standard specification has been set by the FDOT for rut resistance mixtures. An upper limit of 8 mm rut depth in the AP A is the standard specification. Two specimens of 75 mm and 115 mm thickne ss and 150 mm diameter were tested in the Asphalt Pavement Analyzer (APA) at 7% air voids. The test was conducted by Howard Moseley of the FDOT using standard method AASHTO TP 63-03. The specimen was tested at 64 C at 8000 cycles in a bi-d irectional mode under a 100 5psi pressurized rubber hose. All specimens we re plant mixtures sampled from delivery trucks compacted in a Pine Gyratory Compactor. 3.13.2 Servopac Gyratory Compactor Testing The mixture’s rutting performance pred iction was evaluated using the Servopac Gyratory Compactor with shear and density measuring capabilitie s. The Servopac Version 1.23 developed by Industrial Process Co ntrol (IPC) of Australia was used. Both the top and bottom lifts were test ed in the Servopac Gyratory Compactor using the testing procedures devel oped by Birgisson, et al (2002). Four 4.5 kg specimens were prepared from plant mixtures sampled from delivery trucks and laboratory batched specimens usi ng the fine-graded and coarse-graded JMF. The specimens were compacted using the ser vopac gyratory compactor Two specimens each were compacted to Nmax of 160 gyrations representing traffic level D at a gyration angle of 1.25 600kpa of ram pressure, 30 revolu tions per minute (rpm) and at 300 F. The servopac gyratory compactor automatically measured the density and gyratory shear at every gyration. The bulk density of th e compacted specimen was determined and the

PAGE 50

37 machine measured densities corrected. Th e number of gyration to obtain 7% air void was determined. The two other specimens were compacted at 1.25 to number of gyrations to obtained 7% air voids. Th en, the angle of gyration was changed to 2.5 and compacted for 100 more gyrations. Ram pressure, rpm and temperature remained constant. The gyratory slope a nd the vertical failure strain we re determined from the test results. 3.13.3 SuperPaveTM Indirect Tensile Test Superpave IDT test as recommended by Roque et al (2004) was performed on plant mixtures sampled from delivery trucks during th e construction of the test tracks. Truck 1, 2 and 3 contained coarse-gra ded mix which was used to construct lanes 4 and 5. Truck 4, 5, 6, 7 and 8 were fine-graded mixtures used in the construction of lane 1, 2 and 3 of the test track. Only the top lift specimens were tested in the IDT. Two (2) sets of 3 specimens from each truck sample were tested using the MTS Superpave IDT testing Machines. Specimens were made from shot-term oven aged specimen compacted to 7 0.5% air voids in the Servopac Superpave Comp actor. Test temperature was 10 C for conducting the Resilient Modulus, Creep and Tens ile Strength Test on all specimens. 3.14 Condition Survey All the 12 pavement sections were inspected before and after the HVS runs. The main purpose of the survey was to find out whether any other pavement distress apart from rut has occurred due to the application of the loads. The inspection was a visual survey and any distress found was marked for further evaluation. Where cracks were

PAGE 51

38 found, the width and total area were measured and cores taken so as to assess the causes for these cracks.

PAGE 52

39 CHAPTER 4 RESULTS, ANALYSIS AND DISCUSSION 4.0 Evaluation of Rut Profiles Rut profile data give considerable info rmation on the causes, types and mode of pavements or HMA mixture rutting. The shap e of the transverse profile for instance could be used to determine whether the rutti ng was due to the surface layers (instability rutting) or was probable from the bases or the subgrade la yers (structural). 4.1 Introduction Both the rut depth and transverse paveme nt profiles after the HVS loadings were analyzed. The average differential rut de pth was plotted versus number of HVS passes from 100 to 90000 passes. The differential rut depth was initially ev aluated at every 100 passes and then at 1000 to 5000 as the number of passes grew. The transverse profiles at 90000 passes were used to compute the Area Parameter. The area parameter compares the volume of the humps to that of the wheel path depression and gives an indication of whether rutting was due shear flow or densification. 4.2 Rut Depth Figure 4.1 shows the differentia l rut depths of the various test sections versus number of HVS wheel passes. Both the fine-graded and coarse-graded mixt ures show similar trends. Rut depth increased rapidly and exponentia lly with number of passes initi ally and then flattens out on the second portion as the number of passes gr ew. The steep rise comes as a result of the very large rut depth per pass of around 0.024mm/pass to about 0.000058mm/pass of

PAGE 53

40 the wheel. Rut depths development on each te st sections increased rapidly initially to between 2500 and 3000 passes. This was followed by for the very slow rate of rut development until termination. The rate of growth rut per pass wa s within a range of 2.7-6.2 x 10-5mm/pass at 90,000 passes for the sec tions at termination. The initial rate was more than 400 times the rut growth at te rmination for the sections. The minimum and maximum rut depths are 9.78 mm and 16.3 mm for the fine-graded mixture and a minimum of 13.47 mm and maximum of 17.13 mm for the coarse-graded mixtures respectively. The average rut depth wa s 12.8 mm for the fine-graded mixture and 15.1 mm for coarse-graded mixture. Figure 4.2 shows the maximum differential rut depth for all the sect ions of the fine and coarse graded mixtures. Figure 4.3 s hows the maximum absolute rut depth for all the test sections 0.0 5.0 10.0 15.0 20.0 0.010000.020000.030000.040000.050000.060000.070000.080000.090000.0100000.0NUMBER OF PASSESRUT DEPTH (mm ) 3-3A 3-3B 3-3C 3-5B 3-5C 3-5A 3-2C 3-2B 3-4A 3-4B 3-2A 3-4C Figure 4.1. Plot of change in rut depth versus number of HVS wheel passes.

PAGE 54

41 0 5 10 15 20 3-2A3-2B3-2C3-3A3-3B3-3CAVFAVC3-4A3-4B3-4C3-5A3-5B3-5CSECTIONSRUT DEPTH (mm) Fine Coarse Figure 4.2 Maximum differential ru t depths for the sections. 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 2A2B2C3A3B3CAVFAVC4A4B4C5A5B5CSECTIONSRUT DEPTH (m m COARSE FINE Figure 4.3 show the maximum absolute ru t depths for all the test sections. From the analysis above, there seemed be a difference in rut performance of the two mixtures at 50 C. The fine-graded mixture seemed to have out performed the coarse-graded mixture in both differential and absolute rutting. However statistical analysis of the means of di fferential rut depth of the fine-graded and coarse-graded

PAGE 55

42 mixtures showed that there are no significant di fferences in the rut depths. The Student t-test statistics at 95% level of confidence was 1.81 at 7 degrees of pooled-freedom whilst the observed “t” for the differential rut de pth was 1.06 indicating th e null hypothesis of equality of the mean of the fine-graded and coarse-graded mixtures could not be rejected. 4.3 Transverse Rut Profile Considerable amount of information was obtai ned from the transverse profile of the sections. As early as 100 passes of the te st wheel, both the fine-graded and the coarsegraded mixtures have developed humps just out side the wheel paths. Humps develop as a result of shear deformation or instability at relatively low air void content. However the top lift layers of both mixtures deve loped humps early in their service life. The mixtures were compacted to an in itial air voids ranging from 92.1% to 93.7% and 92.4% to 93.5% of Gmm for the fine-graded and coarse-g raded mixtures respectively. After 90000 passes, the air voids of the sect ions ranged from a minimum of 4.4% to maximum of 6.7% at termination for the coarse-graded mixtures and 4.3% to 6.4% respectively for the fine-mixture. This mean s that humps have developed at relatively higher air voids for the mixtures giving an i ndication that most of what was contributing to the rutting must be shear deformation of the HMA mixture and not densification. Figure 4.4 show the differential transverse profiles at 100 passes for the fine-graded mixes and Figure 4.5 shows the same prof iles at 100 passes and a profile of a 90000 passes shown on it for comparison purpose. Figure 4.6 and 4.7 show the same plots for the coarse-graded mixes. The height of the hump at 100 passes for the fine mixture was about 0.8 mm and that of the coarse mixture was over 1 mm

PAGE 56

43 -2.000 -1.500 -1.000 -0.500 0.000 0.500 1.000 010203040506070 TRANSVERSE SWEEP (IN)DIFFERENTIAL RUT DEPTH ( M 3-2A 3-2B 3-2C 3-3A 3-3B 3-3C Figure 4.4 Differential transverse profile s at 100 passes of the fine-graded mix. -12.000 -10.000 -8.000 -6.000 -4.000 -2.000 0.000 2.000 4.000 6.000 8.000 10.000 010203040506070 TRANSVERSE SWEEP (IN)DIFFERENTIAL RUT DEPTH ( M 3-2A 3-2B 3-2C 3-3A 3-3B 3-3C 3-3C(90000)passes Figure 4.5. Differential transver se profiles at 100 passes as compared with one of 90000 passes for the fine-graded mixture.

PAGE 57

44 -3.000 -2.500 -2.000 -1.500 -1.000 -0.500 0.000 0.500 1.000 1.500 010203040506070 TRANSVERSE SWEEP (IN)DIFFERENTIAL RUT DEPTH ( M 3-4A 3-4B 3-4C 3-5A 3-5B 3-5C Figure 4.6. Differential rut profiles at 100 passes of the coarse-graded mixture. -12.0 -10.0 -8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0 010203040506070 TRANSVERSE SWEEP (IN)DIFFERENTIAL RUT DEPTH ( M 3-4A 3-4B 3-4C 3-5A 3-5B 3-5C 3-5A(90000)passes Figure 4.7. Differential transver se profile at 100 passes as compare with one the 90000 passes for the coarse-graded mixtures.

PAGE 58

45 The evolutions of the transverse profile of the fine-graded and coarse-graded mixtures are similar. In both cases, the first 100 passes gave the largest single differential rut depth. Approximately, bot h mixtures recorded just around 2 mm depression in the wheel path and humps of a bout 1 mm in the first 100 passes. As the number of passes grew from 100 passes to 9000 passes, both the depression in wheel paths and the humps also grew in magnitude and in similar propor tion as the initial 100 passes. The downward movement of the wh eel path, and the upward progression of the humps indicated a combination of shear flow and densification occurred during the test Figures 4.8 and 4.9 show the evolution of th e transverse profiles of the fine-graded and coarse-graded mixtures section 3C and 5A. -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 020406080100120140 Laser Travel (0.5 in)Differential Transverse RutT Depth (mm ) 100 passes 200 passes 300 passes 1000 passes 2500 passes 5000 passes 7000 passes 10000 passes 20000 passes 50000 passes 70000 passes 90000 passes Figure 4.8. Evolution of transverse profile of fine-graded mixture section 3C.

PAGE 59

46 -12 -10 -8 -6 -4 -2 0 2 4 6 8 020406080100120140 Laser Travel (0.5 in)Differential Transverse Rut Depth (mm ) 100 passes 200 passes 300 passes 1000 passes 2500 passes 5000 passes 7000 passes 10000 passes 20000 passes 55000 passes 70000 passes 90000 passes Figure 4.9. Evolution of transverse profile of coarse-graded mixture section 5A. 4.4 Area Parameter Change Method Evalua tion of Transverse Rut Profiles The laser rut profilers attached to HVS m easured the initial surface profiles before the application of the loads thus rut progr ession had a reference surface for comparison. The area difference enclosed between the initial surface and th e terminal surface profile was used to determine the type of rutti ng occurring under the lo ad. Drakos et al, (2002) using transverse profile data from th e Modified APA showed that a net positive area indicated a predominance of instability rutting whilst a net negative area showed mainly densification rutting. The positive ar eas are the humps and the negative areas are the depressions in the wheel paths. Figur e 4.8 and 4.9 show the initi al and final surface profiles for section 3-3C and 3-5A. These sections (3-3C and 3-5A) have the maximum differential rut depths among all the test se ctions. The sectio ns with to lowest differential rut depths were 2A and 3B a nd were recorded on the fine-graded mixture

PAGE 60

47 298.0 300.0 302.0 304.0 306.0 308.0 310.0 312.0 314.0 316.0 318.0 320.0 010203040506070 LASER TRAVEL (in)RUT DEPTH (mm ) 90000passes zeropass Figure 4.10. Initial and final surface prof ile of a fine-graded section 3-3C. 300.0 302.0 304.0 306.0 308.0 310.0 312.0 314.0 316.0 318.0 320.0 322.0 010203040506070 LASER TRAVEL (in)RUT DEPTH (mm ) 90000passes zeropass Figure 4.11. Initial and final surface profile of a coarse-graded section 3-5A. The Area Change Parameter (ACP) can be ca lculated for any transverse profile. Using Loess curve fitting function or polynomial after linearization of the rut profiles, the areas under the initial and final surface profiles were calculate d by integrating the

PAGE 61

48 polynomial from one end of the cross section to the other of the transverse profile for all the 12 sections. Loess allows for modeling of the parametric regression surface of the transverse rut profiles so they could be inte grated. Table 4.1 s hows the areas under the original and final surface profiles and the AC P as well the Differential and Absolute rut depths of the sections. Table 4.1. Area change parameters of the fine-graded and coarse-graded sections. Section ID Initial Area (in2) Final Area (in2) Absolute Rut Depth Differential Rut Depth Area Change (in2) Area Change Parameter 3-2A734.581732.94411.229.81.6370.223 3-2B734.581732.94410.3710.11.6370.223 3-2C734.581732.94413.1213.81.6370.223 AVERAGE734.581732.94411.5711.21.6370.223 3-3A728.754726.21610.5812.32.5380.348 3-3B612.683609.13511.2212.73.5480.579 3-3C728.571724.57815.2416.93.9930.548 AVERAGE690.003686.64312.3514.03.3600.492 Section ID Initial Area (in2) Final Area (in2) Absolute Rut Depth Differential Rut Depth Area Change (in2) Percentage Area Change 3-4A742.882740.9113.5513.61.9720.265 3-4B734.558732.95414.1813.51.6040.218 3-4C728.055726.82517.0315.81.2300.169 AVERAGE735.165733.56314.9214.31.6020.218 3-5A736.521732.95415.0317.13.5670.484 3-5B725.882724.61311.0114.21.2690.175 3-5C736.172732.74812.2816.23.4240.465 AVERAGE732.858730.10512.7715.82.7530.375 FINE-GRADED MIX COARSE-GRADED As shown the positive ACP indicates th at primarily both the fine-graded and coarse-graded sections exhi bited instability rutting. There is a theoretical basis for the ACP. Consider the transverse profile of a rutted surface superimposed on the initial or existing surface before trafficking as shown below.

PAGE 62

49 Typically this is how HMA pavements with ad equate structure behave under loads when instability rutting is the rule. If for some reason for instance, higher initial air voids contents immediately after compaction of the mi xture, then consolidation rutting with the absence humps at the edges but only depressions in the wheel paths will be seen from the transverse profile plots Figure 4.12. Positive A1 and negative A2 areas of a transverse rut profile. If HMA material was moved from A2 and equal amount was transferred to A1 then shear deformation was primarily contributing to the rutting. On the other hand if less was transferred that is A1 less than A2 then, some considerable amount of consolidation would have occurred and consolidat ion would be controlling rutting.

PAGE 63

50 Table 4.1 shows the differences of the Ar ea Change Parameter of both the finegraded and coarse-graded mixtures. The hi ghest area-change determined for the finegraded mixture was 0.579 in2 for section 3B which record ed a differential rut of 12.7 mm. In section 3C where the highest differential rut de pth recorded was 16.9 mm the area-change parameter determined was 0.548 in2. In general there seem to be no direct relationship between the differentia l rut depth and the Ar ea Change Parameter. Since both fine-graded and coarse-graded mixtures exhibited shear deformation primarily, the ACP was unable to distinguish their rutting performance. The Area Change Parameter can be used in the evaluation of transverse rut profiles to determine the type of rutting occurring on HMA pavements. The calculations of the Area Change Pa rameter are as shown in Appendix C. 4.5 Evaluation of Core Densities Tables 4.2 and 4.3 show the air voids conten ts of the cores from the wheel paths and the edges of the wheel paths after the HVS runs for the fine-graded and coarsegraded mixtures respectively. Bulk densities of cores extracted from th e test sections were determined according to AASHTO T 166-93 and the air vo ids calculated using the Gmm for each truck sample. The Gmm was determined using the Rice method for the determination of the Maximum specific gravity of bitumi nous paving mixtures, AASHTO MT 321.

PAGE 64

51 Table 4.2. Air voids level of the cores from th e wheel paths and edges of the wheel paths lane 3 fine-graded mixture sections. lanesectionlocation Sample No Wt in Air (g) Wt in H2O (g) SSD Wt (g) Gmb (Mg/m3) Gmm Air Voids (%) Average decrease wheel path air voids ( % ) Average Initial thickness (mm) Rut due to densification (mm) Total Rut depth after HVS 90000 passes ( mm ) Total percentage rut due to densification 22171.61299.72172.22.4892.6024.3 42154.912862155.62.4782.6024.8 12343.713652345.62.3902.6028.1 32295.11332.62296.92.3802.6028.5 222121319.322132.4752.6024.9 42333.91396.52339.92.4742.6024.9 12356.81384.82359.32.4182.6027.1 32311.71341.72313.22.3802.6028.6 222041310.82204.62.4662.5914.8 42321.61384.42322.82.4742.5914.5 122821308.22284.12.3382.5919.8 325801498.82582.42.3812.5918.1 22258.41352.82259.12.4922.6074.4 42239.21322.92240.12.4412.6076.4 12233.51323.82234.42.4532.6075.9 32242.813422243.22.4892.6074.5 22084.71244.72085.42.4802.6074.9 41987.21188.41987.82.4862.6074.6 12488.51471.22489.92.4432.6076.3 32250.61316.12251.82.4052.6077.7 22010.11204.42010.72.4932.6094.4 41803.61079.51804.32.4882.6094.6 12135.21259.12136.22.4342.6096.7 321001234.121022.4202.6097.3 57.8 58.6 53.4 53.5 2.3 2.4 3.8 55.5 2.9 4.3 -0.2 51.61.26 1.25 -0.09 2.46 1.55 2.02 bottomC bottomB bottomA topC topB 3 edge center edge center topA 3 edge center edge center edge center center edge 3 3 3 3 12.30 12.70 16.90 15.67 22.07 22.03 Generally there was a larger reduction of air voids in the wheel paths for the top and bottom lifts of the fine mix gradation than that the coarse-grade d mixtures. At the top lift, air voids reduction ranged from 2.9 to 4.3% As expected the bottom layers have less air void reduction and at se ction 3-3A there was no reducti on at all at the bottom lift. The overall contribution of densificati on for the fine-graded mixture to rutting ranged from 16% to 22%. Thus much of the rutting was due to shear deformation creating humps at the edge of the wheel paths.

PAGE 65

52 Table 4.3. Air voids level of the cores from th e wheel paths and edges of the wheel paths of lane 5, coarse mixture. lanesectionlocation Sample No Wt in Air (g) Wt in H2O (g) SSD Wt (g) Gmb (Mg/m3) Gmm Air Voids (%) Average decrease wheel path air voids (%) Average Initial thickness (mm) Rut due to densification (mm) Total Rut depth after HVS 90000 passes (mm) Total percentage rut due to densification 22335.11385.52336.12.4562.5734.5 42239.71328.42240.82.4552.5734.6 12370.51385.72377.22.3912.5737.1 32580.21500.42585.72.3772.5737.6 22263.61344.82265.32.4592.5734.4 42273.71346.92274.82.4502.5734.8 12343.81371.42348.62.3982.5736.8 32485.31457.824902.4082.5736.4 21944.21139.81947.62.4072.5796.7 42177.31293.12178.52.4592.5794.6 12058.71211.720622.4212.5796.1 32225.51315.82227.22.4422.5795.3 21833.511041834.22.5112.5722.4 42061.81229.62062.82.4752.5723.8 120641218.42065.52.4372.5725.3 32168.21279.12169.32.4362.5725.3 22124.81273.22125.82.4922.5723.1 41970.41175.91971.52.4772.5723.7 12156.21276.82157.12.4492.5724.8 31886.21117.11886.92.4502.5724.7 21988.811841989.62.4692.5683.9 41861.41115.21862.22.4922.5683.0 11902.41121.51903.62.4322.5685.3 32197.81301.72198.52.4512.5684.6 0.78 0.72 1.15 0.03 5 5 5 5 5 5bottomC bottomB bottomA topC topB topA center edge center edge center edge center edge center edge center edge 2.854.517.1015.59 1.51 2.053.614.2012.62 1.07 0.153.216.205.02 2.252.2 1.353.3 1.552.0 There was considerably less reduction in air voids of the wheel paths of the coarse-graded sections than th e fine-graded sections. Th e air void reduction of the top lift ranged from 0.1 to 2.8% and the bottom lif t from 1.3 to 2.2%. The percentage of rutting due to densification ranged from 5 to 16 % indicating that much more shearing deformation occurred in the coarse-grade d mixture than the fine-graded mixture. It was observed that sections with lower densification show higher rut depths. This suggests that at optimal (6 to 8%) level of field compaction of the mixtures, rutting was associated primarily with mixture’s shear re sistance measured by resistance to aggregate sliding and rotation within the mixture. Shearing was induced by lateral stresses

PAGE 66

53 generated by radial tires and it seems the co arse-graded mixtures with greater proportion of coarse aggregates developed bigger hum ps than finer mixture under the same wheel loads. The change in air voids between the wheel paths and humps quantifies the proportion of the total rutting a ssociated with densification. 4.6 Evaluation of Recovered Asphalt Table 4.4 shows the average viscosity of the Asphalt recovered from the wheel paths, top lift, top lift humps and bottom lift wheel path, for both fine-graded and coarsegraded mixtures. The viscosities at 60 C of the recovered aspha lt are in Appendix C. Table 4.4. Viscosity of recovered asphalt In the wheel paths for both fine-grade d and coarse-graded mixtures, recovered asphalt viscosities did not show any significan t differences. The fine-graded mixture has an average viscosity of 27,539 poises whilst the coarse-graded mi xture has average FINE MIXCOARSE MIX SectionLocationSample IDTop lift SectionLocationSample IDTop lift I 2711720K2867903 I3 2526104K32551666 J 3023872L2546778 average 2753898.67 average 2655449 E 4162000G E3 4543666G32907439 **F 2960397**H655747 average 4352833 average 2907439 SectionLocationSample IDBottom lift SectionLocationSample IDbottom lift A 1198604C1333246 A3 758256C3 B 1986190D11168119 average 1314350 average 1250682.5edgeVISCOSITY OF RECOVERED ASPHALT AT 60 C POISE 3Bcenter5Bcenter edge 3C 5A 3Acenter5Cedge

PAGE 67

54 viscosity of 26, 555 poises. Similar comparis on could be made of the bottom lifts. The fine-graded mixture recorded an average of 13,144 poises in the wheel path whilst the coarse-graded mixture recorded 12,509 poises in the edges. Thus there are no significant differences in viscosities for the bottom lifts of the two mixtures. In the humps both the fine-graded and coarse-graded mixtures reco rded viscosities which are two times their respective viscosities of the bottom lifts. Th at indicated the effects of the ageing process of the environment on the top 50 mm of the pavement. It was also observed that the fine-graded mixture had tw o times the viscosity of the coarse-graded mixture for top lift in the hum ps as show for sections 3C and 5A in Table 4.4 however, differential rut depth reco rded from the HVS show that sections 3C and 5A have the highest and similar ru t depths of 16.9 and 17.1mm for fine-graded and coarse-graded mixtures respectively. From Table 4.2 and 4.3, the average termination air voids in the humps of sections 3C and 5A are 8.95 and 7.35% indicating a diffe rence 1.6%. The wheel paths air voids were relative the same for the sectio ns averaging 4.65 and 4.55% for 3C and 5A respectively. The 1.6% higher field air void s content in the hump s of the fine-graded mixture probably was responsible for doubling the asphalt viscosity in the fine-graded section. Due to the limited da ta the above analysis did not show conclusively whether or not asphalt viscosities had any influence on the rutting patterns of the sections. Specimen F, G and H got contaminated and we re excluded from the calculations. 4.7 Asphalt Pavement Analyzer Test The mixtures were assessed based on their APA test results. Generally APA test result gives an indication of mixture future field rutting performance. However APA

PAGE 68

55 results have not been known to distingui sh between fine-graded and coarse-graded SuperPaveTM mixture rutting potential. The APA test was conducted by the FDOT at 8000 cycles and at 64 C. The results are as shown in Table 4.5. The fine-gra de mixture has APA rut depth ranging from 3.3 mm to 4.0 mm with a mean of 3.7 mm for th e 75 mm specimen and 3.4 to 3.6 mm with an average of 3.5 mm for the 115 mm specimen for the top lift. Table 4.5. Asphalt pavement analyzer rut dept h for both fine-graded and coarse-graded mixture. 75 mm specimen115 mm specimen 2A3.393.74.03.6 2B3.393.44.03.6 2C3.6933.53.4 Average3.493.43.833.53 3A4.392.63.33.5 3B4.392.13.33.5 3C3.392.14.03.6 Average3.9792.273.533.53 4A4.393.73.13.3 4B4.392.63.13.3 4C2.792.12.82.9 Average3.7792.803.003.17 5A4.592.62.33.3 5B4.592.62.33.3 5C4.392.63.13.3 Average4.4392.602.573.30 2A3.393.34.32.9 2B3.994.23.83.0 2C2.992.83.83.0 Average3.3793.433.972.97 3A4.092.42.52.9 3B4.092.52.52.9 3C4.892.5not tested3.0 Average4.2792.472.502.93 4A4.392.62.82.9 4B4.393.62.82.9 4C4.493.72.82.6 Average4.3393.302.802.80 5A4.593.62.82.6 5B4.593.72.82.6 5C4.393.33.22.9 Average4.4393.532.932.70-14.97-14.30 13.6 17.1 14.2 13.93 13.6 13.5 15.8fine-graded Coarse-gradedHVS Rut Depth (mm) 90,000 passes* 9.8 10.1 14.8 11.57 12.2 12.7 16.9Top Lift Bottm LiftAPA Rut Depth (mm), 8000 cycles Density (%) Gmm Lab Air Voids HVS Lane Mix Type Liftfine-graded Coarse-graded The coarse-graded mixture has ranges of 2.3 mm to 3.1 mm and average of 2.8 mm for 75 mm pills and 2.9 mm to 3.3 mm with an average 3.0 mm for the top lift

PAGE 69

56 respectively. The bottom lifts for both the fine-graded and coarse-g raded mixtures were 3.4 mm and 2.9 mm for 75 mm pills a nd 3.0 mm and 2.8 mm for the 115 mm pills respectively. 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.02A2B2C3A3B3CAVFAVC4A4B4C5A5B5C APA 75mm APA 115mm HVS RUTS FINE COARSE A VERAGES 115MM 75MM HVS Figure 4.13. Asphalt pavement analyzer compar isons and HVS rut depths for the finegraded and coarse-graded mixture of the top lift. The APA rut depths for the fine-graded mixture has average of 3.7 mm and 3.5 mm for the 75 mm and 115 mm respectively whilst the coarse-graded mixture has average rut depths of 2.8 and 3.0 mm for the 75 and 115 mm. Examination of results indicated no significant differences for the 115 mm specimens but the 75 mm thick specimens, th e coarse-graded mixtures seemed to have out performed the fine-graded mixture. Drakos (2002) concluded that the APA hose does not simulate the stress state in the field during rutting and may not distinguish well the field pe rformance of fine-graded and coarse-graded SuperPaveTM mixtures

PAGE 70

57 4.8 SuperPaveTM Servopac Gyratory Compaction Results The Superpave servopac gyratory results were used as rutting performance predictor using the “pass-failure” criteria developed at the University of Florida. The gyratory shear slope against th e initial vertical strain obt ained by changing the gyratory angle from 1.25 to 2.5 could be used to evaluate mixtures rutting resistance. 0 100 200 300 400 500 600 700 800 050100150200 NUMBER OF GYRATIONAVERAGE SHEAR STRESS KP A fine-graded mixture coarse-graded mixture Figure 4.14. Aaverage gyratory shear stress ve rsus number of gyrat ions for bottom lift. Generally the gyratory shear stress of the fi ne-graded mixtures appears to be higher for both the top and bottom lifts. The figur es below show the average gyratory shear stress against the number of gyra tions for the bottom and top lift. The magnitude of the gyratory shear stress in itself does not mean much in terms of the rutting potential. It is not a fundamental property of a mixture and could therefore not be used to evaluate the rut resistan ce of neither fine-graded nor coarse-graded SuperPaveTM mixtures.

PAGE 71

58 However the slope of the gyratory curve in the air voids range of 4% to7% in conjunction with the initial vertical strain obtained from the gyratory compactor can be used to screen out mixtures with a high rutting potential. 0 100 200 300 400 500 600 700 800 900 050100150200 NUMBER OF GYRATIONSGYRATORY SHEAR STRESS KP A fine mixture coarse mixture Figure 4.15. Average gyratory shear stress versus number of gyrations for the top lift. Figure 4.16 shows the gyratory shear slope for section 3-3C between air voids of 4% to 7% of the top lift of th e fine-graded mixture. The slope of the plot of change in gyratory shear versus the change in air voids between 4% and 7% gi ves an indication of the mixtures rut resistances. Values a bove 15 kPa have been shown to give good field rutting resistance (Darku 2003). The ability of a mixture to generate e nough shearing response to resist straining after the gyration angle is changed from 1.25 to 2.5 has been shown to affect the rutting resistance of mixtures. Figure 4.17 shows the plot of th e gyratory shear versus the number gyrations during the change of the gyratory angle for calc ulating the strains.

PAGE 72

59 y = 24.00x + 672.09 R2 = 0.98740 745 750 755 760 765 770 3.03.54.04.55.0Natural Log RevolutionsGyratory Shear 7% to 4% AV Figure 4.16. Gyratory shear stress versus numbe r of gyrations of the servopac compactor for section 3C fine-graded section. 0 100 200 300 400 500 600 700 800 900 1000 020406080100120140Number of GyrationsShear Stress(kPa) Shear Str. 1.25 2.5 2.5 Figure 4.17. Gyratory shear stress versus number of gyrations for 3-3C with a change of gyration angle from 1.25 to 2.5

PAGE 73

60 Trucks 1,2 and 3 are coarse-graded mixtures Trucks 4, 5, 6, 7 and 8 are fine-graded mixtures for the top lift. Trucks 1, 2, 3 and 4 are fine-graded mixtures. Trucks 6, 7 and 8 are coarse-graded mixtures for the bottom lift. Generally the fine-graded mixture have higher gyratory sl opes but lower vertical strains whilst the coarse-graded mixtures showed lower slopes and marginally higher vertical strains. The fine-graded mixtures could be described as brittle according Birg isson et al (2002) “Pass/Failure Criteria” for both top and bottom lifts. Figure 4.18 shows the plots of the gyratory shear slope versus the in itial failure strains of the top lift. 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 1.001.201.401.601.802.002.202.40 Vertical Failure StrainGyratory Shear Slope(Gs/logN ) truck 1 truck 2 truck 3 truck 4 truck5 truck6 truck7 truck8 jmf fine jmf coarse coarse fine Figure 4.18. Gyratory slope and initial failure st rain of the top lift fine-graded and coarse-graded mixtures According to Birgisson et al (2002) the fine-graded mixtur es could be described as brittle whilst the coarse-graded mixtures coul d be described as low shear strength and of marginally brittle. Both mixtures may ru t when load but from different mechanisms. The coarse-graded mixtures may have rutt ed due their low gyratory shear strength.

PAGE 74

61 0 5 10 15 20 25 30 11.21.41.61.822.22.4 Strain at Initial Minimum Shear(%)Gyratory Shear Slope(Gs/Log(N ) optimum plastic brittle low shear strength Figure 4.19. Pass/Fail criteria for evaluation of rut resistance. The gyratory slope ranged from 7.33 kpa to 13.4 kpa whilst the differential rut depth ranged from 13.5 mm 17.1 mm. Fi gure 4.20 shows the relation between the gyratory slope and the differential rut depth fo r the coarse-graded mixture of the top lift with R2 of 0.91. y = -0.5153x + 20.363 R2 = 0.9138 0 2 4 6 8 10 12 14 16 18 20 579111315 GYRATORY SLOPE (kpa)DIFFERENTIAL RUT DEPTH ( m differential rut depth 4A, 4B 4C 5A Figure 4.20. Differential rut depths versus gyrat ory slope of the coarse-graded mixture of the top lift.

PAGE 75

62 y = -0.1256x + 15.346 R2 = 0.0305 0 2 4 6 8 10 12 14 16 18 051015202530 rut depth vrs slope Figure 4.21. Differential rut depths versus gyrat ory slope of the fine-graded mixtures of the top lift The rutting patterns of the fine-graded mi xtures in relation to the slope of the gyratory shear stress and the vertical failure strains are not good. The gyratory shear slope ranged from 16.79 kpa for sections 3A and 3B to 34.5 for lane 1 where there no was rutting study. The differential rut dept hs recorded for the fine-graded sections ranged from 9.8 mm for section 2A to 16.9 mm for section 3C. Figure 4.21 shows the relation between the slope and the differential ru t depth of the fine-graded mixture. The R2 obtained was 0.03 implying a very weak correlation. According to Darku (2003) mixtures outsi de the optimal zone of Figure 4.19 such as the fine-graded mixtures may show rutting an d or cracking with load application early in-service. 4.9 Evaluation of Gradation of HVS Track Mixture Figure 4.22 shows the gradation of the top li ft gradation for both the fine-graded and coarse-graded mixtures.

PAGE 76

63 0 10 20 30 40 50 60 70 80 90 100 0.000.501.001.502.002.503.003.504.00 SIEVE SIZE^0.45PERCENTAGE PASSING ( % JMF fine truck 7 truck 8 truck 6 truck 5 truck 4 truck 3 coarse CONTROL POINTS RZ truck 2 coarse truck 1 coarse JMF coarse Figure 4.22. Gradation of the fine-graded and co arse-graded plant mix mixtures of the top lift. Gradation is the most important parameter th at affects the resistance of the mixtures to carry load. Aggregate in a mixture may work together to resist deformation when they interact among themselves especially the coarse fractio ns. Coarse fraction in a mixture is defined as the materi als retained on the 1.18 mm sieve. Lambe and Whitman (1969) presented data to show that the maximum porosity of the loose dry soil in contact ranged from 45% to around 50%. In this state, particles interaction is at the minimum and the resistan ce to deformation is also at the minimum. Asphalt concrete mixtures are in general l oose soils particles gl ued together with asphalt binder and could be likened to soils in the way they pack. Figures 4.23 and 4.24 show the unit interaction plots of the fine-graded mixture.

PAGE 77

64 0 10 20 30 40 50 60 70 80 90 100 12.59.5 9.54.75 4.752.36 2.361.18 1.180.6 0.6-0.30.30.15 0.150.075 0.075-0 contiguous sizes, mmBig particle % retained JMF T-1-A T-1-B T-1-C T-2-A Limits Figure 4.23. Interaction unit check for lane 1 and section 2A of lane 2 of the fine-graded mixture top lift. 0 10 20 30 40 50 60 70 80 90 100 12.59.5 9.54.75 4.752.36 2.361.18 1.180.6 0.6-0.30.30.15 0.150.075 0.075-0 contiguous sizes, mmBig particle % retained T-2-B T-2-C T-3-A T-3-B T-3-C Limits Figure 4.24. Interaction unit check for lanes 2 a nd 3 for the fine-graded mixture top lift. Analysis performed by the Materials Group at The University of Florida on the spacing characteristics of aggregates in contact showed aggregates particles interact as a

PAGE 78

65 unit to resist loads if the re lative proportion of one of the sieves size in two contiguous sieves is between 70 and 30%. Fi gure 4.25 shows the porosity of the fine-graded mixtures for the range of particles between 4.75 mm and 1.18 mm for no interaction and interact ion among the particles and their relation with the differential rut depth of fine-grade mixture. 0 10 20 30 40 50 60 70 2A2B2C3A3B3CSectionsPorosity,Differential Rut Depth (%,mm) No.interaction With interaction Differential rut depth No interacttion With interaction Figure 4.25 Porosity of the fine-graded mi xture for interaction and no interaction. Considering the no interaction for the size range 4.75 mm to 1.182 mm, the porosity calculated seems to relate the differe ntial rut depth shown in Figure 4.25. That is the higher the porosity the higher the diffe rential rut depth. Where interaction is considered, the unit interaction plot show that section 2C a nd 3A were the most out of tolerance (close 70%) meaning that the stre ss resisting interacti on is less and hence higher susceptibility to differentia l rutting and higher rut depths.

PAGE 79

66 0 10 20 30 40 50 60 70 80 90 100 12.59.5 9.54.75 4.752.36 2.361.18 1.180.6 0.6-0.30.30.15 0.150.075 0.0750 contiguous sizes, mmBig particle % retained JMF T-4-A T-4-B T-4-C T-5-A Limits Figure 4.26 Unit interaction plot for the co arse-graded mixture of the top lift. The coarse-graded mixtures show sim ilar trend. Figure 4.26 shows the unit interaction plot for the co arse-graded mixture for s ection 4A, 4B, 4C and 5A. The differential rut depth was 17.1 mm and 15.8 mm for sections 5A and 4C respectively. Considering interaction for the two conti guous sieves of 2.36 mm and 1.18 mm range, 5A was the most out tolerance followe d by 4C. Thus the relative potential of mixtures to rut could be predicted during design using the combin ation of porosity and unit interaction calculated for the coarse aggregate fractions in any mixture. 4.10 Evaluation of SuperPaveTM Indirect Tensile Test Results The results of the SuperPaveTM IDT test included the re silient modulus, m-value, strength and dissipated creep strain energy. These parameters were used to calculate the energy ratio (ER) of the mixtures evaluated.

PAGE 80

67 The energy ratio is defined as the dissip ated creep strain energy of the mixture divided by the minimum threshold dissipated cr eep strain energy to resist cracking. A mixture with a low ER would a hi gher potential to top-down cracking. Table 4.6 shows the results of IDT test parameters for Truck 1-3 for the coarsegraded mixtures and that of fine-grade d sections represented by Trucks 4-8. Table 4.6. Indirect tensile test parame ter versus servopac test parameters. Truck No IDm-valueD1St (Mpa) MR (Gpa) FE (kJ/m3) DCSE HMA (kJ/m3) Stress (psi) a DCSE MIN (kJ/m3)ERSlope kPa Vertical strain 50.503945E-071.710.761.10.9657061505.05-E81.3983910.69058 60.45234E-071.6513.576670.80.6997361505.08E-80.7423620.94258 50.403225E-072.0511.686671.61.4202011504.9E-080.7530771.88587 60.467454E-072.2912.293332.72.486711504.7E-080.8598452.89204 50.462534E-072.1411.3866721.7989051504.8E-080.884442.03395 60.485676E-071.9810.752.72.5176561504.9E-081.4637761.71997 50.439494E-072.3512.966672.11.887051504.7E-080.7908472.38611 60.418184E-072.3112.541.61.3872371504.7E-080.6997971.98234 5* 60.471094E-072.1213.423331.81.632591504.8E-080.9046061.80475 50.41744E-072.1311.621.51.3047811504.8E-080.648062.01336 60.442225E-072.3213.046672.32.0937251504.7E-080.9015372.3224 50.504995E-072.5211.993.83.5351791504.6E-081.2803892.76102 60.523774E-072.3513.266672.82.5918661504.7E-081.2088242.14412 5* 60.500595E-072.4212.063334.34.0572641504.7E-081.2376163.27829 6 1 2 7.31.4 13.21.4 7.81.4 7 8 3 4 5 16.81.25 241.15 17.71.1 13.91.1 34.51.2 The ER calculated ranged from 0.691 to 2.89 with an average of 1.69 for the coarse-graded mixture and from 1.8 to 3.28 w ith an average of 2.34 for the fine-graded mixture. With exception of the specimen from Truck 1 the re st of the mixtures evaluated had ER greater than 1, the threshold below which mixtures are susceptible to top-down cracking. Generally the fine-g raded mixture appears to have higher ER than the coarsegraded however statistically th ere are no significant differenc es in the means of the ER and the failure strains. The relationship between ER of the mixtures and the gyratory

PAGE 81

68 shear slope of both mixtures was found to be fair with R2 of 0.5049 as shown in Figure 4.23. y = 9.1228x 2.4521 R2 = 0.5049 0 5 10 15 20 25 30 35 40 01234 Energy ratioGyratory slope (kPa ) Figure 4.27. Relationship between gyratory shea r slope and the energy ratio of both the fine-graded and coarse-graded mixtures. 4.11 Further Evaluation of the Fine-Graded Mixture Further evaluation of the di fferential rut depth was done to compare the fine-graded mixture to coarse-graded mixtur e after eliminating sections 3C and 5A. Sections 3C and 5A recorded the highest differe ntial rut depth after running th e HVS. Section 3C and 5A recorded 16.9 mm and 17.1 mm differential rut depths respec tively. The relatively high rut depths of these two sections could be attr ibuted to the higher coarse-fraction porosity or the marginal interacting unit checks as shown in Figures 4.24 and 4.25. Figure 4.28 shows the differential rut depth for the fine-g raded and coarse-graded mixtures with the exclusion of sections 3C and 5A. The aver age differential rut depths of the fine-graded and coarse-graded mixtures ar e 11.3 mm and 13.22 mm, respectively.

PAGE 82

69 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 2A2B2C3A3BAVFAVC4A4B4C5B5C SectionsDifferential rut depth (mm) Fine Coarse Figure 4.28. Comparison of fine-graded and coarse-graded differential rut depths. Comparing the difference between the mean rut depths of the fine-graded mixture and the coarse-graded, we realized that statis tically there was no signi ficant difference in the mean differential rut dept h at 95% confidence level. 4.12 Comparison of the Fine-graded Mi xture of HVS Round 1 to HVS Round 3 Research conducted at the FDOT Materials Office in Gainesville in 2002 compared the rutting resistance of fine-graded SuperPaveTM mixtures with and without SBS polymer modified binder and the conclusion drawn was that, the SBS modified mixture out-performed the mixture without SBS polymer modified binder. In our analysis we compared the rutting re sistance of the fine-graded mixture to the fine-graded unmodified mixtur e in “HVS Round 1” using th e differential rut depths at 50000 passes of the test wheel. The HVS runni ng modes and the temperatures of testing were the same in the two projects. The type of aggregates used for both projects were different. “HVS Round 1” was constructed with limestone from south Florida whilst

PAGE 83

70 “HVS Round 3” was constructed with granite from Georgia. Figure 4.29 and 4.30 shows the gradation of the Job mix formula and the differential rut profiles for both fine-graded mixtures. 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0 0.000.501.001.502.002.503.003.504.00 Sieve size (mm) ^0.45Percentage Passing (% ) JMF FINE HVS # 3 JMF FINE HVS #1 Figure 4.29. Job mix formula HVS Round 1 a nd HVS Round 3 for the fine-graded mixtures. 0.0 5.0 10.0 15.0 20.0 25.0 0.020000.040000.060000.080000.0100000.0120000.0 NUMBER OF PASSESRUT DEPTH (mm ) 3A round #3 3B round #3 3C round #3 2C round #3 2B round #3 2A round #3 4A round #1 4B round #1 5A round #1 5B round #1 Figure 4.30. Differential rut depth of the fi ne-graded mixture HVS Round 1 and HVS Round 3

PAGE 84

71 The effective asphalt content of the JMF for “HVS Round 1” was 4.9% and that of HVS Round 3 was 4.5%. Differential rut depth recorded in HVS Round 1 (lanes 4 and 5) appears to be higher than that HVS Round 3 (lanes 2 and 3) at 50000 passes. HVS Round 1 recorded a minimum of 13.2 mm and a maximum 19.5 mm with an average of 15.8 mm whilst HVS Round 3 recorded a minimum and a maximu m of 8.9 mm and 14.9 respectively with an average of 11.3 mm. Gradation of the two mixtures are differ ent and is the parameter with significant influence on mixture or pavement rutting perf ormance that could have contributed to the differences in rutting performance between HVS Round 1 and HVS Round 3. Analysis of the gradations using th e interacting unit check and the porosity of the coarse aggregate-fraction (12.5 mm to 1.18 mm) were done to see if any differences exist for the two mixtures. Figures 4.31 and 4.32 show th e interacting unit check and porosity plots of the gradations of the JMF of the two mixt ures as well as lanes 4 and 5 of “HVS Round 1”. The interacting unit check indicate s that aggregate size-range 4.75 mm to 1.18 mm are acting together to resist stresses within the two mixtures. In both cases, there is a break in interaction at 9.5 mm to 4.75 mm. “HVS Round 1 mixtures (JMF and lanes 4 and 5) aggregates-size ranging from 12.5 mm to 9.5 mm are floating in the mixture and do not contribute shearing stress resistance of the mixture. For the interaction at 2.36 mm to 1.18 mm HVS Round 1 mi xtures are closer to the tolerance limits and has higher porosities at all interaction ranges than HVS Round 3 mixture which makes it more susceptible to rut ting. It is clear from the above analyses

PAGE 85

72 that HVS Round 3fine-graded mixture has be tter rutting resistan ce than HVS Round 1 unmodified fine-graded mixture. 12.5-9.5 9.5-4.75 4.75-2.36 2.36-1.18 1.18-0.6 0.6-0.3 0.3-0.15 0.15-0.075 0.075-0Contiguous sieve sizes, mmLarge/Small Particle Proportion hvs 3 JMF hvs 1JMF Limits lane 4 lane 5 100/0 90/10 70/30 80/20 40/60 30/70 20/80 10/90 0/100 60/40 50/50 Figure 4.31. Interacting unit check for “JMF” “HVS Round 1 and 3” and lanes 4 and 5. 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 2.36-1.184.75-1.189.5-1.1812.5-.1.18 Contiguous sizes (mm)Porosity (% ) HVS3 HVS 1 Figure 4.32. Porosity of job mix formula for “HVS Round 1” and “HVS Round 3” fine-graded mixture.

PAGE 86

73 4.13 Condition Survey Results The condition survey undertaken on all the 12 sections showed no visible tire imprints, raveling or cracking except section 3B where two longitudinal cracks of width of about 0.5mm and length 4 – 5 inches have appeared. Figure 4.33 shows one of such cracks. Figure 4.33. Longitudinal crack on section 3B When cores were taken, it was realized that the crack wa s induced by a strain gage inserted between the bottom lift and the top lif t for strain measurement. It was also realized that the strain gages in the locations of cracks were nearer to the surface of the top lift. The mixture seemed loose around the gages suggesting some amount of segregation had occurred during the laying process. Segrega tion of the mixture could be observed on the sides of the cores taken at the section at about 1 inch to 2 inches of the top lift. The mixture had also stripped s lightly and the fines pumped to the surface as shown in Figure 4.33. Figure 4.34 shows the core with the strain-gag e induced crack at

PAGE 87

74 the top and mix segregation around the strain ga ge as well as the joint between the top lift and the bottom lift. Figure 4.34. Strain gage induced crack on th e top lift of sectio n 3-3B fine-graded mixture.

PAGE 88

75 CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS 5.1 Conclusion Analysis of the results of both the aver age differential and absolute rut depth measurement show that the difference in the mean rut depths of the coarse-graded and fine-graded mixture was st atistically insignificant. From the limited data collected, we can not conclude that the SuperPaveTM Coarse-graded mixture is either be tter or worse than the SuperPaveTM fine-graded mixture. When the transverse profiles were eval uated, the Area Change Parameter results show that both the fine-grade d and coarse-graded mixture e xhibited primarily instability rutting. The shape of the tr ansverse profile at 100 passes of the HVS wheel shows that the instability rutting of both mixtures develope d relatively early in the service life of the pavements. Based on the limited Servopac test results, the plots of the gyratory shear slope versus the initial vertical failure strain for the fine-graded and coarse-graded mixture show that they were both “brittle”. Th e fine-graded mixture however had relatively higher gyratory shear strength a nd lower strain potential than the coarse-graded mixture. Examination of the test results of the vi scosities of the recovered asphalt from the cores were very close to another and that its effect on mixture rutting could not be effectively evaluated

PAGE 89

76 Previous conclusions based on the Interac ting Unit Check and their Porosity effects on mixture performances were verified in th is study. The usefulness of the gradation guideless in explaining differen ces in pavements and mixtures performance could be used for the evaluation of mixtures during mix design. Results of the APA test data did not show any clear differences between the Fine-graded and coarse graded mixtures a nd also between APA rut depth and HVS rut depths. Results of the IDT test parameters sh ow no statistical si gnificant difference between the fine-graded and the coarse grad ed mixtures top-down cracking potential. The fine-graded mixture used in this study appears to have a better rutting resistance than an unmodified fine-gra ded mixture used in a previous study. 5.2 Recommendation There is need to do more evaluation of th e coarse-graded and fine-graded mixtures, so that a larger statistical would be availa ble for evaluation of the effects of aggregate gradation on rutting resistance. Evaluation of different grades of asphalt using the same aggregates source and gradation should be investigated for the evaluatio n of instability rutting. Hard grades of asphalts processed in such as to control the effects of ageing for cracking control are on the market and provide an alternative for cont rol of instability rutt ing whilst maintaining its ductility.

PAGE 90

77 APPENDIX A MIX DESIGNS Address Fax No.E-mail Type Mix D 100 F.D.O.T. CODEPIT NO. TM-561 1. 43GA-553 TM-561 2. GA-553 TM-561 3. GA-553 Starvation 4. Hill 5. 6. 29%35%28%8% JOB MIX 123456FORMULA3/4" 19.0mm100100100100 100 100 E1/2" 12.5mm94100100100 9890-100 Z3/8" 9.5mm66100100100 90 -90 INo. 4 4.75mm183798100 54 SNo. 8 2.36mm6771100 3228-5839.1-39.1No. 16 1.18mm3345100 23 25.6-31.6ENo. 30 600m323093 17 19.1-23.1VNo. 50 300m222049 11 ENo. 100 150m211210 5 INo. 200 75m1.51.07.63.6 4.52-10 SGSB2.8092.7992.7702.626 2.779 LD 04-2543A (TL-D) StructuralThe mix properties of the Job Mix Formula have been conditionally verified, pending successful final verification during produc tion at the assigned plant, the mix design is approved subject to F.D.O.T. specifications.Local Sand 20 12 / 22 / 2004 12 / 22 / 2004 Junction City Mining, L.L.C. 12 / 22 / 2004 Junction City Mining, L.L.C. 51 Howie MoseleySP-12.5 Intended Use of Mix Thomas O. MalerkOri g inal document retained at the State Materials OfficeNo. 200 reflects aggregate changes expected during production. Director, State Materials Office STATE OF FLORIDA DEPARTMENT OF TRANSPORTATION STATEMENT OF SOURCE OF MATERIALS AND JOB MIX FORMULA FOR BITUMINOUS CONCRETE SUBMIT TO THE STATE MATERIALS ENGINEER, CENTRAL BITUMINOUS LABORATORY, 5007 NORTHEAST 39TH AVENUE, GAINESVILLE, FLA. 32609Contractor V. E. Whitehurst & Sons2230 N. W. 73rd Place, Gainesville, FL 32653 Phone No. Design Traffic Level Submitted By Gyrations @ N des (352) 573-3816(352) 373-3314 Coarse TYPE MATERIALPRODUCERDATE SAMPLED PERCENTAGE BY WEIGHT TOTAL AGGREGATE PASSING SIEVES #78 Stone #89 Stone W-10 Screenings PG 67-22 12 / 22 / 2004 Blend Number CONTROL POINTS RESTRICTED ZONE Junction City Mining, L.L.C. V.E. Whitehurst & Sons A-1 Gradation and sources of a ggregate for the coarse mixture

PAGE 91

78 VaVMAVFA %Gmm @ Nmax4.014.673 97.2 3.714.675 97.5 3.214.578 98.0 4.5% 46 %F 154 C 155.1 Lbs/Ft 3 2485 Kg/m396.0F 154 C 14.6% -0.23 Additives %% 87.0 87.3 87.7 1.0 1.0 2.485 2.488 2.495 4.5 4.6 4.8 310 2.589 2.583 2.577 Gmb @ NdesP0.075 / PbePbGmm Pbe4.5 4.7 1.0 4.4 HOT MIX DESIGN DATA SHEET LD 04-2543A (TL-D) NCAT Oven Calibration Factor VMA Total Binder ContentFAA %Gmm @ Nini Compaction Temperature 310Ar-Maz AdHere LOF 65-00 (M-0014)%Gmm @ NdesLab. Density(+To Be Added)/(-To Be Subtracted) Mixing Temperature Antistrip 0.75 95.9 96.1 96.3 96.5 96.7 96.9 4.04.55.0 % Asphalt %Gmm@Ndes 14.3 14.4 14.5 14.6 14.7 14.8 4.04.55.0 % Asphalt %VMA 72 73 75 76 78 79 4.04.55.0 % Asphalt %VFA A-2. Hot mix design data sheet for coarse-graded mixture

PAGE 92

79 Address Fax No.E-mail Type Mix D 100 F.D.O.T. CODEPIT NO. TM-561 1. 43GA-553 TM-561 2. GA-553 TM-561 3. GA-553 Starvation 4. Hill 5. 6. 28%12%50%10% JOB MIX 123456FORMULA3/4" 19.0mm100100100100 100 100 E1/2" 12.5mm94100100100 9890-100 Z3/8" 9.5mm66100100100 90 -90 INo. 4 4.75mm183798100 68 SNo. 8 2.36mm6771100 4828-5839.1-39.1No. 16 1.18mm3345100 34 25.6-31.6ENo. 30 600m323093 25 19.1-23.1VNo. 50 300m222049 16 ENo. 100 150m211210 8 INo. 200 75m1.51.07.63.6 4.92-10 SGSB2.8092.7992.7702.626 2.769 LD 04-2544A (TL-D) RESTRICTED ZONE Junction City Mining, L.L.C. V.E. Whitehurst & Sons Blend Number CONTROL POINTS TYPE MATERIALPRODUCERDATE SAMPLED PERCENTAGE BY WEIGHT TOTAL AGGREGATE PASSING SIEVES #78 Stone #89 Stone W-10 Screenings PG 67-22 12 / 22 / 2004 Phone No. Design Traffic Level Submitted By Gyrations @ N des (352) 573-3816(352) 373-3314 Fine STATE OF FLORIDA DEPARTMENT OF TRANSPORTATION STATEMENT OF SOURCE OF MATERIALS AND JOB MIX FORMULA FOR BITUMINOUS CONCRETE SUBMIT TO THE STATE MATERIALS ENGINEER, CENTRAL BITUMINOUS LABORATORY, 5007 NORTHEAST 39TH AVENUE, GAINESVILLE, FLA. 32609Contractor V. E. Whitehurst & Sons2230 N. W. 73rd Place, Gainesville, FL 32653 Thomas O. MalerkOri g inal document retained at the State Materials OfficeNo. 200 reflects aggregate changes expected during production. Director, State Materials Office Howie MoseleySP-12.5 Intended Use of Mix 20 12 / 22 / 2004 12 / 22 / 2004 Junction City Mining, L.L.C. 12 / 22 / 2004 Junction City Mining, L.L.C. 51 StructuralThe mix properties of the Job Mix Formula have been conditionally verified, pending successful final verification during produc tion at the assigned plant, the mix design is approved subject to F.D.O.T. specifications.Local Sand A-3. Gradation and sources of aggreg ate for the fine-graded mixture

PAGE 93

80 VaVMAVFA %Gmm @ Nmax4.214.771 97.2 4.014.773 97.0 4.6% 46 %F 154 C 154.4 Lbs/Ft 3 2475 Kg/m396.0F 154 C 14.7% -0.28 Additives %% %Gmm @ NdesLab. Density(+To Be Added)/(-To Be Subtracted) Mixing Temperature Antistrip 0.75 Compaction Temperature 310Ar-Maz AdHere LOF 65-00 (M-0014)HOT MIX DESIGN DATA SHEET LD 04-2544A (TL-D) NCAT Oven Calibration Factor VMA Total Binder ContentFAA %Gmm @ Nini Pbe4.5 1.1 4.4 P0.075 / PbePbGmmGmb @ Ndes310 2.583 2.579 4.5 4.6 2.474 2.475 88.4 88.6 1.1 95.7 95.8 95.9 95.9 96.0 96.1 4.14.65.1 % Asphalt%Gmm @ Ndes 14.4 14.5 14.6 14.7 14.8 14.9 4.14.65.1 % Asphalt% VMA 70 71 72 72 73 74 4.14.65.1 % Asphalt% VFA A-4. Hot mix data sheet for the fine-graded mixture

PAGE 94

81 APPENDIX B RECOVERED VISCOSITIES VISCOSITY OF RECOVERED ASPHALT SECTION: 3ALIFT:BOTTOM MIXTURE TYPE:FINE SPECIMEN ID: A SPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE RPM%D/cm2 SEC-1 C 12750001.215.357120.7160.2 11880954.249.9169662.1460.2 11323076.573.6250243.660.2 11642854.248.9166262.1460.2 12333331.214.850320.7160.2 AVERAGE1198604 VISCOSITY OF RECOVERED ASPHALT SECTION: 3ALIFT:BOTTOM MIXTURE TYPE:FINE SPECIMEN ID: A3 SPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE RPM%D/cm2 SEC-1 C 8000002.116.852020.460.2 7523806.347.4161161.4360.2 71509410.675.8257722.2160.2 7380956.346.6158101.4360.2 7857142.116.556100.4160.2 AVERAGE758256.6 VISCOSITY OF RECOVERED ASPHALT SECTION: 3ALIFT:BOTTOM MIXTURE TYPE:FINE SPECIMEN ID: B SPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE RPM%D/cm2 SEC-1 C 22285710.715.653040.2160.2 1983333359.5202301.0260.2 18666663.667.2228481.2260.2 1866666356190401.0260.2 19857140.713.947260.2460.2 AVERAGE1986190 Cp VISCOSITY Cp VISCOSITY Cp VISCOSITY B-1. Viscosity of recovered asphalt section 3A bottom lift

PAGE 95

82 VISCOSITY OF RECOVERED ASPHALT SECTION: 3BLIFT:TOP MIXTURE TYPE:FINE SPECIMEN ID: I SPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE RPM%D/cm2 SEC-1 C 28833330.617.358820.260.2 27277781.849.1166940.6160.2 25419353.178.8267921.0560.2 26222221.847.2160480.6160.2 27833330.616.756780.260.2 AVERAGE2711720.2 VISCOSITY OF RECOVERED ASPHALT SECTION: 3BLIFT:TOP MIXTURE TYPE:FINE SPECIMEN ID: I3 SPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE RPM%D/cm2 SEC-1 C 27500000.616.556100.260.2 25000002.152.5178500.760.2 23757583.378.4266561.1260.2 24047622.150.5171700.7160.2 26000000.615.653040.260.2 AVERAGE2526104 VISCOSITY OF RECOVERED ASPHALT SECTION: 3BLIFT:TOP MIXTURE TYPE:FINE SPECIMEN ID:J SPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE RPM%D/cm2 SEC-1 C 32600000.516.355420.1760.2 28705881.748.8165920.5860.2 27655172.980.2272680.9960.2 28058821.747.7162180.5860.2 31000000.515.552700.1760.2 AVERAGE2960397.4 cP VISCOSITY Cp VISCOSITY cP VISCOSITY B-2. Viscosity of recovered asphalt section 3B top lift

PAGE 96

83 VISCOSITY OF RECOVERED ASPHALT SECTION: 3CLIFT:TOP MIXTURE TYPE:FINE SPECIMEN ID: E SPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE RPM%D/cm2 SEC-1 C 45000000.522.576500.1760.1 41666671.250170000.4160.2 3915000278.3266220.6860.2 40083331.248.1163540.4160.1 42200000.521.171740.1760.2 AVERAGE4162000 VISCOSITY OF RECOVERED ASPHALT SECTION: 3CLIFT:TOP MIXTURE TYPE:FINE SPECIMEN ID: E3 SPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE RPM%D/cm2 SEC-1 C 49000000.314.749980.160.2 44333331.253.2180880.4160.2 42100002.284.2286280.6860.2 43750001.252.5178500.4160.2 48000000.314.448960.160.2 AVERAGE4543666.6 VISCOSITY OF RECOVERED ASPHALT SECTION: 3CLIFT:TOP MIXTURE TYPE:FINE SPECIMEN ID: F SPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE RPM%D/cm2 SEC-1 C 32600000.516.355420.1760.2 28705881.748.8165920.5860.2 27655172.980.2272680.9960.2 28058821.747.7162180.5860.2 31000000.515.552700.1760.2 AVERAGE2960397.4 cP VISCOSITY Cp VISCOSITY cP VISCOSITY B-3. Viscosity of recovered asphalt section 3C top lift

PAGE 97

84 VISCOSITY OF RECOVERED ASPHALT SECTION: 5BLIFT:TOP MIXTURE TYPE:COARSESPECIMEN ID:K SPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE RPM%D/cm2 SEC-1 C 30600000.515.352020.1760.2 28187501.645.1153340.5460.2 26807692.669.7236980.8860.2 28000001.644.8152320.5460.2 29800000.514.950660.1760.2 AVERAGE2867903.8 VISCOSITY OF RECOVERED ASPHALT SECTION:5BLIFT:TOP MIXTURE TYPE:COARSESPECIMEN ID: K3 SPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE RPM%D/cm2 SEC-1 C 27000000.616.255080.260.2 24791672.459.5202300.8260.2 24000003.481.6277441.1660.2 24625002.459.1200940.8260.2 27166670.616.355420.260.2 AVERAGE2551666.8 VISCOSITY OF RECOVERED ASPHALT SECTION:5BLIFT:TOP MIXTURE TYPE:COARSESPECIMEN ID:L SPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE RPM%D/cm2 SEC-1 C 27166670.616.355420.260.2 25190482.152.9179860.7160.2 24029413.481.7277781.1660.2 24619052.151.7175780.7160.2 26333330.615.853720.260.2 AVERAGE cP VISCOSITY Cp VISCOSITY cP VISCOSITY B-4. Viscosity of recovered asphalt section 5B top lift

PAGE 98

85 VISCOSITY OF RECOVERED ASPHALT SECTION: 5ALIFT:TOP MIXTURE TYPE:COARSESPECIMEN ID:H SPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE RPM%D/cm2 SEC-1 C 7250002.014.549300.6860.2 6514297.045.6155042.3860.2 60230813.078.3266224.4260.2 6200007.043.4147562.3860.2 6800002.013.646240.6860.2 AVERAGE655747.4 VISCOSITY OF RECOVERED ASPHALT SECTION:5ALIFT:TOP MIXTURE TYPE:COARSESPECIMEN ID: G3 SPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE RPM%D/cm2 SEC-1 C 30800000.515.452360.1760.2 28500001.645.6155040.5460.2 27034482.978.4266560.9960.2 28437501.645.5154700.5460.2 30600000.515.352020.1760.2 AVERAGE2907439.6 VISCOSITY Cp VISCOSITY cP B-5. Viscosity of recovered asphalt section 5A top lift

PAGE 99

86 VISCOSITY OF RECOVERED ASPHALT SECTION:5CLIFT:BOTTOM MIXTURE TYPE:COARSESPECIMEN ID:C3 VISCOSITYSPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE cPRPM%D/cm2SEC-1 C 14545451.11654400.3760.3 13228573.546.3157421.1960.2 12527275.568.9234261.8760.3 12542853.54.9149261.1960.2 13818181.115.251680.3760.2 AVERAGE1333246.4 VISCOSITY OF RECOVERED ASPHALT SECTION:5CLIFT:BOTTOM MIXTURE TYPE:COARSESPECIMEN ID:D1 SPEEDTORQUESHEAR STRESSSHEAR RATETEMPERATURE RPM%D/cm2SEC-1 C 12666661.215.251680.4160.3 1167500446.7158781.3660.2 1081428775.7257382.3860.3 1125000445153001.3660.3 12000001.214.448960.4160.2 AVERAGE cP VISCOSITY B-6. Viscosity of recovered asphalt section 5C bottom lift

PAGE 100

87 APPENDIX C AREA CHANGE PARAMETER C-1. Area-change parameter section 2A

PAGE 101

88 C-2. Area-change parameter section 2B

PAGE 102

89 C-3. Area-change parameter section 2C

PAGE 103

90 C-4. Area-change parameter section 3A

PAGE 104

91 C-5. Area-change parameter section 3B

PAGE 105

92 C-6. Area-change parameter section 3C

PAGE 106

93 C-7. Area-change parameter section 4A

PAGE 107

94 C-8. Area-change parameter section 4B

PAGE 108

95 C-9. Area-change parameter section 4C

PAGE 109

96 C-10. Area-change parameter section 5A

PAGE 110

97 C-11. Area-change parameter section 5B

PAGE 111

98 C-12. Area-change parameter section 5C

PAGE 112

99 LIST OF REFERENCES Birgisson, B., Darku, D., Roque, R. and Page G. C. (2004) “The Need for Inducing Shear Instability to Obtain Relevant Parameters for HMA Rut-Resistance”, Journal of the Association of Asphalt Paving Technologist 73, 23-52. Buchanan, M. S. and Cooley, A.L. Jr. ( 2002) “Case Studies of the Tender Zone in Coarse-Graded Superpave Mixture”, Draft Final Report. NCAT Report 02-01. Corte, Jean-Francois. (2001) “Development and Uses of Hard-Grade Asphalt and of High-Modulus Asphalt Mixes in France”, Perpetual Bituminous Pavements Transportation Research Circular No.503, (Transportation Research Board), 12-30 Cross, S. A., Adu-Osei, A., Hainin, M. R. and Fredrichs, R. K. (1999) “Effects of Gradation on Performance of Asphalt Mixtures”, Proceedings, 78th Annual Meeting, Transportation Research Board Washington, D.C. Darku, D.D. (2003) “Evaluation of SuperPaveTM Gyratory Compactor for Assessing the Rutting Resistance of Asphalt Mixture”, PhD dissertation of the University of Florid, Gainesville. Drakos, C. A., Roque, R., Birgisson, B. a nd Novak, M. (2002) “I dentification of a Physical Model to Evaluate Rutti ng Performance of Asphalt Mixtures”, Journal of Testing and Evaluation 20, Harvey J. and Popescu, L. (2000) “Accelerated Pavement Testing of Rutting Performance of Two Caltrans Overlay Strategies”, Transportation Research Record No 1716, (Transportation Research Board), 116-125. Kandhal, P. S., Cross, S. A. and Brown, E. R. (1990) “Evaluation of Bituminous Pavements for High Pressure Truck Tires”, Final Report, FHWA Report No EHWA-PA-90-008 87-01 Kim, B., Roque, R. and Birgisson, B. (2003) “Laboratory Evaluation of Effects of SBS Modifier on Cracking Resist ance of Asphalt Mixture”, Transportation Research Record, No 1829, (Transportation Research Board), 8-15. Lea, J. and Heath, A. (1999) “Using AP T to Fast-track I nnovative Materials”, International Conference for Accelerated Pavement Testing, Reno, Nevada

PAGE 113

100 Moseley, H. L. (1999) “An Evaluation of SuperPaveTM Compaction and Asphalt Mixture Properties”, M.E thesis of the Univers ity of Florida, Gainesville Myers, L. A. and Roque, R. (2001) “Evaluat ion of Top-Down Cracking in Thick Asphalt Pavements and the Implications for Pavement Design”, Perpetual Bituminous Pavements, Transportation Research Circular, 503, (Transportation Research Board), 79-87 Nukunya, B, Roque, R., Tia, M. and Birgiss on, B. (2001) “Evaluation of VMA and other Volumetric Properties as Criteria for the Design and Acceptance of Superpave Mixtures”, Journal of the A ssociation of Aspha lt Paving Technologist 70, 38-69. Nuureldin, A. S. (1997) “Construction Consid eration in the Field Application of the Superpave Mixture Design System”. Progress of SuperPaveTM: Evaluation and Implementation, ASTM STP 1322 Robertson, W. D. (1997) “Determining th e Winter Design Temperature for Asphalt Pavements”, Journal of the Association of Asphalt Paving Technologist 66, 312-337. Ruth, B. E., Roque, R. and Nukunya, B. ( 2000) “Aggregate Gradation Characterization Factors and Their Relationships to Fr acture and Failure Strain of Asphalt Mixtures”, Journal of the A ssociation of Aspha lt Paving Technologist 69, 310-344. Sargand, S M and Kim, Sang-soo (2003) “Per formance Evaluation of Polymer Modified Mixes Using Laboatory Test and Accelerated Pavement Loading Facility”, Proceedings, 82nd Annual Meeting, Transportation Research Board, Washington D C. Sholar, G A., Page G C., Musselman, J A. a nd Moseley, H L. (2004) “Evaluating the Use of Lower VMA Requirements for SuperPaveTM Mixtures”, Research Report FL/DOT/SMO/04-479. Stroup-Gardinar, M. and Law, M. (2000) “Influence of Segregation in Pavement Performance” Journal of the Association of Asphalt Paving Technologist 69 313-339. Stuart, K D., Mogawer, S W. (1997) “Valida tion of Asphalt Binder and Mixture Test that Predicts Rutting Susceptibility Using the FHWA ALF”, Journal of the Association of Asphalt Paving Technologist 66, 109-152. Tia, M., Roque, R., Sirin, O., Kim, H ong-Joong. (2002) “Evaluation of Superpave Mixtures with and without Polymer Modification by means of Accelerated Pavement Testing”, Final Report U. F Project No: 49104504801-12.

PAGE 114

101 Tia, M., Ruth, B E. (1985) “Basic Rheological Concepts Established by H. E Schweyer Asphalt Rheology, Relationship to Mixture”, Special Technical Publication, ASTM STP 941. Villiers, C., Roque, R. and Diet rich, B. (2005) “Interpretatio n of Transverse Profile to Determine the Source of Rutting within Asphalt Pavement System”, Transportation Research Record No 1950 (Transportation Research Board), 73-81. Villiers, C. (2004) “Evaluati on of Sensitivity of SuperPaveTM Mixtures for Development of Performance Related Specification”, PhD dissertation of the University of Florida, Gainesville. Wu, Z., Hossain, M. and Gisi, A J. (2000) “Performance of Superpave Mixtures under Accelerated Load Testing”, Transportation Research Record No 1716 (Transportation Research Board), 126-134.

PAGE 115

102 BIOGRAPHICAL SKETCH Collins Boadu Donkor was born on August 22, 1964 in Kumasi-Ghana. In 1990, he obtained his Bachelor of Science degr ee in civil engineering from the Kwame Nkrumah University of Science and Technol ogy. He was employed and continued to work for the Ghana Highway Authority of the Ministry of Roads and Transport, Ghana.


Permanent Link: http://ufdc.ufl.edu/UFE0013324/00001

Material Information

Title: Evaluation of Rut Resistance of Superpave Fine-Graded and Coarse-Graded Mixtures
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013324:00001

Permanent Link: http://ufdc.ufl.edu/UFE0013324/00001

Material Information

Title: Evaluation of Rut Resistance of Superpave Fine-Graded and Coarse-Graded Mixtures
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013324:00001


This item has the following downloads:


Full Text












EVALUATION OF RUT RESISTANCE OF SUPERPAVET FINE-GRADED AND
COARSE-GRADED MIXTURES








By

COLLINS BOADU DONKOR


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ENGINEERING

UNIVERSITY OF FLORIDA


2005



























Copyright 2005

by

Collins Boadu Donkor

































To John Osei-Asamoah, Chief Director of the Ministry of Roads and Transport-Ghana














ACKNOWLEDGMENTS

My sincere thanks go to my Supervisory Committee Chair Dr. Mang Tia for his

directions, support, and encouragement in seeing me through this course. To the other

members of my Committee, Dr Reynaldo Roque and Dr Bjorn Birgisson, I say thanks for

all the things you have done for me during my 16-month stay at UF. To Dr. Drakos,

George Lopp, and Tanya Riedhammer, I say thank you for the help you gave during the

testing and data analysis phase for this project.

To friends at the FDOT Materials Office, Greg Sholar, Howie Moseley, Steve Ross

and Salil Gokhale, thanks go to you for allowing me to use your data and facilities and

also for the help with my data analysis.

To all my friends in the University and colleagues at the Materials Department, I

thank you and may God bless all of you.

Finally to my sister, Linda, cousin, Christina, and brother-in-law, Peter, it would

have been very difficult without your encouragement and support. I thank you very

much.
















TABLE OF CONTENTS



L IS T O F T A B L E S ........................... ... .. .. .... .. ...................................................... v iii

LIST OF FIGURES ......... ......................... ...... ........ ............ ix

ABSTRACT .............. ............................................. xii

CHAPTER

1 IN TR OD U CTION ............................................... .. ......................... ..

1.1 B ack g rou n d ................................................................. .. ......................... . 1
1.2 R research H ypothesis......... ........................................ ...... ........ ............. 3
1.3 O bjectiv es ....................................................... 4
1.4 Scope .................. ...................5.............................
1.5 Research Approach ...................................... ....... ........ .. ........ ..

2 LITER A TU R E REV IEW ............................................................. ....................... 7

2.0 Introduction .................................................... ..... .................................
2.1 Characteristics of Mixture Constituents..........................................................8
2.1.1 A asphalt .................................................................. .......... 8
2.1.2 A ggregates ............................................................ 9
2.2 Traffic L loading .................. .................................... .. ........ .... 10
2.3 Environm ental Effects ......................................................... .............. 11
2.4 C construction .................................... ........................ ...... ......12
2.5 R ut M easurem ent ...................................................... ....... ................ .. 13
2.5.1 Non-contact Laser Height-Sensor Rut Depth Measurement .................13
2.5.2 D ifferential R ut D epth ........................................ ........ ............... 14
2.5.3 A absolute R ut D epth................................... ........................ .................. 14
2.6 Mixture Response Characteristics or Parameters .........................................15
2.6.1 Indirect Tension Test (IDT) ................ .......................... ............... 15
2.6.2 Servopac SuperPaveTM Gyratory Compaction............................. 16
2.6.3 Gyratory Testing Machine GTMTM ...............................................17
2.6.4 A asphalt Pavem ent A nalyzer........................................ ............... 17
2.7 A accelerated Pavem ent Testing...................................... ........ ............... 19









3 MATERIALS AND METHODOLOGY.............................................. 21

3 .0 Intro du action .......................................................................... 2 1
3.1 Sum m ary of M ethodology .................................... ........................... .......... 21
3.2 Test Track L ayout .................. .......................... .................... .. 22
3.3 Pavement Structures....................................................... 23
3.4 Asphalt Concrete Mixtures used ....................................... ...............24
3 .4 .1 A g g reg ate s ....................................................................................... 2 7
3.4.2 A asphalt ...................... ............. .............. 28
3.5 Test Tracks Construction and Instrumentation........................... .............28
3.6 Heavy Vehicle Simulator Test Configuration and Instrumentation ................29
3.7 L aser Profiler .......................................... .. .. ............. ......... 30
3.8 Trafficking .............................................. .................................... 31
3.9 Heating and Temperature Control ....................................... ............... 31
3.10 R ut M easurem ent ................................................... .......... .......... .......... 33
3.11 Air Void Content and Thickness Changes ................................................34
3.12 Determination of Viscosity of Cores ........................................ ................35
3.13 Laboratory Testing on Plant and Laboratory Prepared Mixtures ....................35
3.13.1 Asphalt Pavem ent Analyzer Test............................................. 36
3.13.2 Servopac Gyratory Compactor Testing ............................................ 36
3.13.3 SuperPaveTM Indirect Tensile Test .......................................... 37
3.14 C condition Survey ......................... .. .................... .. ...... ........... 37

4 RESULTS, ANALYSIS AND DISCUSSION.........................................................39

4.0 E valuation of R ut Profiles.................................................................... ..... 39
4 .1 In tro d u ctio n ................................................................................................ 3 9
4 .2 R ut D epth ................................................................ ................... 39
4.3 T ransverse R ut P rofile .......................... .................. ......................... .... 42
4.4 Area Parameter Change Method Evaluation of Transverse Rut Profiles ........46
4.5 Evaluation of Core D ensities ........................................ ........ ............... 50
4.6 Evaluation of Recovered Asphalt ...........................................................53
4.7 A sphalt Pavem ent A nalyzer Test................... .................. ................ ... 54
4.8 SuperPaveTM Servopac Gyratory Compaction Results................................57
4.9 Evaluation of Gradation of HVS Track Mixture ...........................................62
4.10 Evaluation of SuperPaveTM Indirect Tensile Test Results.............................66
4.11 Further Evaluation of the Fine-Graded M ixture............................................ 68
4.12 Comparison of the Fine-graded Mixture of HVS Round 1 to HVS Round 3..69
4.13 C condition Survey R results ........................................ .......................... 73

5 CONCLUSIONS AND RECOMMENDATIONS............................................... 75

5 .1 C o n clu sio n .................................................................... 7 5
5.2 R ecom m endation ......................... ...................... .. .. ...... ........... 76









APPENDIX

A M IX DESIGN S .................. .................. .................. ........... .. ............ 77

B RECOVERED VISCOSITIES ...........................................................................81

C AREA CHAN GE PARAM ETER ..................................................................... ..... 87

L IST O F R E F E R E N C E S ...................................... .................................... ....................99

BIOGRAPH ICAL SKETCH .............................................................. ............... 102
















LIST OF TABLES


Table p

3.1. Job mix formula and truck samples gradations, volumetric properties and air voids
for fine-graded mixture bottom lift. .............................................. ............... 25

3.2. Job mix formula and truck samples gradations, volumetric properties and air voids
for the fine-graded m ixture top lift.................................... .................................... 25

3.3. Job mix formula and truck samples gradations, volumetric properties and air voids
for the coarse-graded mixture bottom lift. .................................... .................26

3.4. Job mix formula and truck samples gradations, volumetric properties and air voids
for the coarse-graded m mixture top lift.................................... ....................... 26

3.5. Specification for the asphalt PG 67-22 used for the mixtures. ..................................28

4.1. Area change parameters of the fine-graded and coarse-graded sections .................48

4.2. Air voids level of the cores from the wheel paths and edges of the wheel paths
lane 3 fine-graded m ixture sections. ............................................. ............... 51

4.3. Air voids level of the cores from the wheel paths and edges of the wheel paths of
lane 5, coarse m mixture ....................................... .............. ................. 52

4.4. V iscosity of recovered asphalt.......................... .............................. ............... 53

4.5. Asphalt pavement analyzer rut depth for both fine-graded and coarse-graded
m ixtu re. .............................................................................55

4.6. Indirect tensile test parameter versus servopac test parameters...............................67
















LIST OF FIGURES


Figure page

3.1 Show the layout accelerated pavement testing test track........................ ...............23

3.2. Pavement structure for the test lanes with fine-graded and coarse-graded mixtures..24

3.3. Job mix formula gradation curves of fine-graded and coarse-graded mixtures..........27

3.4. Heavy vehicle simulator test carriage and lasers........................................... ........... 30

3.5. Longitudinal rut imprints on section 3-3B....................................... ............... 31

3.6. Thermocouple assembly on track 3-2A. .......................................... ............... 32

3.7. Heavy vehicle simulator with its insulation assembly.............................................33

3.8. Location of cores to be taken after the HVS runs on section 3-3B...........................34

4.1. Plot of change in rut depth versus number of HVS wheel passes. ...........................40

4.2 Maximum differential rut depths for the sections............................. ...............41

4.3 show the maximum absolute rut depths for all the test sections.............................. 41

4.4 Differential transverse profiles at 100 passes of the fine-graded mix. ........................43

4.5. Differential transverse profiles at 100 passes as compared with one of 90000
passes for the fine-graded mixture. ........................................ ....... ............... 43

4.6. Differential rut profiles at 100 passes of the coarse-graded mixture..........................44

4.7. Differential transverse profile at 100 passes as compare with one the 90000 passes
for the coarse-graded m ixtures. ........................................ ........................... 44

4.8. Evolution of transverse profile of fine-graded mixture section 3C. .........................45

4.9. Evolution of transverse profile of coarse-graded mixture section 5A......................46

4.10. Initial and final surface profile of a fine-graded section 3-3C...............................47

4.11. Initial and final surface profile of a coarse-graded section 3-5A............................47









4.12. Positive A1 and negative A2 areas of a transverse rut profile ................................49

4.13. Asphalt pavement analyzer comparisons and HVS rut depths for the fine-graded
and coarse-graded mixture of the top lift. ................................ ....... ............ 56

4.14. Average gyratory shear stress versus number of gyrations for bottom lift.............57

4.15. Average gyratory shear stress versus number of gyrations for the top lift ..............58

4.16. Gyratory shear stress versus number of gyrations of the servopac compactor for
section 3C fine-graded section. ...... ....................................................................59

4.17. Gyratory shear stress versus number of gyrations for 3-3C with a change of
gyration angle from 1.25 to 2.5 .......................... ................... ..................59

4.18. Gyratory slope and initial failure strain of the top lift fine-graded and coarse-
graded m fixtures .......... ...... ........... .................................. ........ 60

4.19. Pass/Fail criteria for evaluation of rut resistance............................................... 61

4.20. Differential rut depths versus gyratory slope of the coarse-graded mixture of the
top lift. .............................................................................. 6 1

4.21. Differential rut depths versus gyratory slope of the fine-graded mixtures of the
top lift ............. ......... .......... ............ ................................... 62

4.22. Gradation of the fine-graded and coarse-graded plant mix mixtures of the top lift. 63

4.23. Interaction unit check for lane 1 and section 2A of lane 2 of the fine-graded
m mixture top lift.............. .. .............. ................... .. ...... .. .... .............. 64

4.24. Interaction unit check for lanes 2 and 3 for the fine-graded mixture top lift............64

4.25 Porosity of the fine-graded mixture for interaction and no interaction. ...................65

4.26 Unit interaction plot for the coarse-graded mixture of the top lift ..........................66

4.27. Relationship between gyratory shear slope and the energy ratio of both the fine-
graded and coarse-graded mixtures ....... ..................... ...............68

4.28. Comparison of fine-graded and coarse-graded differential rut depths ...................69

4.29. Job mix formula HVS Round 1 and HVS Round 3 for the fine-graded mixture......70

4.30. Differential rut depth of the fine-graded mixture HVS Round 1 and HVS Round
3 ......................................................... .............................. . 7 0

4.31. Interacting unit check for "JMF" "HVS Round 1 and 3" and lanes 4 and 5............72









4.32. Porosity of job mix formula for "HVS Round 1" and "HVS Round 3" fine-
graded m ixtu re....................................................................................... .... 72

4.33. Longitudinal crack on section 3B ........................................ ......................... 73

4.34. Strain gage induced crack on the top lift of section 3-3B fine-graded mixture........74

A-i Gradation and sources of aggregate for the coarse mixture ....................................77

A-2. Hot mix design data sheet for coarse-graded mixture ............................................78

A-3. Gradation and sources of aggregate for the fine-graded mixture ............................79

A-4. Hot mix data sheet for the fine-graded mixture ............... ................ ...................80

B-1. Viscosity of recovered asphalt section 3A bottom lift......................................81

B-2. Viscosity of recovered asphalt section 3B top lift .............................................82

B-3. Viscosity of recovered asphalt section 3C top lift ................... .... ............... 83

B-4. Viscosity of recovered asphalt section 5B top lift ................. .............................84

B-5. Viscosity of recovered asphalt section 5A top lift...............................................85

B-6. Viscosity of recovered asphalt section 5C bottom lift...........................................86

C-1. Area-change param eter section 2A ...................................... ......................... 87

C-2. Area-change param eter section 2B ........................................ ........................ 88

C-3. Area-change param eter section 2C .......................................... ......... ............... 89

C-4. Area-change param eter section 3A ...................................... ......................... 90

C-5. Area-change param eter section 3B ........................................ ........................ 91

C-6. Area-change param eter section 3C ........................................ ........................ 92

C-7. Area-change param eter section 4A ...................................... ......................... 93

C-8. Area-change param eter section 4B ........................................ ........................ 94

C-9. Area-change param eter section 4C ........................................ ........................ 95

C-10. Area-change parameter section 5A............................................................... 96

C- 1 A rea-change param eter section 5B ................................ ......................... .. ......... 97

C-12. Area-change param eter section 5C ............................... ................................. 98














Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Engineering

EVALUATION OF RUT RESISTANCE OF SUPERPAVETM FINE-GRADED AND
COARSE-GRADED MIXTURES

By

Collins Boadu Donkor

December 2005

Chair: Mang Tia
Major Department: Civil and Coastal Engineering

Two gradation specifications were developed as part of the Superior Performing

Pavement Program in 1993. They are known as coarse and fine gradations respectively.

It was recommended to use the coarse gradation in order to achieve better rutting

resistance, however many researchers believe that either of the two gradations will do

well in rutting resistance if they are properly designed and constructed.

This research focuses on evaluating the rut resistance of a fine-graded mixture, as

compared to a coarse-graded mixture in use in the State of Florida.

Both field and laboratory evaluation methods were used. An Accelerated

Pavement Facility was constructed at the State Materials Office in Gainesville, Florida

consisting of 6 sections of a typical fine-graded and a typical coarse-graded mixture in

use in Florida. The loads were applied via a heavy vehicle simulator (HVS)

manufactured in South Africa.









Asphalt Pavement Analyzer (APA) rut depths, Servopac Gyratory Compaction

tests, Indirect Tensile Test (IDT), and Bulk Density tests were performed on the

specimens of the fine-graded and coarse-graded mixtures.

Analyses of the results of both the differential and absolute rut depths show that the

difference in the mean rut depths of the coarse-graded and fine-graded mixture was

statistically insignificant. From the limited data collected we can not conclude that the

SuperPaveTM coarse-graded mixture is either better or worse than the SuperPaveT

fine-graded mixture.














CHAPTER 1
INTRODUCTION

1.1 Background

Rutting is a world-wide performance problem associated with Hot Mix Asphalt

(HMA) mixtures. It manifests itself as a longitudinal bowl-like surface depression in the

wheel paths on flexible pavements with the application of vehicular loads. The gradual

and progressive reduction of wheel path layer thickness leads to functional as well as

structural deficiency of the pavement. Rutting can be due to instability caused by

inadequate shear resistance and or densification in the HMA mixtures.

Where pavement structure is inadequate rutting could occur as a result of

permanent deformation of the subgrade or granular base and subbase materials.

Rutting creates safety, functional as well as structural problems on HMA

pavements. A 10 mm rutting is likely to cause hydroplaning on pavements at 40 mph

traffic speed. Development of depression and shoves increase the surface roughness

resulting in higher vehicle operating cost.

Higher levels of rutting cause reduction in layer thickness which reduces the load

spreading ability of the pavement. Large tensile stresses could develop at the bottom or

on top of the surface layer causing cracking in HMA. Low shear resistance associated

with rutting of HMA mixtures have been found to create shear planes (Birgisson et al.

2002). Such planes of weakness have induced crack development in HMA layers. Top-

down longitudinal cracking have also developed as result of high tensile stress at crowns

of the shoves created by permanent deformation of HMA layers.









Rutting being a performance measuring parameter is associated with traffic loading

as well as mixture quality. Two differently designed mixtures of the same pavement

thickness, and loaded the same way under the same environmental conditions may

behave differently in rut performance. Mixture differences could be exhibited by

gradation, aggregate texture, binder grade and content, dust to effective binder ratio as

well as mixture volumetrics, voids in mineral aggregates (VMA), voids filled with

asphalt (VFA), and air voids (AV).

The State of Florida has experienced some amount of rutting of varying extent and

degrees of severity on at least two pavements constructed with SuperPaveTM coarse

mixtures since its implementation in 1995. The Coarse SuperPaveTM mixtures are the

dominant HMA used in Florida. It accounts for over 75% of all HMA construction in the

state interstate, and heavy traffic highways.

Properly designed SuperPaveTM coarse mixtures have been mandated by the

Florida Department of Transport (FDOT) for heavy-traffic, HMA construction because of

perceived better rut resistance than the fine-graded mixtures, complaints from both

Contractors and FDOT Engineers on the difficulties associated with its use, have called

for the inclusion of the fine-graded SuperPaveTM mixtures for heavy traffic and Interstate

Highways. Contractors struggle to meet minimum voids in mineral aggregate (VMA)

specifications, especially when using aggregates native of Florida (Sholar et al. 2004).

The requirements of VMA in coarse-graded SuperPaveTM mixtures have even been called

into question in several research projects (Nukunya et al. 2001, Anderson et al. 2001, and

Kandhal et al. 1999) which indicated that more serious work remains to be done.









Coarse-graded mixtures are difficult to compact, leading to large field-air voids and

hence high densification and higher rut depth, as exemplified in the Wes Track

experience. It segregates easily causing traveling and moisture-induced damage and

ultimately reduction of pavement performance. Permeability of HMA mix layers are

influenced by the level of compaction. Lower level of compaction increases the

permeability allowing air and moisture ingress that affects the durability of the mix.

Fine-graded SuperPaveTM mixtures do not suffer such difficulties associated, with

graded-coarse SuperPaveTM mixtures however, they are less attractive on the basis of

their higher rut susceptibility.

This research evaluates the long-term rut resistance of SuperPaveTM fine-graded

mixtures used in Florida and makes comparison to the predominant coarse-graded

mixtures in use in the State. The Heavy Vehicle Simulator (HVS) acquired by the

FDOT Materials Office and the constructed Accelerated Pavement Testing (APT)

Facility was used in this study. The HVS applied on a full-scale accelerated testing

facility can be used to evaluate rut resistance of mixtures in situations where knowledge

of long-term performance is not available. The HVS can simulate 20 years of interstate

traffic on a test pavement within a short period of time (Tia et al. 2002).

1.2 Research Hypothesis

The resistance to shear deformation, which is a measure of rut resistance in HMA

mixture, has been attributed mainly to inter granular contact friction and interlocking of

the coarse aggregate particles. Under near-surface, low-confinement loading conditions,

coarse-graded SuperPaveTM mixtures will out perform fine-graded SuperPaveTM mixtures

in rut resistance.









1.3 Objectives

There are three main objectives in this research. Primarily, it is focused on

evaluating the performance of a typical fine-graded and a typical coarse-graded

SuperPaveTM mixture with respect to rutting under the Florida climate. The second

objective is to quantify rut in terms of average rut depth using laser-captured progressive

transverse profiler. This research will also evaluate performance predicting models

developed at the University of Florida. The models include the "pass/fail criteria" to

predict pavement performance based on the SuperPaveTM gyratory compaction, indirect

tensile test (IDT) and asphalt pavement analyzer (APA) characteristics of mixtures.

Hot mix asphalt (HMA) mixtures densify upon the application of wheel loads from

as built air voids contents of 7% to end of design life air void content of 4%. With

further application of loads after the 4% air voids content, a good mix can mobilize

enough shearing resistance to counteract the shearing and tensile stress that causes rutting

at the near-surface and low-confinement of the HMA layer. High-shear and tensile

stresses at the near-surface create shear planes of changing angles

(Darku and Birgisson. 2003) in rut susceptible mixtures resulting in instability rutting of

the HMA layer.

The HVS applied unidirectional radial tire load of 9 kips with four inch wander of

90000 passes to both fine-graded and coarse-graded test sections. A continuous

progressive rut depth measurement will be captured using a laser profiler attached to the

wheel. It is expected that, the result of this study will show clearly which of the two

typical (fine and coarse) Florida mixtures have better rut resistance.









1.4 Scope

The scope of this research is to (1), evaluate the rutting performance of a typical

Florida SuperPaveTM fine-graded and coarse-graded mixture using the HVS.

(2), Conduct a thorough literature review of factors affecting rutting performance of fine-

graded and coarse-graded mixtures and the experiences of some States with Florida-like

climatic conditions, like Texas and Alabama which have switched from coarse-graded to

fine-graded SuperPaveTM mixture and (3), analyze the two mixtures, focusing on their

rutting-resistance characteristics using laboratory test results and performance predicting

models developed at the University of Florida.

The analyses of the test results will be focused on evaluating these performance

models or index tests to accurately and reliably measure a mixture response

characteristics or parameter that is highly correlated to the occurrence of pavement

rutting over a diverse range of traffic and climatic conditions. This will help to predict

long-term performance of different HMA mixtures without having to conduct full-size

long-term APT experiments, which are expensive to perform.

1.5 Research Approach

In order to evaluate the fine-graded SuperPaveTM mixture for interstate traffic and

also make comparison to the coarse-graded mixture for rut performance, an APT facility

was designed and implemented at the FDOT office in Gainesville. The following

activities were executed:

A Heavy Vehicle Simulator (Mark IV) was used to apply a 9 kip single-wheel

115 -psi load, via a radial tire traveling at 8 mph on carefully constructed pavement lanes.

Previous research results (Tia et al. 2002) showed that the effective way of inducing









rutting in APT using HVS was to apply a 9 kip wheel load in a unidirectional mode with

4 inches wander of 1 inch increments via a super single radial tire.

Longitudinal and transverse rut depths were measured using two laser profilers

mounted on the axis on each side of the single wheel of the HVS.

* Analysis of the rut profiles was performed to evaluate rut resistance performance

of the mixtures to determine which type of rutting was predominant densificationn or

shear instability rutting). The area-change parameter (ACP), is a physical methodology

that can be used to determine whether rutting is primarily due to shear instability or

because of densification.

* Cores were extracted from the wheel paths and along the immediate edge of the

wheel paths for density and thickness measurement. The viscosity of the asphalt

recovered from cores taken from the inside and edge of the wheel paths were determined

to evaluate the effects of gradation, environment and traffic.

The mixtures were tested and the data analyzed using these methodologies:

(i) The University of Florida's rutting framework which is a "pass/fail criteria"
relating the slope and the vertical-failure strain of the gyration curve, using the
SuperPaveTM gyratory compactor.

(ii) The University of Florida Energy Ratio "pass/fail" framework for cracking
performance. Mixtures that exhibit greater resistance to rutting may behave
poorly in cracking resistance. All the mixtures were tested for tensile
strength, and their fracture energy and failure strain parameters were
evaluated.

(iii) The APA test was used as rut distinguishing tool as a "pass/fail" framework
for both mixtures.














CHAPTER 2
LITERATURE REVIEW

2.0 Introduction

The Superior Performing Pavements (SuperPaveTM) mixture design system was

introduced as part of the Strategic Highway Research Program (SHRP) to replace the

Marshall and other mixture design procedures 1993 as a rational mix design procedure.

The main objectives of the SuperPaveTM was to provide mixtures with better resistance to

rutting, fatigue, low temperature cracking and moisture induce damage. Two types

namely "Coarse and Fine" mixtures with different gradation characteristics are in use

today. Originally a coarse-graded mixture defined as having gradation passing below the

Restricted Zone (RZ) whilst the fine-graded mixture gradation passes above the RZ.

Many researchers have evaluated the rut resistance of mixtures. It has been held

for a considerable period of time that the coarse-graded mixtures will out perform the

fine-graded mixtures in rut resistance. For instance, stone matrix asphalt is known to

have excellent rutting resistance. However many researcher have found out that there is

no difference in rut resistance between coarse-graded and fine-graded SuperPaveTM

mixtures. (Kandhal and Cooley. 2002) did not find any significant differences in rutting

resistance between coarse and fine Superpave mixtures. (Sargand and Kim. 2003) using

APA rut-depth analyses concluded that neither gradations passing above, through or

below the restricted zone was significant in affecting rut-depth.

There are many factors that account for the behavior of mixtures during rutting, not

just a matter of the mixture being either a coarse-graded or a fine-graded only.









Rutting of HMA mixtures is affected by many factors such as: (1) Characteristics

of mixture constituents. (2) Traffic loading (3) Environmental effects and

(4) Construction.

2.1 Characteristics of Mixture Constituents

Asphalt concrete is composed of about 95% aggregate and 5% asphalt compacted

at elevated temperatures to low air voids. Because the aggregates are subjected to

crushing and abrasive wear during manufacture, placing and compaction, they are

generally required to be hard, tough, strong and durable with cubical shape, of low

porosity, rough textured and proper gradation.

To bind the aggregates together to form mixture to perform under traffic loading,

the asphalt not only should be stiff enough to resist permanent deformation, should be

also flexible enough to resist fracture in cold weather.

2.1.1 Asphalt

Both the amount and grade of asphalt in the mixtures influence rutting potential of

a mixture. Stiffer binder and hence higher G* values increase the resistance to rutting

the asphalt mixture. (Corte 2001) noted that G* varies from a ratio of 1 to 2 for the same

test conditions and that the sensitivity to permanent deformation as indicated by G*/sin6,

is definitely different from one asphalt to another. There is a limit to increasing asphalt

hardness to control rutting. Stiffer binders have increase brittleness at low temperatures

and thus lower its healing capacity.

Mixtures with binder content on either side of the optimum, impacts negatively on

the permanent deformation characteristics of a mixture. Lower than optimum binder

content results in increase air voids, lower cohesion due to lower film thickness and

lower shearing resistance and higher permanent deformation with load application.









Excessive binder content produces in a mix creates pore pressures tending to break the

interlocking of the aggregate particle and results in instability and high permanent

deformation.

2.1.2 Aggregates

There are a whole lot of aggregate properties that affect the rut resistance of

mixtures. It is the aggregate properties that must provide the support to resist permanent

deformation (NCAT, 1996). The aggregate texture, angularity, nominal maximum size

and gradation are critical to good rutting resistance of a mixture. Adhesion and cohesion

of mixtures are influenced directly by aggregate texture. Angular and rough textured

aggregates will provide stronger frictional, bonding and interlocking forces to resist

rutting than smooth, and rounded aggregate even if rounded aggregates compacts better.

Gradation is the most important parameter for rut susceptibility. Gradation affects

the stiffness and frictional resistance of HMA mixture. The proportion and effective size

of aggregates passing the 0.075 mm sieve controls the mass viscosity of the asphalt that

surrounds the coarse aggregates. As the mass viscosity increases with increasing filler

content, the mix become stiffer, increasing its ability to resist permanent deformation.

Increase in the nominal maximum aggregate size and the percentage of coarse aggregates

increases the volume concentration of aggregates with more rutting resistance ability.

(Stuart and Mogawer 1997) concluded that, under accelerated loading conditions,

increase in nominal maximum aggregate size significantly decreased rutting

susceptibility. Various DOT's including the FDOT have limits on the percentage

passing #200 sieve and the nominal maximum size of aggregates for SuperPaveTM

mixtures.









It has been a long held view among researchers and practitioners that generally

coarse-graded SuperPaveTM mixtures have better rutting resistance than finer-graded

SuperPaveTM mixtures.

After the Wes Track experiment, this view point has been discarded. Darku (2002)

observed that, it is the mixture's ability to dilate during shearing that controls its rutting

potential. Increasing ability of a mixture to dilate reduces its vertical strain thus lower

rutting under loads. In fact neither coarse-graded nor fine-graded Superpave mixtures

have superior rutting resistance. Recent studies (Kandhal et al. 2002) and (NCHRP

Project 9-14) have shown that the mechanical properties of HMA appear to be more

sensitive in coarse-graded than in fine-graded mixtures. Coarse-graded, fine-graded or

mixes passing through the restricted zone may perform well under various traffic and

environmental conditions. Roque (1997) pointed out that good shearing resistance can

be achieved with a broad range of aggregate structures as long as suitable gradations are

used. However as noted again by Roque (2002), there is no clear-cut method of

selecting an aggregate gradation to produce good mixtures.

2.2 Traffic Loading

Traffic load induced stress is the major cause of pavement distresses apart from

climatic conditions. These stresses are a function of wheel loads, tire pressure and type

as well as thickness and stiffness of the layers. Collop and Cebon (1995) concluded that

both dynamic vehicle loads and asphalt layer stiffness variation can have significant

influence on long-term flexible pavement performance. Ullidtz and Larsen (1983)

published a model which predicted the performance of flexible in terms of roughness,

rutting and cracking as a function of traffic loading and climate. With the introduction

of radials tires, many States in the United States and countries around the world over









have seen significant instability rutting in HMA. The effect of tire type (radial or bias

ply) could have significantly influence on how HMA ruts. Bigirsson and Roque (2002)

have shown that the high transverse near-surface stress at low confinement in the vicinity

of the edges of radial tires may partly explain the mechanism of instability rutting. The

history of tire pressures on HMA rutting is well known. Kandhal et al (1990) traced the

changes of truck tires pressures from 70 psi to 85 psi during the AASHTO Road Test and

subsequent increases in States like Virginia, Florida, Texas and Illinois with averages of

96 to 1 l0psi and maximum in the order of 155 psi. There are no legal limits against tire

pressures in the United States (Kandhal et al. 1990).

2.3 Environmental Effects

The effects of temperature on rutting have been investigated by many researchers.

Robertson (1997) observed that the shape of the temperature profile in a pavement is

dependent on the air temperature history. Summer time pavement temperatures ranged

for most countries of the world around 60 to 700C. Asphalt concrete being a thermo-

susceptible material will at warm enough temperatures develop significant visco-plastic

strains under wheel loads. The accumulation of large visco-plastic strains results in

permanent deformation of the pavement. Collop and Cebon (1983) observed that

changes in the temperature of the asphalt layers affect the elastic and viscous properties

of the asphalt. Matthews and Monismith (1992) using Hveem Stabilometer and creep

test results concluded that temperature has more influence on rutting than the aggregate

grading. As asphalt oxidizes under temperature and air, it looses its flexibility thus

increases its stiffness with increasing rutting resistance but lose it cracking resistance.









2.4 Construction

Constructional inaccuracies and variability impact negatively on pavement

performances. Villiers (2004) showed that variation in dust content and/or asphalt

content that were perfectly acceptable under the current practice produced significant loss

in performance or unacceptable cracking performance. Hot mix asphalt construction

starts with mixture design and is followed sequentially with production, laying and

compaction. Poor mixture designs have direct bearing on mixture performance. For

instance too well graded aggregate gradation leads to low VMA and lower than

acceptable asphalt content. Rogue (1997) stated that mixtures that are poorly designed,

produced, and constructed can result in rutting due to plastic deformation. Mixtures with

too much rounded natural sand particles have resulted in tender mixes and poor rutting

performance (Buchanan and Cooley 2002). Inaccurate production or lack of proper

production control of mixtures leads to pavements of variable performance. Nouredin

(1997) concluded that, the overall pavement performance life may be significantly

affected when the specific asphalt content, aggregate gradation (job mix formula) and the

degree of compaction are not achieved in situ. The success of any HMA pavement

construction lies with careful and accurate laying operation. Mat thickness, segregation

and moisture induced damage control results from proper and efficient laying operations.

Stroup-Gardiner et al (2000) reported that when there are higher levels of segregation, the

failure mode shifts from fatigue to compression i.e. rutting. Adu-Osei et al (1999)

concluded that construction variability can have significant adverse effect on the

performance of mixtures. Compaction is to reduce the air void level of the mixture to

increase the aggregate to aggregate contact and interlock thus, increase mixture resistance

to shearing forces. Excessive air voids due to inadequate compaction increase rut depths









on application of traffic loads. It should be noted that, for adequate rutting resistance,

stone-to-stone contact is critical. HMA layers consolidate more with higher initial air

voids content. Harvey and Popescu (2000) showed that good construction compaction

helps to reduce the amount of densification that occurs under trafficking and reduces the

amount of rutting caused by densification. Peterson et al (2004) observed that achieving

proper compaction of asphalt pavement is crucial to its longevity and acceptable

performance.

2.5 Rut Measurement

Accurate rut measurement is important safety consideration on our roads ways.

Rut depth of over 10 mm provides considerable risk to motorist in either rainy or icy

weather. Hydroplaning caused by pond or slickly ice sheet in longitudinal rut depression

jeopardizes the safety of motoring public. Rutting has been known to contribute

insignificantly to longitudinal roughness in pavement. In fact the contribution of rutting

to serviceability is in order of -1.3(RD)2 where RD is rut depth in inches and very small

compared to surface roughness in IRI. However rut depth of 0.5 inches could have

considerable safety as well as structural implications. It is important therefore to

measure and quantify rutting accurately in order to select an appropriate repair strategy.

2.5.1 Non-contact Laser Height-Sensor Rut Depth Measurement

Advancement in technology has provided engineers with a fast, efficient and

automated rut depth measurement using laser non-contact height-sensors called laser

profilers. Traditionally rut depth measurements using strings and or straight edges and

wedges have been dangerous and also time consuming. It is expensive when used for

either project or network level evaluations and usually impractical to measure at regular

intervals. Gokail (2003) observed the impracticality of measuring ruts conventionally









when using the HVS under APT applications. The space restrictions underneath the

HVS loading system will not allow accurate rut data collection with conventional

methods. The non-contact laser measuring systems have rendered these manual methods

obsolete in many countries. The non-contact method consists of measuring the

transverse profile by digitizing the pavement surface and then analyzing the data to

calculate the rut depth using a simulated straight-edge at normal traffic speeds. The

distance between the simulated straight-edge and the lowest point along the transverse

profile is calculated as the rut depth. Tia (2002) showed that two very different rut

measurements can be calculated using the transverse surface profile out from the laser

profiler. The "differential rut depth" and "absolute rut depth" can be calculated from

transverse surface profile.

2.5.2 Differential Rut Depth

When a straight line is drawn to touch the two highest point of the differential

surface profile obtained by subtracting the initial surface profile from the surface profile

after some trafficking, the greatest distance between the straight line and differential

surface profile is the "differential rut depth". The function of this parameter is to

incorporate the instability characteristics of the material into the rutting prediction Drakos

(2002) and includes the dilation portion of the deformed material into the rut

measurement.

2.5.3 Absolute Rut Depth

Absolute rut depth is the difference between the lowest points on the initial surface

profile and the surface profile after some period of trafficking. Both the initial and as

trafficked rut depths are measured the same way. A straight line is drawn to touch the

highest points on the surface profile and the greatest distance to the lowest point on the









surface profile measured as the rut depth. Villiers (2004) concluded that absolute rutting

cannot be used as a measure of mixture performance. One must evaluate each section

carefully to assess the contribution of different layers.

2.6 Mixture Response Characteristics or Parameters

Several mixture response characteristics or parameters exist for both laboratory as

well as field evaluation of HMA. Permanent deformation and fatigue response have

been measured in the laboratory using the Servopac SuperPaveTM Gyratory Compaction

(SGC), Asphalt Pavement Analyser (APA), SuperPaveTM Indirect Tension Test (IDT)

characteristics and the Gyratory Testing Machine (GTM). The slope and the vertical

failure strain parameters from the Servopac SGC as proposed by Birgisson et al (2002)

and the APA rut depth are tools for rut depth evaluation. Low m-value and high

Dissipated Creep Strain Energy (DCSE) have been known to associate higher fracture

resistance as calculated from the stress-strain curve of HMA under stated temperature and

time of loading (Roque and Butler 1997). Nukunya (2002) evaluated mixture rutting

resistance using IDT creep results. The ratio of initial to final gyration angle as

measured with the GTM has been used by many researchers to evaluate mixture

instability.

2.6.1 Indirect Tension Test (IDT)

Top down cracking is the primary mode of pavement distress in Florida.

Approximately 80% of the State's deficient highways are due to top down cracking

Sholar (2004). Mixtures that exhibit high resistance to rutting may exhibit high

propensity to fracture because of its excessive stiffness and low healing potential at low

and intermediate temperatures. Such mixtures may exhibit high creeping compliance.

Kim (2003) showed that there is a direct relationship between the rate of creep and the









rate of micro-damage accumulation while investigating the effect of Styrene butadiene

styrene (SBS) on coarse SuperPaveTM fracture potential. The SuperPaveTM IDT tests

can be used to measure the diametric Resilient Modulus (MR), Creep Compliance (CC)

and Tensile Strength (TS) of HMA. (Roque et al. 1999) developed a framework for

evaluating the fracture properties of HMA using Energy Ratios (ER) from calculated

Dissipated Creep Stain Energy (DCSE), Fracture Energy Limit (FE), and MR.

2.6.2 Servopac SuperPaveTM Gyratory Compaction

The Servopac Super pave Gyratory Compactor (SGC) has excellent versatility in

what parameters it could measure and also the ease which control parameters could be

changed to simulate field conditions. Measurement during compaction such height,

density, air voids and shear strength can be obtained with this compactor. The ultimate

field and initial densities of a mix can be simulated using the SGC.

(Birgisson and Darku 2002) proposed a "Pass Failure" framework for evaluating

mixtures rutting potential based on the Slope of the gyratory shear against number

gyrations and the vertical failure strain. The minimum acceptable gyratory shear slope is

15 kPa and the optimum range for good rutting resistance is between 1.4% and 2.0% for

vertical failure strain. Mixture with failure strain less than 1.4% are considered brittle

whilst that greater than 2.0% are considered plastic.

Instability rutting is manifested in a rearrangement of the aggregate structure

(Birgisson and Darku 2002). This rearrangement is produced during compaction by

changing the gyration angle of the Servopac SGC from 1.25 to 2.5 at 7% air voids

content of the mixture. The straining of the mixture during this change of gyration angle

and the degree of rearrangement afterwards gives an indication of the mixtures rutting

resistance in the field.









The Vertical failure Strain is the ratio of the difference in specimen height at

maximum shear stress to the height of specimen at the first minimum shear stress after

the change of angle of gyration from 1.250 to 2.50.

2.6.3 Gyratory Testing Machine GTMT

The Gyratory Testing Machine (GTMTM) was developed by the Army Corp of

Engineers in the late 1950 at the Waterways Experimental Station. It is a compactor as

well as a tester. The GMTTM combines vertical pressure and shear displacement to

simulate field roller compaction and future traffic densification thus, mixture properties

measured during compaction and densification with the GMTTM could be used to

evaluate mixture performance. The measure of mixture's stability is the Gyratory

Stability Index (GSI) and is related the mixture's resistance to rutting. The maximum

gyration angle divided by the initial gyration angle is the GSI. GSI values of more than

1.1 indicate instability whilst values close to 1 shows the mixture is stable. The GTM is

the only device capable of monitoring changes in mixture response with densification.

Moseley (1999) concluded that the GTMTM give a good indication of mixture

performance by measuring shear resistance.

2.6.4 Asphalt Pavement Analyzer

The Asphalt Pavement Analyzer (APA) was a development from the Georgia Load

Wheel Tester to evaluate rutting and moisture susceptibility of HMA mixtures in the mid

1990s. Asphalt mixtures compacted mainly in the SGC are placed in mold and subjected

to 8000 cycles of loaded 100 psi pressure hose in an environmental chamber maintained

at 640C.









In Florida, mixtures that exhibits rut depth of over 8 mm after the 8000 cycles are

considered unacceptable for field rutting performance. Rut depth is measured as the

difference between lowest point before and after the 8000 cycle application.

Various researchers have used the APA to characterize HMA mixtures. Research

at NCAT showed the APA was sensitive to mixtures with different asphalt binder and

varying gradation. Thus coarse-graded and fine-graded mixtures would show different

results. This observation is not shared by the FWHA at Turner-Fairbanks which

compared APT and Load Wheel Test (LWT) results for rutting. They concluded that

none of the LWT could distinguish between good and poor performing mixtures which,

was clearly distinguished by the APT. Sargand and Kim (2003) using APA rut-depth

analyses concluded that neither gradations passing above, through or below the restricted

zone was significant in affecting rut-depth.

Drakos (2002) showed the APA hose does not capture the critical lateral stresses

found to be detrimental to rutting as well as cracking of HMA pavement and modified the

loading mechanism from a hose to a strip know as the New APA to simulating stress

distribution in a single radial tire rib. Drakos (2002) again showed that neither the New

APA nor the original APA could distinguish between coarse-graded and fine-graded

SuperPaveTM mixture performance at 4% air voids and at 640C. They reported

significant difference in performance at 700C using the New APA at the same air void.

Rut depths were measured using Absolute Rut Depth Method.









2.7 Accelerated Pavement Testing

The history of Accelerated Pavement Testing (APT) dates back to late 1900s when

Michigan implemented the first APT facility with the view to determine pavement

response and performance under a controlled, accelerated accumulation of damage.

Accelerated pavement testing offers enormous potential for studying pavement

distress mechanisms and for evaluating performance of asphalt mixtures and pavement in

relatively short period of time, (Roque and Tia 2005).

(Lea and Heath 1997) have observed that, APT has been used primarily for long-

term plan to systematically test and analyze existing materials, using a defined testing

matrix over a large number of test sections. The Federal Highway Administration

conducted test on mixtures of different gradations and asphalt contents at the Accelerated

Loading Facility at Turner Fairbanks Highway Center. (Romero and Stuart 1998) using

the data from the ALF at Turner Fairbanks Highway Center showed significant field

performance of mixtures with different gradation but of the same asphalt content.

The Texas Accelerated Pavement Test Center (TxAPT) has a new fixed facility that

can yield information for accelerated pavement test. The Mobile Load Simulator (MLS)

and new instrumentation has been used to demonstrate the effect of heavy loads on thin

load-zone pavements and comparison of this pavement with good and superior quality

crushed stone bases. The MLS use two wheels instead of a single wheel which is used in

Florida. (Hudson et al. 2004) concluded that fixed APT centers provide more output

from the equipment and thus better payout on the investment over the short term than

using in-place highways.

One role of APT, which was often neglected but have gained popular recently, is in

evaluating innovative materials. In South Africa, the Council of Scientific and Industrial









Research (CSIR) used the HVS and the APT facility to evaluate what they called

"inverted pavement". It was found that the presence of a crushed stone base course over

a stabilized subbase course in a pavement has a bridging effect at transferring cracks in

the stabilized subbase and inhibits the rate of crack propagation to the surface".

Evaluation of new materials, in order to cut down on the cost of experimental

evaluation and bring the products to the markets quicker has been done well using APT

facility as shown in the evaluation of SBS modifiers in Florida.

Styrene Butadine Styrene (SBS) modifier's effect on rutting resistance of

Superpave fine mixture was evaluated using the APT facility at the Florida's Department

Transport Material Office in Gainesville.














CHAPTER 3
MATERIALS AND METHODOLOGY

3.0 Introduction

To be able to accomplish the objectives of this research, experiments were designed

that enables us to apply accelerated wheel loads using the South African Heavy Vehicle

Simulator (HVS) Mark IV on test tracks constructed with the fine-graded mixture and

coarse-graded SuperPaveTM mixture. This research has combined accelerated pavement

testing with laboratory testing using various equipment and data analysis tool to

characterize and to predict the behavior of the mixtures using theoretical models,

materials response characteristics as well as pavement performance models.

3.1 Summary of Methodology

The Test Tracks and the HVS were equipped with appropriate instrumentation to

measure pavement temperatures, rut depths, strains and pressures. Rut data were

analyzed procedures to predict mixture performance. The laboratory tests were

conducted to measure the volumetric and performance parameters of the mixtures. These

tests measured the Air Voids, VMA, VFA, APA, Gyratory Slope and Vertical Failure

Strain and IDT Creep, Resilient Modulus and Indirect Tensile Strength. Mixture

Stability calculated as the Gyratory Stability Index, which is related to rutting potential of

the mixtures were measure. Analysis of the gradations of the Job Mix Formula and Plant

mixtures were performed using the Florida HMA Gradation Guidelines (FHMAGG)

recently developed at the University of Florida. The mixtures Dominant Aggregate Size

Range (DASR), Porosity, Interaction Characteristics and the Interstitial Volume









properties were evaluated to characterize the mixtures. Visual condition survey was

undertaken to observe the severity and extent of pavement distresses that have occurred

so as to measure pavement performance of the mixtures under real load conditions.

3.2 Test Track Layout

The construction of the test track was started in January and was completed in the

same month in 2005. The test track is located within the FDOT State Materials Office in

Gainesville, Florida. The construction of the asphalt concrete surface was done on an

existing limesrock base and granular subbase. Each lane has two lifts of asphalt concrete

from an asphalt plant located in the City of Gainesville. The accelerated pavement test

tracks consisted of three (3) lanes paved with fine-graded and two (2) lanes paved with

coarse-graded SuperPaveTM mixtures. Each lane is 132 feet long and is divided into

three (3) sections of A, B and C each 44 feet. All the lanes are of equal width of 12 feet

and were constructed by conventional plants mixing, placement and compaction

processes that exemplify the real world situation of construction and material variability

including mixture's asphalt content and gradation variations. Each of the sections of 34

feet long had 20 feet of test area and 7 feet at each end for acceleration and deceleration

of the wheel. The south end of the tracks was reserved for maneuvering the HVS. It was

important that no preloading of the tracks were introduced by the HVS during the

maneuvering because the 40 ton weight of machine could cause damage to the pavement.

The testing sequence was arranged such the effect of time on each lane could

average out.

Figure 3.1 shows the layout of the test tracks and the instrumentation plan and

locations.











AGING STUDY FINE GRADED COARSE GRADED

LANE 1 LANE2 LANE3 LANE LANE

17fl Il" I" l If \

2 2 I f I2 2 fl
S24ft 1

2fl 2 0
ftl I ft l


I ? I ~ -

1l I I i I D0
I, I 'i | 'o Ig ^


II I-- I
L i 2 21 2fl I

4 + staggerin Isa steggenn



S' I I I
Si P;

i t-I-


B B 1 B slaggeBnn I 2fl 1 s aggr I


-- 2f2 o 2
o17f ft l I CL
3F 6F -I 1 2 F
I I staggennc I I1 agsggemn ,

flI 2f 2f |

IIIatE 8 off IPatE i 12
__ _3-F .. __ -F ___ i2fTCL iI, 1 ,CL i o1ffCLT
+- 13-FT--- +-6-FT-- -2-FT-12- FT-------12-FT--------2-FT-F


EXPERIMENT-3 INSTRUMENTATION PLAN


KEY

Strain Gauge, Top
Layer
I Strain Gauge,
Bottom Layer
S Pressure Cell,
Bottom Layer

Total Strain Gauges = 7C
Total Pressure Cells = 9


Not to Scale


Figure 3.1 Show the layout accelerated pavement testing test track.


3.3 Pavement Structures


The pavement structure consists of a prepared sandy subgrade, a layer of over 10.5


in (265 mm) of limerock base course and 12 in (305 mm) of granular subbase. Lanes 1


through to lane 3 were constructed with two lifts of 2 in (50 mm) fine-graded


SuperPaveTM mixture surfacing whilst Lanes 3 and 4 were finished with two lifts of 2 in


(50 mm) coarse-graded SuperPaveTM mixture. Figure 3.2 shows the pavement structure


for the APT test track.

























Fine-graded mixture tracks Coarse-graded mixture track

Figure 3.2. Pavement structure for the test lanes with fine-graded and coarse-graded
mixtures

3.4 Asphalt Concrete Mixtures used

Two mixtures were used for this study. Both are 12.5 mm nominal size mixtures

and vary only on definition of being a SuperPaveTM fine-graded or a coarse-graded

mixture. They were made using Georgia granite aggregates, natural sand from a local

sand pit at Gainesville and PG 67-22 asphalt. The mixes were designed by the FDOT

using traffic correspond to class D (10-30 millions) ESAL's over the design period of 20

years. The mix was produced by a batch plant located in Gainesville. Even though it

was planned not to have significant gradation variations within each mix type, samples

from delivery trucks used for the construction of the test tracks show considerable

variations. This may result in significant aggregate structural differences and

performance variations. Tables 3.1, 3.2 3.3 and 3.4 show the Job Mix Formula gradations

and grading of samples from the trucks used for the construction of the bottom and top

lifts lanes for both coarse-graded and fine-graded mixtures. Mixture volumetric

properties and asphalt contents are also shown. The optimum asphalts chosen for the

fine-graded and coarse-grade mixtures are 4.6 and 4.5% respectively.


50 mm AC Toplift


50 mm AC Bottomlift


265 mm Limerock base



305 mm Granular subbase


50 mm AC Toplift


50 mm AC Bottomlift


265 mm Limerock base



305 mm Granular subbase













Table 3.1. Job mix formula and truck samples gradations, volumetric properties and air
voids for fine-graded mixture bottom lift.


HVS Round 3 (Fine Gradation) Summary Data
Bottom Lift
Fine Graded Mix
Lane 1 1 1 2 2 2 3 3 3
Section A B C A B C A B C
Truck JMF 1 1 2 2 3 3 4 4 5
Gmm 2.579 2.585 2.585 2.578 2.578 2.572 2.572 2.607 2.607 2.609
Grb 2.475 2.487 2.487 2.493 2.493 2.497 2.497 2.504 2.504 2.484
AC content 4.6 4.1 4.1 4.6 4.6 4.3 4.3 4.1 4.1 4.3
Air Voids 4.0 3.8 3.8 3.3 3.3 2.9 2.9 4.0 4.0 4.8
VMA 14.7 13.9 13.9 14.1 14.1 13.7 13.7 13.3 13.3 14.1
VFA 73 73 73 77 77 79 79 70 70 66
Pbe 4.5 4.1 4.1 4.5 4.5 4.3 4.3 3.8 3.8 3.9
Dust Ratio 1.1 1.0 1.0 1.0 1.0 1.1 1.1 1.1 1.1 1.2
19 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
12.5 98.0 97.6 97.6 98.6 98.6 95.9 95.9 97.2 97.2 96.8
9.5 90.0 86.8 86.8 90.0 90.0 86.7 86.7 84.8 84.8 87.7
4.75 68.0 54.7 54.7 63.4 63.4 60.5 60.5 56.7 56.7 60.3
2.36 48.0 38.3 38.3 44.7 44.7 42.8 42.8 39.9 39.9 43.2
1.28 34.0 29.5 29.5 32.9 32.9 32.1 32.1 30.5 30.5 32.8
0.425 25.0 23.6 23.6 26.0 26.0 25.3 25.3 24.6 24.6 26.4
0.3 16.0 15.3 15.3 16.2 16.2 17.0 17.0 15.8 15.8 17.6
0.15 8.0 6.8 6.8 7.6 7.6 7.9 7.9 7.5 7.5 8.0
0.075 4.9 4.1 4.1 4.7 4.7 4.7 4.7 4.4 4.4 4.7
Density 93.0 93.1 92.8 93.3 93.3 94.2 93.7 92.5 92.1 91.9


Table 3.2. Job mix formula and truck samples gradations, volumetric properties and air
voids for the fine-graded mixture top lift.


HVS Round 3 (Fine Gradation) Summary Data
Top Lift
Fine Graded Mix
Lane 1 1 1 2 2 2 3 3 3
Section A B C A B C A B C
Truck JMF 7 7&8 8 5 5&6 6 4 4 5
Gmm 2.579 2.594 2.594 2.590 2.591 2.591 2.598 2.602 2.602 2.591
Grb 2.475 2.493 2.493 2.487 2.506 2.506 2.504 2.491 2.491 2.506
AC content 4.6 4.4 4.4 4.3 4.2 4.2 4.2 4.1 4.1 4.2
Air Voids 4.0 3.9 3.9 4.0 3.3 3.3 3.6 4.3 4.3 3.3
VMA 14.7 14.2 14.2 14.3 13.6 13.6 13.7 14.0 14.0 13.6
VFA 73 73 73 72 76 76 74 69 69 76
Pbe 4.5 4.3 4.3 4.3 4.2 4.2 4.2 4.1 4.1 4.2
Dust Ratio 1.1 1.2 1.2 1.0 1.2 1.2 1.2 1.3 1.3 1.2
19 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
12.5 98.0 97.7 97.7 96.8 97.0 97.0 97.4 97.5 97.5 97.0
9.5 90.0 88.5 88.5 85.4 85.2 85.2 85.9 88.2 88.2 85.2
4.75 68.0 61.6 61.6 58.9 59.1 59.1 57.3 61.6 61.6 59.1
2.36 48.0 44.5 44.5 42.5 43.0 43.0 41.8 44.6 44.6 43.0
1.18 34.0 34.1 34.1 32.7 32.8 32.8 32.2 33.7 33.7 32.8
0.425 25.0 27.3 27.3 26.2 26.4 26.4 26.0 26.9 26.9 26.4
0.3 16.0 17.5 17.5 16.5 17.2 17.2 17.0 17.2 17.2 17.2
0.15 8.0 8.2 8.2 7.6 8.2 8.2 8.0 8.2 8.2 8.2
0.075 4.9 4.9 4.9 4.4 5.1 5.1 4.9 5.2 5.2 5.1
Density 93.0 92.8 92.1 92.6 93.7 93.4 93.9 92.7 91.4 92.3












Table 3.3. Job mix formula and truck samples gradations, volumetric properties and air
voids for the coarse-graded mixture bottom lift.


HVS Round 3 (Coarse) Summary Data
Bottomlift
Coarse Graded Mix
Lane 4 4 4 5 5 5
Section A B C A B C
Truck JMF 6 6 7 7 7 8
Gmm 2.589 2.573 2.573 2.572 2.572 2.572 2.568
Gmb 2.485 2.461 2.461 2.548 2.548 2.548 2.451
AC content 4.5 4.8 4.8 4.6 4.6 4.6 4.6
Air Voids 4.0 4.3 4.3 4.4 4.4 4.4 4.6
VMA 14.6 15.7 15.7 15.6 15.6 15.6 15.8
VFA 73 72 72 72 72 72 71
Pbe 4.4 4.7 4.7 4.6 4.6 4.6 4.6
Dust Ratio 1.0 0.9 0.9 0.8 0.8 0.8 0.8
19 100.0 100.0 100.0 100.0 100.0 100.0 100.0
12.5 98.0 98.2 98.2 97.2 97.2 97.2 96.9
9.5 90.0 89.5 89.5 86.4 86.4 86.4 85.9
4.75 54.0 48.5 48.5 44.4 44.4 44.4 43.0
2.36 32.0 29.8 29.8 28.1 28.1 28.1 27.0
1.28 23.0 22.5 22.5 21.8 21.8 21.8 21.3
0.425 17.0 18.2 18.2 17.9 17.9 17.9 17.6
0.3 11.0 11.6 11.6 11.9 11.9 11.9 11.6
0.15 5.0 6.1 6.1 5.9 5.9 5.9 5.8
0.075 4.5 4.0 4.0 3.8 3.8 3.8 3.7
Density 94.5 93.9 93.6 94.2 93.3 92.8 94.4


Table 3.4. Job mix formula and truck samples gradations, volumetric properties and air
voids for the coarse-graded mixture top lift


HVS Round 3 (Coarse) Summary Data
Toplift
Coarse Graded Mix
Lane 4 4 4 5 5 5
Section A B C A B C
Truck JMF 2 2 3 1 1 2
Gmm 2.589 2.579 2.579 2.583 2.573 2.573 2.579
Gmb 2.485 2.468 2.468 2.514 2.457 2.457 2.468
AC content 4.5 4.6 4.6 4.9 4.5 4.5 4.6
Air Voids 4.0 4.3 4.3 2.7 4.5 4.5 4.3
VMA 14.6 15.2 15.2 14.0 15.6 15.6 15.2
VFA 73 72 72 81 71 71 72
Pbe 4.4 4.6 4.6 4.6 4.5 4.5 4.6
Dust Ratio 1.0 0.8 0.8 0.9 0.8 0.8 0.8
19 100.0 100.0 100.0 100.0 100.0 100.0 100.0
12.5 98.0 96.6 96.6 97.8 96.8 96.8 96.6
9.5 90.0 85.8 85.8 89.2 88.0 88.0 85.8
4.75 54.0 47.8 47.8 53.7 47.3 47.3 47.8
2.36 32.0 30.1 30.1 33.7 29.5 29.5 30.1
1.28 23.0 23.1 23.1 25.8 22.8 22.8 23.1
0.425 17.0 19.0 19.0 21.0 18.7 18.7 19.0
0.15 11.0 12.6 12.6 13.9 12.1 12.1 12.6
0.3 5.0 5.9 5.9 6.5 5.7 5.7 5.9
0.075 4.5 3.8 3.8 4.1 3.6 3.6 3.8
Density 94.5 92.7 92.6 92.1 93.4 93.7 93.4











3.4.1 Aggregates

The coarse mix was made up of 27% of #78 stone with Georgia code GA-553 and

Florida code 43, 33% #89 GA-553 code 51, 32% W-10 Screening GA-553 code 20 and

8% natural from Starvation Hill Pit. The proportions are by weight of total aggregates.

The fine-graded mixture was made from the same aggregate source with the

following mix proportions, 27%, 10%, 53% and 10% of Florida code 43, 51, 20 and

Starvation Hill sand respectively. These gradations are typical for asphalt pavement

construction in Florida. The same aggregate was used for the production of both mixtures

but vary only on the gradation requirements. Figure 3-3 shows the gradation of the

coarse-graded and fine-graded mixtures.


JMF FINE JMF COARSE

100.0

90.0 0

80.0

S70.0

S60.0
0,
S50.0
a1)
S40.0

S30.0

20.0

10.0 03

0.0 0
0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00
Sieve size (rrn) ^0.45


Figure 3.3. Job mix formula gradation curves of fine-graded and coarse-graded mixtures.









3.4.2 Asphalt

The asphalt used for the design and production of the two mixtures was a grade PG

67-22 (AC-30) which is typically used for high traffic volume asphalt pavement

construction in Florida. The AC has considerable stiffness at in service temperature of

Florida's climate and influences rutting performance. In this research no asphalt

modifier or additive was used in the mixtures. Table 3.5 shows the test requirements and

specification for the asphalt used for production of both fine-graded and coarse-graded

mixtures.

Table 3.5. Specification for the asphalt PG 67-22 used for the mixtures.

Original Asphalt
Test Specification TemperatureC
Viscosity Max. 3 Pas 135
G*/Sin 0 Min. 1.0 kPa 64
RTOF
Test Specification TemperatureC
G*/Sin 0 Min. 2.2 kPa 64
PAV
Test Specification TemperatureC
G*Sin 0 Max. 5000 Pas 25
Creep
Stiffness 0.300 < CS < 300 -12 @ 60 sec

The optimum asphalt content for the fine-graded and coarse-graded mixtures were

4.6% and 4.5% respectively. The optimum asphalt content was determined at Ndes of

100 gyrations with the Pine Gyratory Compactor and at 4% air voids content. Appendix

A shows the Mix designs for both the fine-graded and coarse-graded mixtures

3.5 Test Tracks Construction and Instrumentation

All five test lanes, which were made up of three (3) lanes of fine-graded mixture

and two (2) lanes of coarse-graded mixture, were constructed at the APT site at the

FDOT State Materials Office in Gainesville. Lane 1 was for aging the studies, lanes 2









and 3 were for fine-graded mixtures whilst lane 4 and 5 were coarse-graded mixture

rutting studies. Each lane has two lifts of 50 mm asphalt concrete surface. The lanes

were compacted to 7+1% field air voids at the optimum asphalt content. Samples of the

mixtures were taken from the trucks and tested for its asphalt content, gradation and

theoretical maximum specific gravity Gmm. After compaction, cores were taken to

determine air void content and lift thickness of each section. These tests were carried

out by the FDOT State Materials Office. The Tables in Appendix B shows the

thicknesses and air voids test results of each lane/section of the test tracks.

Figure 3.1 shows the instrumentation of the test truck. A total of nine (9) pressure

cells were installed at the bottom layer. Seventy (70) Strain gages were installed at

various locations and at the top of the top lift and top of the bottom lift during

construction. Six (6) K-type thermocouples were installed at the beginning of

application of wheel loading for each section. The thermocouples measured

temperatures at a depth of 50 mm from the surface of the pavement.

3.6 Heavy Vehicle Simulator Test Configuration and Instrumentation

The HVS was manufactured in of South Africa. It is self propelled, 40 foot long

and weighs 40 tons. The HVS is capable of applying different loads and could be

equipped with different tires in either singe or dual type. In this research, the HVS was

equipped with a test carriage having a 16in wide super single radial tire of 18000 lbs of

load at 115 psi pressure. The test carriage is mounted on a test beam that allows both

longitudinal and transverse movement. The load via the radial tire makes a 4 inch of

transverse movement called wander in one inch increment for 30 feet Uni-Directional









longitudinal traverse. Tia et al (2002) found that this is the HVS loading configuration

that simulates actual field situation.

3.7 Laser Profiler

The Laser Profiler, model SLS 5000TM manufactured by LMI Selcom was

mounted on a test carriage, one each side and 30 inches apart. Each pass of the laser's

records 58 data points. This process is repeated until each laser covers the lateral

distance between them, making a total of 61 sweeps. There is a coincidence of laser

points at the 61st point for the two lasers. The last sweep of the right laser overlaps the

first sweep of the left laser. The total lateral distance the lasers make sums up to be 60

inches. Data captured by computers connected the HVS was used to calculate the

longitudinal profile and the transverse profiles using a computer software written by Tom

Byron of the FDOT. The figure below shows the HVS Test Carriage and Lasers.

B^^-B^-,


Figure 3.4. Heavy vehicle simulator test carriage and lasers.









3.8 Trafficking

A total of 90,000 passes of a 9000 lb, via a 12 inches wide radial tire at 115 psi was

applied on each section of the Test Track at speed 8 mph in one direction only

(unidirectional mode). Fourteen thousand (14000) passes were applied each day running

continuously and only stopped for repairs and servicing of the HVS or if any of the

pavement heaters malfunctioned. Testing of the tracks was terminated after a total of

90000 passes were applied to each section of the lanes. Florida States requires some for

maintenance interventions when the average rut depth exceeds 12.5mm of routes carriage

heavy traffic. The figure below shows the longitudinal rut profiles made after the HVS

run with 90,000 passes for the sections.























Figure 3.5. Longitudinal rut imprints on section 3-3B.

3.9 Heating and Temperature Control

The effects of temperature on rutting have been investigated by many researchers.

Robertson (1997) observed that the shape of the temperature profile in a pavement is









dependent on the air temperature history. There could be several inflection points,

depending on air temperature history of the location.

Thus to simulate Florida climatic conditions, the pavement was heated to 500C.

The heating, measuring and temperature control system of the test track pavement

consisted of three pairs of Watlow Raymax 1525 radiant heaters capable of heating the

space enclosed by the HVS and insulators to a temperature of 500C. Pavement

temperature was measured by 6 K-type thermocouple inserted into the pavement one

each side of the wheel path at 3 pair locations. The average temperature was measured

by the thermocouples at a depth of 2 inches of the pavement and was recorded by the

monitoring computer. To avoid the situation of varying air temperature of the

Gainesville area causing variations in the pavement temperatures, the track was shielded

from the effects of the environment using 3in thick Styrofoam board covered with an

0.08in thick aluminum sheeting. Figure 3.6 and 3.7 shows the thermocouple layout on a

test section and the HVS and insulation assembly.


Figure 3.6. Thermocouple assembly on track 3-2A.





























Figure 3.7. Heavy vehicle simulator with its insulation assembly.

3.10 Rut Measurement

The two non-contact lasers mounted on the Test Carriage measures the vertical

distance between the surface of the pavement and the laser position. The initial runs of

the HVS was set to apply no load as it moves longitudinal with the specified maximum

wander of 4 inches in 1 inch increments. Such runs set the baseline reference for

subsequent runs with load application. The LMI Silicon Lasers have a 0.025%

resolution and collect and output real time data every 4 inches taking 58 transverse data

per longitudinal pass. The two lasers, left and right as they are designated make a

"straight" and diagonal longitudinal movement interchangeably to complete the

maximum wander of 4 inches. The process would be repeated 302 times until each

laser would sweep over a lateral distance of 30 inches. Analysis of the profiler data was

done using a Computer Software written by Tom Byron of the FDOT. The initial

surface profile before the test is subtracted from the measured surface after the test to

obtain the differential surface profile.









3.11 Air Void Content and Thickness Changes

The layer thickness and air void content change with the application of wheel loads

in and just outside the wheel path as a result of rutting of the HMA mixture. The initial

as constructed layer thicknesses and air voids were determined by taking 2 cores each per

section per layer. Their thicknesses were measured with a caliper. The bulk specific

gravity was determined and air voids calculated. After the HVS runs 4 cores each were

taken from each section. Two (2) of those cores were from inside the wheel paths and

the 2 others from just the edge of the wheel path in the humps created by the shearing of

the mixtures. The thickness was measured transverse to the direction of travel of the

wheel along the side of the core after which the top and bottom lift where separated using

a diamond saw. The figure below shows the core locations comparing center of wheel

path to edge of wheel path


Figure 3.8. Location of cores to be taken after the HVS runs on section 3-3B.









3.12 Determination of Viscosity of Cores

Two cores were taken to determine the viscosity of the asphalt shortly after laying.

The asphalt was extracted according to the Reflux Asphalt Extraction procedure, ASTM

2171-95 and recovered from Trichloroethylene solvent using ASTM 5404-97 from cores

taken from the section before and after the HVS runs. Two cores each were taken from

the sections before the runs. After the runs, cores were taken from lanes 3 and 5 only.

Lane 3 has the fine graded mixture while lane 5 has the coarse-graded mixture. In

lane 3, two specimens were taken the bottom lift of section 3A, two taken from wheel

path top lift of section 3B and two specimens taken from the edge (hump) top lift of

section 3C. In lane 5, two specimens were taken from the hump of section 5A in the top

lift, two from the top lift wheel path of section 5B whilst two cores were taken from the

bottom lift of section 5C. The viscosity at 600C of the recovered asphalt was measured

using the Brookfield Viscometer DV III+ according to ASTM D 4402. All asphalt

samples were tested in triplicates. Three shear rates were used in the test. An initial

shear rate of 15 % torque was varied to 50 % and then to 85 % at 1.5 minutes intervals.

The average viscosity was calculated as the viscosity at the different rate of shear at

600C.

3.13 Laboratory Testing on Plant and Laboratory Prepared Mixtures

A series of laboratory tests were performed to determine the mixture response

characteristics and performance prediction parameters of the two mixtures taken from

delivery trucks and stored during the construction phase of the test tracks. Some

specimens were also batched in the laboratory and tested for similar parameters.









3.13.1 Asphalt Pavement Analyzer Test

Rut susceptibility has been correlated to APA rut depths in many researches and a

standard specification has been set by the FDOT for rut resistance mixtures. An upper

limit of 8 mm rut depth in the APA is the standard specification.

Two specimens of 75 mm and 115 mm thickness and 150 mm diameter were tested

in the Asphalt Pavement Analyzer (APA) at 7% air voids. The test was conducted by

Howard Moseley of the FDOT using standard method AASHTO TP 63-03. The

specimen was tested at 640C at 8000 cycles in a bi-directional mode under a 100 + 5psi

pressurized rubber hose. All specimens were plant mixtures sampled from delivery

trucks compacted in a Pine Gyratory Compactor.

3.13.2 Servopac Gyratory Compactor Testing

The mixture's rutting performance prediction was evaluated using the Servopac

Gyratory Compactor with shear and density measuring capabilities. The Servopac

Version 1.23 developed by Industrial Process Control (IPC) of Australia was used.

Both the top and bottom lifts were tested in the Servopac Gyratory Compactor

using the testing procedures developed by Birgisson, et al (2002).

Four 4.5 kg specimens were prepared from plant mixtures sampled from delivery

trucks and laboratory batched specimens using the fine-graded and coarse-graded JMF.

The specimens were compacted using the servopac gyratory compactor. Two specimens

each were compacted to Nmax of 160 gyrations representing traffic level D at a gyration

angle of 1.250, 600kpa of ram pressure, 30 revolutions per minute (rpm) and at 3000F.

The servopac gyratory compactor automatically measured the density and gyratory shear

at every gyration. The bulk density of the compacted specimen was determined and the









machine measured densities corrected. The number of gyration to obtain 7% air void

was determined. The two other specimens were compacted at 1.250 to number of

gyrations to obtained 7% air voids. Then, the angle of gyration was changed to 2.50 and

compacted for 100 more gyrations. Ram pressure, rpm and temperature remained

constant. The gyratory slope and the vertical failure strain were determined from the test

results.

3.13.3 SuperPaveTM Indirect Tensile Test

Superpave IDT test as recommended by Roque et al (2004) was performed on plant

mixtures sampled from delivery trucks during the construction of the test tracks. Truck

1, 2 and 3 contained coarse-graded mix which was used to construct lanes 4 and 5.

Truck 4, 5, 6, 7 and 8 were fine-graded mixtures used in the construction of lane 1, 2 and

3 of the test track.

Only the top lift specimens were tested in the IDT. Two (2) sets of 3 specimens

from each truck sample were tested using the MTS Superpave IDT testing Machines.

Specimens were made from shot-term oven aged specimen compacted to 7+0.5% air

voids in the Servopac Superpave Compactor. Test temperature was 100C for conducting

the Resilient Modulus, Creep and Tensile Strength Test on all specimens.

3.14 Condition Survey

All the 12 pavement sections were inspected before and after the HVS runs. The

main purpose of the survey was to find out whether any other pavement distress apart

from rut has occurred due to the application of the loads. The inspection was a visual

survey and any distress found was marked for further evaluation. Where cracks were






38


found, the width and total area were measured and cores taken so as to assess the causes

for these cracks.














CHAPTER 4
RESULTS, ANALYSIS AND DISCUSSION

4.0 Evaluation of Rut Profiles

Rut profile data give considerable information on the causes, types and mode of

pavements or HMA mixture rutting. The shape of the transverse profile for instance

could be used to determine whether the rutting was due to the surface layers (instability

rutting) or was probable from the bases or the subgrade layers (structural).

4.1 Introduction

Both the rut depth and transverse pavement profiles after the HVS loadings were

analyzed. The average differential rut depth was plotted versus number of HVS passes

from 100 to 90000 passes. The differential rut depth was initially evaluated at every 100

passes and then at 1000 to 5000 as the number of passes grew.

The transverse profiles at 90000 passes were used to compute the Area Parameter.

The area parameter compares the volume of the humps to that of the wheel path

depression and gives an indication of whether rutting was due shear flow or densification.

4.2 Rut Depth

Figure 4.1 shows the differential rut depths of the various test sections versus

number of HVS wheel passes.

Both the fine-graded and coarse-graded mixtures show similar trends. Rut depth

increased rapidly and exponentially with number of passes initially and then flattens out

on the second portion as the number of passes grew. The steep rise comes as a result of

the very large rut depth per pass of around 0.024mm/pass to about 0.000058mm/pass of







40


the wheel. Rut depths development on each test sections increased rapidly initially to

between 2500 and 3000 passes. This was followed by for the very slow rate of rut

development until termination. The rate of growth rut per pass was within a range of

2.7-6.2 x 10-5mm/pass at 90,000 passes for the sections at termination. The initial rate


was more than 400 times the rut growth at termination for the sections. The minimum

and maximum rut depths are 9.78 mm and 16.3 mm for the fine-graded mixture and a

minimum of 13.47 mm and maximum of 17.13 mm for the coarse-graded mixtures


respectively. The average rut depth was 12.8 mm for the fine-graded mixture and 15.1

mm for coarse-graded mixture.

Figure 4.2 shows the maximum differential rut depth for all the sections of the fine

and coarse graded mixtures. Figure 4.3 shows the maximum absolute rut depth for all

the test sections


200

-*-3-3A
,- 3-3B
150 3-3C
3-5B
E ............. .......3-5C
E 3-5C
e 3-5A
41 3-2C
3-2B
3-4A
50 3-4B
3-2A
~- 3-4C

00
00 100000 200000 300000 400000 500000 600000 700000 800000 900000 1000000
NUMBER OF PASSES



Figure 4.1. Plot of change in rut depth versus number of HVS wheel passes.





















E
E

I-
LU
In


Fine
20



15
2 0-------------------


1-







3-2A 3-2B 3-2C 3-3A 3-3B 3-3C AVF


I i Coarse






I










AVC 3-4A 3-4B 3-4C 3-5A 3-5B 3-5C
SECTIONS


Figure 4.2 Maximum differential rut depths for the sections.


20.00
FINE I COARSE
18.oo I00
16.00

14.00

T 12.00


I- 8.00
D
6.00

4.00

2.00

0.00
2A 2B 2C 3A 3B 3C AVF AVC 4A 4B 4C 5A 5B 5C
SECTIONS



Figure 4.3 show the maximum absolute rut depths for all the test sections.


From the analysis above, there seemed be a difference in rut performance of the


two mixtures at 500C. The fine-graded mixture seemed to have out performed the


coarse-graded mixture in both differential and absolute rutting. However statistical


analysis of the means of differential rut depth of the fine-graded and coarse-graded









mixtures showed that there are no significant differences in the rut depths. The Student

t-test statistics at 95% level of confidence was 1.81 at 7 degrees of pooled-freedom whilst

the observed "t" for the differential rut depth was 1.06 indicating the null hypothesis of

equality of the mean of the fine-graded and coarse-graded mixtures could not be rejected.

4.3 Transverse Rut Profile

Considerable amount of information was obtained from the transverse profile of the

sections. As early as 100 passes of the test wheel, both the fine-graded and the coarse-

graded mixtures have developed humps just outside the wheel paths. Humps develop as

a result of shear deformation or instability at relatively low air void content. However

the top lift layers of both mixtures developed humps early in their service life.

The mixtures were compacted to an initial air voids ranging from 92.1% to 93.7%

and 92.4% to 93.5% of Gmm for the fine-graded and coarse-graded mixtures respectively.

After 90000 passes, the air voids of the sections ranged from a minimum of 4.4% to

maximum of 6.7% at termination for the coarse-graded mixtures and 4.3% to 6.4%

respectively for the fine-mixture. This means that humps have developed at relatively

higher air voids for the mixtures giving an indication that most of what was contributing

to the rutting must be shear deformation of the HMA mixture and not densification.

Figure 4.4 show the differential transverse profiles at 100 passes for the fine-graded

mixes and Figure 4.5 shows the same profiles at 100 passes and a profile of a 90000

passes shown on it for comparison purpose. Figure 4.6 and 4.7 show the same plots for

the coarse-graded mixes. The height of the hump at 100 passes for the fine mixture was

about 0.8 mm and that of the coarse mixture was over 1 mm

















1.0 -4------2A--U---------3-2--3-3A----------B---------
1.000



0.500


I--
LL 0.000
o 10 20 30 ) 50 60 7
I--

j -0.500
I--
LU
0 -1.000
LL
U-


-1.500



-2.000
TRANSVERSE SWEEP (IN)



Figure 4.4 Differential transverse profiles at 100 passes of the fine-graded mix.


-- 3-2A -a-3-2B


3-2C 3-3A --3-3 B -*-3-3C 3-33C(90000)passes


10.000

8.000

6.000

S4.000

o 2.000
F--o.ooo

0 ) -2.000 /10 50 60
5 -2.000
I--
L -4.000
LU
-L -6.000
U-
-8.000

-10.000

-12.000
TRANSVERSE SWEEP (IN)



Figure 4.5. Differential transverse profiles at 100 passes as compared with one of 90000
passes for the fine-graded mixture.


- -3-2A -u-3-2B


3-2C


3-3A -3-3B -*-3-3C
















3-5A -MN 3-5B -*-3-5C


1.500


1.000


S0.500
-r
I-
S0.000
i-
S-0.500
-j
-1.000
z

LL
LL
o -2.000


-2.500


-3.000


TRANSVERSE SWEEP (IN)


Figure 4.6. Differential rut profiles at 100 passes of the coarse-graded mixture.


-4-3-4A -3-4B


3-4C 3-5A ---3-5B --3-5C ---3-5A(90000)passes


d.U

6.0

4.0

2.0

0.0


10 2 >ii 40


50 60


TRANSVERSE SWEEP(IN)



Figure 4.7. Differential transverse profile at 100 passes as compare with one the 90000

passes for the coarse-graded mixtures.


-3-4A 3-4B


3-4C










The evolutions of the transverse profile of the fine-graded and coarse-graded

mixtures are similar. In both cases, the first 100 passes gave the largest single

differential rut depth. Approximately, both mixtures recorded just around 2 mm

depression in the wheel path and humps of about 1 mm in the first 100 passes. As the

number of passes grew from 100 passes to 9000 passes, both the depression in wheel

paths and the humps also grew in magnitude and in similar proportion as the initial 100

passes. The downward movement of the wheel path, and the upward progression of the

humps indicated a combination of shear flow and densification occurred during the test

Figures 4.8 and 4.9 show the evolution of the transverse profiles of the fine-graded

and coarse-graded mixtures section 3C and 5A.


10

8
100 passes
E 6
E 200 passes
4 300 passes
2 1000 passes
H- 2
2500 passes
S0 5000 passes
i 20 40 60 100 120 1 0
> 2 7000 passes
-- 10000 passes
-4
m 20000 passes
S-6 50000 passes
-- 70000 passes
a -8
10 90000 passes
-10

-12 -
Laser Travel (0.5 in)



Figure 4.8. Evolution of transverse profile of fine-graded mixture section 3C.














8

6 100 passes

4 -- 200 passes
300 passes
22-
S1000 passes
0 -- 2500 passes
20 60 80 100 120 10 5000 passes
2 -
.> 7000 passes
P -4 10000 passes
1--
S-6 20000 passes
55000 passes
S-8 -- 70000 passes

-10 -- 90000 passes

-12
Laser Travel (0.5 in)



Figure 4.9. Evolution of transverse profile of coarse-graded mixture section 5A.

4.4 Area Parameter Change Method Evaluation of Transverse Rut Profiles

The laser rut profilers attached to HVS measured the initial surface profiles before

the application of the loads thus rut progression had a reference surface for comparison.

The area difference enclosed between the initial surface and the terminal surface

profile was used to determine the type of rutting occurring under the load. Drakos et al,

(2002) using transverse profile data from the Modified APA showed that a net positive

area indicated a predominance of instability rutting whilst a net negative area showed

mainly densification rutting. The positive areas are the humps and the negative areas are

the depressions in the wheel paths. Figure 4.8 and 4.9 show the initial and final surface

profiles for section 3-3C and 3-5A. These sections (3-3C and 3-5A) have the maximum

differential rut depths among all the test sections. The sections with to lowest

differential rut depths were 2A and 3B and were recorded on the fine-graded mixture














320.0
318.0
316.0
314.0
312.0
310.0
308.0
306.0
304.0
302.0
300.0
298.0


0 10 20 30 40 50 60 70
LASERTRAVEL (in)



Figure 4.10. Initial and final surface profile of a fine-graded section 3-3C.


322.0
320.0
318.0
316.0
314.0
312.0
310.0
308.0
306.0
304.0
302.0
300.0


0 10 20 30 40 50 60 70
LASER TRAVEL (in)


Figure 4.11. Initial and final surface profile of a coarse-graded section 3-5A.

The Area Change Parameter (ACP) can be calculated for any transverse profile.


Using Loess curve fitting function or polynomial after linearization of the rut profiles, the


areas under the initial and final surface profiles were calculated by integrating the











polynomial from one end of the cross section to the other of the transverse profile for all


the 12 sections. Loess allows for modeling of the parametric regression surface of the


transverse rut profiles so they could be integrated. Table 4.1 shows the areas under the


original and final surface profiles and the ACP as well the Differential and Absolute rut


depths of the sections.


Table 4.1. Area change parameters of the fine-graded and coarse-graded sections.


FINE-GRADED MIX
Initial Area Absolute Rut Differential Area Change Area Change
Section ID (in2) Final Area (in2) Depth Rut Depth (in2) Parameter
3-2A 734.581 732.944 11.22 9.8 1.637 0.223
3-2B 734.581 732.944 10.37 10.1 1.637 0.223
3-2C 734.581 732.944 13.12 13.8 1.637 0.223
AVERAGE 734.581 732.944 11.57 11.2 1.637 0.223

3-3A 728.754 726.216 10.58 12.3 2.538 0.348
3-3B 612.683 609.135 11.22 12.7 3.548 0.579
3-3C 728.571 724.578 15.24 16.9 3.993 0.548
AVERAGE 690.003 686.643 12.35 14.0 3.360 0.492


COARSE-GRADED
Initial Area Absolute Rut Differential Area Change Percentage
Section ID (in2) Final Area (in2) Depth Rut Depth (in2) Area Change
3-4A 742.882 740.91 13.55 13.6 1.972 0.265

3-4B 734.558 732.954 14.18 13.5 1.604 0.218
3-4C 728.055 726.825 17.03 15.8 1.230 0.169
AVERAGE 735.165 733.563 14.92 14.3 1.602 0.218

3-5A 736.521 732.954 15.03 17.1 3.567 0.484
3-5B 725.882 724.613 11.01 14.2 1.269 0.175
3-5C 736.172 732.748 12.28 16.2 3.424 0.465
AVERAGE 732.858 730.105 12.77 15.8 2.753 0.375


As shown the positive ACP indicates that primarily both the fine-graded and


coarse-graded sections exhibited instability rutting.


There is a theoretical basis for the ACP. Consider the transverse profile of a rutted


surface superimposed on the initial or existing surface before trafficking as shown below.








Typically this is how HMA pavements with adequate structure behave under loads when

instability rutting is the rule. If for some reason for instance, higher initial air voids

contents immediately after compaction of the mixture, then consolidation rutting with the

absence humps at the edges but only depressions in the wheel paths will be seen from the

transverse profile plots

nginal Protile



'- DeCfoiinedPfile A







.J.1,A -... F.,* Ir I




I Ii! JI ,





Figure 4.12. Positive A1 and negative A2 areas of a transverse rut profile.

If HMA material was moved from A2 and equal amount was transferred to A1 then

shear deformation was primarily contributing to the rutting. On the other hand if less

was transferred that is A1 less than A2 then, some considerable amount of consolidation

would have occurred and consolidation would be controlling rutting.









Table 4.1 shows the differences of the Area Change Parameter of both the fine-

graded and coarse-graded mixtures. The highest area-change determined for the fine-

graded mixture was 0.579 in2 for section 3B which recorded a differential rut of

12.7 mm. In section 3C where the highest differential rut depth recorded was 16.9

mm the area-change parameter determined was 0.548 in2. In general there seem to be no

direct relationship between the differential rut depth and the Area Change Parameter.

Since both fine-graded and coarse-graded mixtures exhibited shear deformation

primarily, the ACP was unable to distinguish their rutting performance. The Area

Change Parameter can be used in the evaluation of transverse rut profiles to determine the

type of rutting occurring on HMA pavements.

The calculations of the Area Change Parameter are as shown in Appendix C.

4.5 Evaluation of Core Densities

Tables 4.2 and 4.3 show the air voids contents of the cores from the wheel paths

and the edges of the wheel paths after the HVS runs for the fine-graded and coarse-

graded mixtures respectively.

Bulk densities of cores extracted from the test sections were determined according

to AASHTO T 166-93 and the air voids calculated using the Gmm for each truck sample.

The Gmm was determined using the Rice method for the determination of the

Maximum specific gravity of bituminous paving mixtures, AASHTO MT 321.









Table 4.2. Air voids level of the cores from the wheel paths and edges of the wheel paths
lane 3 fine-graded mixture sections.


Generally there was a larger reduction of air voids in the wheel paths for the top

and bottom lifts of the fine mix gradation than that the coarse-graded mixtures. At the

top lift, air voids reduction ranged from 2.9 to 4.3%. As expected the bottom layers have

less air void reduction and at section 3-3A there was no reduction at all at the bottom lift.

The overall contribution of densification for the fine-graded mixture to rutting

ranged from 16% to 22%. Thus much of the rutting was due to shear deformation

creating humps at the edge of the wheel paths.











Table 4.3. Air voids level of the cores from the wheel paths and edges of the wheel paths
of lane 5, coarse mixture.


age Age Total Rut
Average Total
Air decrease Rut due to depth after
Sample Wt in Wt in SSD Gmb Initial percentage
lane section location N Gmm Voids wheel path t densification HVS 90000 eret
No Air(g) H20(g) ,, ., thickness rut due to
(%) air voids (mm) passes ific
(mm) densification
(%) (mm)
2 2335.1 1385.5 2336.1 2.456 2.573 4.5 8 55 11 1
center 2.8 54.5 1.51 17.10 15.59
4 2239.7 1328.4 2240.8 2.455 2.573 4.6
topA 1 2370.5 1385.7 2377.2 2.391 2.573 7.1
edge 3 2580.2 1500.4 2585.7 2.377 2.573 7.6

2 2263.6 1344.8 2265.3 2.459 2.573 4.4
center 2.0 53.6 1.07 14.20 12.62
4 2273.7 1346.9 2274.8 2.450 2.573 4.8
op1 2343.8 1371.4 2348.6 2.398 2.573 6.8
edge 3 2485.3 1457.8 2490 2.408 2.573 6.4

2 1944.2 1139.8 1947.6 2.407 2.579 6.7
center 0.1 53.2 0.03 16.20 5.02
r 4 2177.3 1293.1 2178.5 2.459 2.579 4.6
opC 1 2058.7 1211.7 2062 2.421 2.579 6.1
edge 3 2225.5 1315.8 2227.2 2.442 2.579 5.3

S 2 1833.5 1104 1834.2 2.511 2.572 2.4 22
center 2.2 52.2 1.15
S tt4 2061.8 1229.6 2062.8 2.475 2.572 3.8
e d 1 2064 1218.4 2065.5 2.437 2.572 5.3
edge 3 2168.2 1279.1 2169.3 2.436 2.572 5.3

2 2124.8 1273.2 2125.8 2.492 2.572 3.1 1
center 1.3 53.3 0.72
54 1970.4 1175.9 1971.5 2.477 2.572 3.7
1 2156.2 1276.8 2157.1 2.449 2.572 4.8
edge 3 1886.2 1117.1 1886.9 2.450 2.572 4.7

2 1988.8 1184 1989.6 2.469 2.568 3.9 520 078
5 4 1861.4 1115.2 1862.2 2.492 2.568 3.0
S bottomC
1 1902.4 1121.5 1903.6 2.432 2.568 5.3
edge 3 2197.8 1301.7 2198.5 2.451 2.568 4.6


There was considerably less reduction in air voids of the wheel paths of the


coarse-graded sections than the fine-graded sections. The air void reduction of the top


lift ranged from 0.1 to 2.8% and the bottom lift from 1.3 to 2.2%. The percentage of


rutting due to densification ranged from 5 to 16 % indicating that much more shearing


deformation occurred in the coarse-graded mixture than the fine-graded mixture.


It was observed that sections with lower densification show higher rut depths. This


suggests that at optimal (6 to 8%) level of field compaction of the mixtures, rutting was


associated primarily with mixture's shear resistance measured by resistance to aggregate


sliding and rotation within the mixture. Shearing was induced by lateral stresses








53



generated by radial tires and it seems the coarse-graded mixtures with greater proportion


of coarse aggregates developed bigger humps than finer mixture under the same wheel


loads.


The change in air voids between the wheel paths and humps quantifies the


proportion of the total rutting associated with densification.


4.6 Evaluation of Recovered Asphalt


Table 4.4 shows the average viscosity of the Asphalt recovered from the wheel


paths, top lift, top lift humps and bottom lift wheel path, for both fine-graded and coarse-


graded mixtures. The viscosities at 60 OC of the recovered asphalt are in Appendix C.


Table 4.4. Viscosity of recovered asphalt


VISCOSITY OF RECOVERED ASPHALT AT 60C POISE
FINE MIX

Section Location Sample ID Top lift
I 2711720

13 2526104
3B center
S 3023872
average 2753898.67

E 4162000

3C edge E3 4543666
3C edge
*F 2960397

average 4352833


Section Location Sample ID Bottom lift
A 1198604

A3 758256
3A center
B 1986190

average 1314350


Section Location Sample ID

K

5B center
L
average

G

5A edge G3
**H
average


Section Location Sample ID

C

5C edge C3
D1
average


In the wheel paths for both fine-graded and coarse-graded mixtures, recovered


asphalt viscosities did not show any significant differences. The fine-graded mixture has


an average viscosity of 27,539 poises whilst the coarse-graded mixture has average


COARSE MIX

Top lift

2867903

2551666

2546778

2655449



2907439

655747

2907439


bottom lift

1333246



1168119

S1250682.5









viscosity of 26, 555 poises. Similar comparison could be made of the bottom lifts. The

fine-graded mixture recorded an average of 13,144 poises in the wheel path whilst the

coarse-graded mixture recorded 12,509 poises in the edges. Thus there are no significant

differences in viscosities for the bottom lifts of the two mixtures. In the humps both the

fine-graded and coarse-graded mixtures recorded viscosities which are two times their

respective viscosities of the bottom lifts. That indicated the effects of the ageing process

of the environment on the top 50 mm of the pavement.

It was also observed that the fine-graded mixture had two times the viscosity of the

coarse-graded mixture for top lift in the humps as show for sections 3C and 5A in

Table 4.4 however, differential rut depth recorded from the HVS show that sections

3C and 5A have the highest and similar rut depths of 16.9 and 17.1mm for fine-graded

and coarse-graded mixtures respectively.

From Table 4.2 and 4.3, the average termination air voids in the humps of sections

3C and 5A are 8.95 and 7.35% indicating a difference 1.6%. The wheel paths air voids

were relative the same for the sections averaging 4.65 and 4.55% for 3C and 5A

respectively. The 1.6% higher field air voids content in the humps of the fine-graded

mixture probably was responsible for doubling the asphalt viscosity in the fine-graded

section. Due to the limited data the above analysis did not show conclusively whether or

not asphalt viscosities had any influence on the rutting patterns of the sections. Specimen

F, G and H got contaminated and were excluded from the calculations.

4.7 Asphalt Pavement Analyzer Test

The mixtures were assessed based on their APA test results. Generally APA test

result gives an indication of mixture future field rutting performance. However APA












results have not been known to distinguish between fine-graded and coarse-graded


SuperPaveTM mixture rutting potential.


The APA test was conducted by the FDOT at 8000 cycles and at 640C. The results


are as shown in Table 4.5. The fine-grade mixture has APA rut depth ranging from 3.3


mm to 4.0 mm with a mean of 3.7 mm for the 75 mm specimen and 3.4 to 3.6 mm with


an average of 3.5 mm for the 115 mm specimen for the top lift.


Table 4.5. Asphalt pavement analyzer rut depth for both fine-graded and coarse-graded
mixture.


Mix Density (%) APA Rut Depth (mm), 8000 cycles HVS Rut Depth (mm)
Lift Type HVS Lane Lab Air Voids Gmm 75 mm specimen 115 mm specimen 90,000 passes*
2A 3.3 93.7 4.0 3.6 9.8
2B 3.3 93.4 4.0 3.6 10.1
2- 2C 3.6 93 3.5 3.4 14.8
2 Average 3.4 93.4 3.83 3.53 11.57
S 3A 4.3 92.6 3.3 3.5 12.2
S 3B 4.3 92.1 3.3 3.5 12.7
3C 3.3 92.1 4.0 3.6 16.9
Average 3.97 92.27 3.53 3.53 13.93
0" 4A 4.3 93.7 3.1 3.3 13.6
I.-
4B 4.3 92.6 3.1 3.3 13.5
4C 2.7 92.1 2.8 2.9 15.8
SAverage 3.77 92.80 3.00 3.17 14.30
5A 4.5 92.6 2.3 3.3 13.6
o 5B 4.5 92.6 2.3 3.3 17.1
5C 4.3 92.6 3.1 3.3 14.2
Average 4.43 92.60 2.57 3.30 14.97
2A 3.3 93.3 4.3 2.9
2B 3.9 94.2 3.8 3.0
2C 2.9 92.8 3.8 3.0_
S Average 3.37 93.43 3.97 2.97_
3A 4.0 92.4 2.5 2.9
S 3B 4.0 92.5 2.5 2.9
3C 4.8 92.5 not tested 3.0_
S Average 4.27 92.47 2.50 2.93_
4A 4.3 92.6 2.8 2.9
4B 4.3 93.6 2.8 2.9
4C 4.4 93.7 2.8 2.6_
SAverage 4.33 93.30 2.80 2.80
5A 4.5 93.6 2.8 2.6
o 5B 4.5 93.7 2.8 2.6
5C 4.3 93.3 3.2 2.9
Average 4.43 93.53 1 2.93 2.70


The coarse-graded mixture has ranges of 2.3 mm to 3.1 mm and average of 2.8 mm


for 75 mm pills and 2.9 mm to 3.3 mm with an average 3.0 mm for the top lift









respectively. The bottom lifts for both the fine-graded and coarse-graded mixtures were

3.4 mm and 2.9 mm for 75 mm pills and 3.0 mm and 2.8 mm for the 115 mm pills

respectively.

APA 75mm APA115mm O HVS RUTS
18.0
115M M
16.0 FINE COARSE

14.0 AVERAGS 75M M
I r 75MM
12.0
HVS
10.0 I

8.0

6.0

4.0

2.0

0.0 lill
2A 2B 2C 3A 3B 3C AVF AVC 4A 4B 4C 5A 5B 5C


Figure 4.13. Asphalt pavement analyzer comparisons and HVS rut depths for the fine-
graded and coarse-graded mixture of the top lift.

The APA rut depths for the fine-graded mixture has average of 3.7 mm and 3.5 mm

for the 75 mm and 115 mm respectively whilst the coarse-graded mixture has average rut

depths of 2.8 and 3.0 mm for the 75 and 115 mm.

Examination of results indicated no significant differences for the 115 mm

specimens but the 75 mm thick specimens, the coarse-graded mixtures seemed to have

out performed the fine-graded mixture.

Drakos (2002) concluded that the APA hose does not simulate the stress state in the

field during rutting and may not distinguish well the field performance of fine-graded and

coarse-graded SuperPaveTM mixtures










4.8 SuperPaveTM Servopac Gyratory Compaction Results

The Superpave servopac gyratory results were used as rutting performance

predictor using the "pass-failure" criteria developed at the University of Florida. The

gyratory shear slope against the initial vertical strain obtained by changing the gyratory

angle from 1.250 to 2.50 could be used to evaluate mixtures rutting resistance.


800

700

600

500

400 -- fine-graded mixture
coarse-graded mixture
W 300

200

100

0
0 50 100 150 200
NUMBER OF GYRATION



Figure 4.14. Average gyratory shear stress versus number of gyrations for bottom lift.

Generally the gyratory shear stress of the fine-graded mixtures appears to be higher

for both the top and bottom lifts. The figures below show the average gyratory shear

stress against the number of gyrations for the bottom and top lift.

The magnitude of the gyratory shear stress in itself does not mean much in terms of

the rutting potential. It is not a fundamental property of a mixture and could therefore

not be used to evaluate the rut resistance of neither fine-graded nor coarse-graded

SuperPaveTM mixtures.










However the slope of the gyratory curve in the air voids range of 4% to7% in

conjunction with the initial vertical strain obtained from the gyratory compactor can be

used to screen out mixtures with a high rutting potential.




900
-< fine mixture
800
-- -coarse mixture
uo 700
600
U)
5 500
= 400
UO
I 300
0
200
> 100
0
0 50 100 150 200
NUMBER OF GYRATIONS



Figure 4.15. Average gyratory shear stress versus number of gyrations for the top lift.

Figure 4.16 shows the gyratory shear slope for section 3-3C between air voids of

4% to 7% of the top lift of the fine-graded mixture. The slope of the plot of change in

gyratory shear versus the change in air voids between 4% and 7% gives an indication of

the mixtures rut resistances. Values above 15 kPa have been shown to give good field

rutting resistance (Darku 2003).

The ability of a mixture to generate enough shearing response to resist straining

after the gyration angle is changed from 1.250 to 2.50 has been shown to affect the rutting

resistance of mixtures. Figure 4.17 shows the plot of the gyratory shear versus the

number gyrations during the change of the gyratory angle for calculating the strains.















y = 24.00x+ 672.09
R2 0.98





7% to 4% AV


) 3.5 4.0 4.5
Natural Log Revolutions


Figure 4.16. Gyratory shear stress versus number of gyrations of the servopac compactor
for section 3C fine-graded section.




1000

900 -

800
02.50
700

600 1.25

500 -

400 -

W 300 -

200

100

0 -
0 20 40 60 80 100 120 140
Number of Gyrations
-- Shear Str.



Figure 4.17. Gyratory shear stress versus number of gyrations for 3-3C with a change of
gyration angle from 1.250 to 2.50.










Trucks 1,2 and 3 are coarse-graded mixtures, Trucks 4, 5, 6, 7 and 8 are fine-graded

mixtures for the top lift. Trucks 1, 2, 3 and 4 are fine-graded mixtures.

Trucks 6, 7 and 8 are coarse-graded mixtures for the bottom lift. Generally the

fine-graded mixture have higher gyratory slopes but lower vertical strains whilst the

coarse-graded mixtures showed lower slopes and marginally higher vertical strains. The

fine-graded mixtures could be described as brittle according Birgisson et al (2002)

"Pass/Failure Criteria" for both top and bottom lifts. Figure 4.18 shows the plots

of the gyratory shear slope versus the initial failure strains of the top lift.





40.0
S35.0 fine truck 1
I
65 30.0 truck 2
S25.0 coarse truck 3
S20.0 / truck 4
S* truck5
150 -----------. truck
10.0 truck
5.0 I truck
0.0 '----- A jmf fine
1.00 1.20 1.40 1.60 1.80 2.00 2.20 2.40 jmf coarse
Vertical Failure Strain



Figure 4.18. Gyratory slope and initial failure strain of the top lift fine-graded and
coarse-graded mixtures

According to Birgisson et al (2002) the fine-graded mixtures could be described as

brittle whilst the coarse-graded mixtures could be described as low shear strength and of

marginally brittle. Both mixtures may rut when load but from different mechanisms.

The coarse-graded mixtures may have rutted due their low gyratory shear strength.







61






30

I I
25 I
I I

(9 20 brittle optimum | plastic
0
15 --_ --- -- 1- --I -------
I I
10 I
SloA? shear strength
5 5
I I
0
1 1.2 1.4 1.6 1.8 2 2.2 2.4
Strain at Initial Minimum Shear(%)


Figure 4.19. Pass/Fail criteria for evaluation of rut resistance.

The gyratory slope ranged from 7.33 kpa to 13.4 kpa whilst the differential rut

depth ranged from 13.5 mm 17.1 mm. Figure 4.20 shows the relation between the

gyratory slope and the differential rut depth for the coarse-graded mixture of the top lift

with R2 of 0.91.




2 20
I18 5A y -0.5153x + 20.363
16R2 = 0.9138
14 4C
S12
lO ^4A, 4B
10
8
6 differential rut depth
S4
,L 2
0
5 7 9 11 13 15
GYRATORY SLOPE (kpa)



Figure 4.20. Differential rut depths versus gyratory slope of the coarse-graded mixture of
the top lift.












y =-0.1256x + 15.346
18 R2 = 0.0305
16
14
12 -
10
8
6 -* rut depth vrs slope
4
2
0
0 5 10 15 20 25 30



Figure 4.21. Differential rut depths versus gyratory slope of the fine-graded mixtures of
the top lift

The rutting patterns of the fine-graded mixtures in relation to the slope of the

gyratory shear stress and the vertical failure strains are not good. The gyratory shear

slope ranged from 16.79 kpa for sections 3A and 3B to 34.5 for lane 1 where there no

was rutting study. The differential rut depths recorded for the fine-graded sections

ranged from 9.8 mm for section 2A to 16.9 mm for section 3C. Figure 4.21 shows the

relation between the slope and the differential rut depth of the fine-graded mixture. The

R2 obtained was 0.03 implying a very weak correlation.

According to Darku (2003) mixtures outside the optimal zone of Figure 4.19 such

as the fine-graded mixtures may show rutting and or cracking with load application early

in-service.

4.9 Evaluation of Gradation of HVS Track Mixture

Figure 4.22 shows the gradation of the top lift gradation for both the fine-graded

and coarse-graded mixtures.







63




---JMF fine
100
100 truck
90 0 truck
80 truck
z 70 -- -truck5
60 truck
c' 60
9 truck3 coarse
W 50
,, CONTROL POINTS
4 40
z RZ
30 truck 2 coarse
20 ---truck 1 coarse
10 _- JMF coarse
0 *
0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00
SI EVE SIZ 0.45


Figure 4.22. Gradation of the fine-graded and coarse-graded plant mix mixtures of the top
lift.

Gradation is the most important parameter that affects the resistance of the mixtures

to carry load. Aggregate in a mixture may work together to resist deformation when

they interact among themselves especially the coarse fractions. Coarse fraction in a

mixture is defined as the materials retained on the 1.18 mm sieve.

Lambe and Whitman (1969) presented data to show that the maximum porosity of

the loose dry soil in contact ranged from 45% to around 50%. In this state, particles

interaction is at the minimum and the resistance to deformation is also at the minimum.

Asphalt concrete mixtures are in general loose soils particles glued together with

asphalt binder and could be likened to soils in the way they pack. Figures 4.23 and 4.24

show the unit interaction plots of the fine-graded mixture.




















--*-JMF
-_ -T-1-A
-A- T-1-B
--T-1-C
---T-2-A
- Limits


12.5- 9.5- 4.75- 2.36- 1.18- 0.6-0.3
9.5 4.75 2.36 1.18 0.6
contiguous sizes, mm


0.3- 0.15- 0.075-0
0.15 0.075


Figure 4.23. Interaction unit check for lane 1 and section 2A of lane 2 of the fine-graded
mixture top lift.


-- T-2-B
--_ -T-2-C
-- T-3-A
----T-3-B
--- T-3-C
- Lirrits


12.5- 9.5- 4.75- 2.36- 1.18- 0.6-0.3 0.3-
9.5 4.75 2.36 1.18 0.6 0.15
contiguous sizes, mm


0.15- 0.075-0
0.075


Figure 4.24. Interaction unit check for lanes 2 and 3 for the fine-graded mixture top lift.


Analysis performed by the Materials Group at The University of Florida on the


spacing characteristics of aggregates in contact showed aggregates particles interact as a










unit to resist loads if the relative proportion of one of the sieves size in two contiguous

sieves is between 70 and 30%. Figure 4.25 shows the porosity of the

fine-graded mixtures for the range of particles between 4.75 mm and 1.18 mm for no

interaction and interaction among the particles and their relation with the differential rut

depth of fine-grade mixture.




E 70
E
-60
S No interaction With interaction
50
Ca
S40

S30
I
20 -
0

o 0
2A 2B 2C 3A 3B 3C
No. interaction
Sections With interaction
Differential rut depth


Figure 4.25 Porosity of the fine-graded mixture for interaction and no interaction.

Considering the no interaction for the size range 4.75 mm to 1.182 mm, the

porosity calculated seems to relate the differential rut depth shown in Figure 4.25. That

is the higher the porosity the higher the differential rut depth. Where interaction is

considered, the unit interaction plot show that section 2C and 3A were the most out of

tolerance (close 70%) meaning that the stress resisting interaction is less and hence

higher susceptibility to differential rutting and higher rut depths.







66





100
90
80
S70 - -- - - - -- JMF
60 --- T-4-A
50 -- T-4-B
40 T-4-C
0- T-5-A
30- ----- -------------------------
- Limits
20
10
0
12.5- 9.5- 4.75- 2.36- 1.18- 0.6-0.3 0.3- 0.15- 0.075-
9.5 4.75 2.36 1.18 0.6 0.15 0.075 0
contiguous sizes, mm



Figure 4.26 Unit interaction plot for the coarse-graded mixture of the top lift.

The coarse-graded mixtures show similar trend. Figure 4.26 shows the unit

interaction plot for the coarse-graded mixture for section 4A, 4B, 4C and 5A.

The differential rut depth was 17.1 mm and 15.8 mm for sections 5A and 4C

respectively. Considering interaction for the two contiguous sieves of 2.36 mm and 1.18

mm range, 5A was the most out tolerance followed by 4C. Thus the relative potential of

mixtures to rut could be predicted during design using the combination of porosity and

unit interaction calculated for the coarse aggregate fractions in any mixture.

4.10 Evaluation of SuperPaveTM Indirect Tensile Test Results

The results of the SuperPaveTM IDT test included the resilient modulus, m-value,

strength and dissipated creep strain energy. These parameters were used to calculate the

energy ratio (ER) of the mixtures evaluated.










The energy ratio is defined as the dissipated creep strain energy of the mixture

divided by the minimum threshold dissipated creep strain energy to resist cracking. A

mixture with a low ER would a higher potential to top-down cracking.

Table 4.6 shows the results of IDT test parameters for Truck 1-3 for the coarse-

graded mixtures and that of fine-graded sections represented by Trucks 4-8.

Table 4.6. Indirect tensile test parameter versus servopac test parameters.

Truck St MR FE UU EHMA Stress UU MIN Vertical
No ID m-value D, (Mpa) (Gpa) (kJ/m3) (kJ/m3) (psi) a (kJ/m3) ER Slope kPa strain
1 5 0.50394 5E-07 1.7 10.76 1.1 0.965706 150 5.05-E8 1.398391 0.69058 73
6 0.4523 4E-07 1.65 13.57667 0.8 0.699736 150 5.08E-8 0.742362 0.94258
5 0.40322 5E-07 2.05 11.68667 1.6 1.420201 150 4.9E-08 0.753077 1.88587
2 ---- -- 13.2 1.4
6 0.46745 4E-07 2.29 12.29333 2.7 2.48671 150 4.7E-08 0.859845 2.89204
5 0.46253 4E-07 2.14 11.38667 2 1.798905 150 4.8E-08 0.88444 2.03395 7.8 1.4
3 7.8 1.4
6 0.48567 6E-07 1.98 10.75 2.7 2.517656 150 4.9E-08 1.463776 1.71997
5 0.43949 4E-07 2.35 12.96667 2.1 1.88705 150 4.7E-08 0.790847 2.38611 16.8 1.25
4 16.8 1.25
6 0.41818 4E-07 2.31 12.54 1.6 1.387237 150 4.7E-08 0.699797 1.98234
5 5* 24 1.15
6 0.47109 4E-07 2.12 13.42333 1.8 1.63259 150 4.8E-08 0.904606 1.80475
6 5 0.4174 4E-07 2.13 11.62 1.5 1.304781 150 4.8E-08 0.64806 2.01336 17.7 1.1
6 17.7 1.1
6 0.44222 5E-07 2.32 13.04667 2.3 2.093725 150 4.7E-08 0.901537 2.3224
5 0.50499 5E-07 2.52 11.99 3.8 3.535179 150 4.6E-08 1.280389 2.76102
7 13.9 1.1
6 0.52377 4E-07 2.35 13.26667 2.8 2.591866 150 4.7E-08 1.208824 2.14412

8 --- 34.5 1.2
6 0.50059 5E-07 2.42 12.06333 4.3 4.057264 150 4.7E-08 1.237616 3.27829


The ER calculated ranged from 0.691 to 2.89 with an average of 1.69 for the

coarse-graded mixture and from 1.8 to 3.28 with an average of 2.34 for the fine-graded

mixture. With exception of the specimen from Truck 1 the rest of the mixtures evaluated

had ER greater than 1, the threshold below which mixtures are susceptible to top-down

cracking. Generally the fine-graded mixture appears to have higher ER than the coarse-

graded however statistically there are no significant differences in the means of the ER

and the failure strains. The relationship between ER of the mixtures and the gyratory









shear slope of both mixtures was found to be fair with R2 of 0.5049 as shown in Figure

4.23.





40
4y = 9.1228x 2.4521
35 R2 = 0.5049
S30
S25 -
0

o 15 -
10 -
5
0
0 1 2 3 4
Energy ratio



Figure 4.27. Relationship between gyratory shear slope and the energy ratio of both the
fine-graded and coarse-graded mixtures.

4.11 Further Evaluation of the Fine-Graded Mixture

Further evaluation of the differential rut depth was done to compare the fine-graded

mixture to coarse-graded mixture after eliminating sections 3C and 5A. Sections 3C and

5A recorded the highest differential rut depth after running the HVS. Section 3C and 5A

recorded 16.9 mm and 17.1 mm differential rut depths respectively. The relatively high

rut depths of these two sections could be attributed to the higher coarse-fraction porosity

or the marginal interacting unit checks as shown in Figures 4.24 and 4.25. Figure 4.28

shows the differential rut depth for the fine-graded and coarse-graded mixtures with the

exclusion of sections 3C and 5A. The average differential rut depths of the fine-graded

and coarse-graded mixtures are 11.3 mm and 13.22 mm, respectively.













16.00

14.00
E
E 12.00
r-
- 10.00
-a
B 8.00

6.00

S4.00
o
2.00

0.00


2A 2B 2C 3A 3B AVF AVC 4A 4B 4C 5B 5C
Sections



Figure 4.28. Comparison of fine-graded and coarse-graded differential rut depths.

Comparing the difference between the mean rut depths of the fine-graded mixture

and the coarse-graded, we realized that statistically there was no significant difference in

the mean differential rut depth at 95% confidence level.

4.12 Comparison of the Fine-graded Mixture of HVS Round 1 to HVS Round 3

Research conducted at the FDOT Materials Office in Gainesville in 2002 compared

the rutting resistance of fine-graded SuperPaveTM mixtures with and without SBS

polymer modified binder and the conclusion drawn was that, the SBS modified mixture

out-performed the mixture without SBS polymer modified binder.

In our analysis we compared the rutting resistance of the fine-graded mixture to the

fine-graded unmodified mixture in "HVS Round 1" using the differential rut depths at

50000 passes of the test wheel. The HVS running modes and the temperatures of testing

were the same in the two projects. The type of aggregates used for both projects were

different. "HVS Round 1" was constructed with limestone from south Florida whilst


Fine I Coarse
-i








70



"HVS Round 3" was constructed with granite from Georgia. Figure 4.29 and 4.30 shows


the gradation of the Job mix formula and the differential rut profiles for both fine-graded


mixtures.


100.0

90.0

80.0

70.0

60.0

50.0

40.0

30.0

20.0

10.0


0.0 +-
0.00


1.50 2.00 2.50
Sieve size (nmm) ^0.45


Figure 4.29. Job mix formula HVS Round 1 and HVS Round 3 for the fine-graded
mixtures.




25.0

3A round
20.0 3B round
20.0
3C round
--2C round
15.0 2B round
-- 2A round
u4A round
10.0 10 4B round
5A round
5B round
5.0



0.0 ..
0.0 20000.0 40000.0 60000.0 80000.0 100000.0 120000.0
NUMBER OF PASSES



Figure 4.30. Differential rut depth of the fine-graded mixture HVS Round 1 and HVS
Round 3









The effective asphalt content of the JMF for "HVS Round 1" was 4.9% and that of

HVS Round 3 was 4.5%.

Differential rut depth recorded in HVS Round 1 (lanes 4 and 5) appears to be

higher than that HVS Round 3 (lanes 2 and 3) at 50000 passes. HVS Round 1 recorded a

minimum of 13.2 mm and a maximum 19.5 mm with an average of 15.8 mm whilst HVS

Round 3 recorded a minimum and a maximum of 8.9 mm and 14.9 respectively with an

average of 11.3 mm.

Gradation of the two mixtures are different and is the parameter with significant

influence on mixture or pavement rutting performance that could have contributed to the

differences in rutting performance between HVS Round 1 and HVS Round 3. Analysis

of the gradations using the interacting unit check and the porosity of the coarse

aggregate-fraction (12.5 mm to 1.18mm) were done to see if any differences exist for the

two mixtures. Figures 4.31 and 4.32 show the interacting unit check and porosity plots

of the gradations of the JMF of the two mixtures as well as lanes 4 and 5 of "HVS Round

1". The interacting unit check indicates that aggregate size-range 4.75 mm to 1.18 mm

are acting together to resist stresses within the two mixtures. In both cases, there is a

break in interaction at 9.5 mm to 4.75 mm. "HVS Round 1 mixtures (JMF and lanes 4

and 5) aggregates-size ranging from 12.5 mm to 9.5 mm are floating in the mixture and

do not contribute shearing stress resistance of the mixture.

For the interaction at 2.36 mm to 1.18 mm HVS Round 1 mixtures are closer to the

tolerance limits and has higher porosities at all interaction ranges than HVS Round 3

mixture which makes it more susceptible to rutting. It is clear from the above analyses







72


that HVS Round 3fine-graded mixture has better rutting resistance than HVS Round 1


unmodified fine-graded mixture.


100/0
90/10
80/20
0 70/30

60/40 -
p. *hs 3 JMF
(, 50/50
S\ hv\ 1JMF
40/60 - Limits
= 30/70 ----lane 4

E 20/80 -- ----------- -- - ---- ---lane 5
10/90
0/100

Lo LO (D 0- (o co ,- O o.
d d I ?
1- C- d
0 i
Contiguous sieve sizes, mm


Figure 4.31. Interacting unit check for "JMF" "HVS Round 1 and 3" and lanes 4 and 5.


70.00

60.00

50.00

40.00 -
U HVS3
S30.00 aHVS 1
o

20.00

10.00

0.00
2.36-1.18 4.75-1.18 9.5-1.18 12.5-.1.18

Contiguous sizes (mm)



Figure 4.32. Porosity of job mix formula for "HVS Round 1" and "HVS Round 3"
fine-graded mixture.









4.13 Condition Survey Results

The condition survey undertaken on all the 12 sections showed no visible tire

imprints, traveling or cracking except section 3B where two longitudinal cracks of width

of about 0.5mm and length 4 5 inches have appeared. Figure 4.33 shows one of such

cracks.





















Figure 4.33. Longitudinal crack on section 3B

When cores were taken, it was realized that the crack was induced by a strain gage

inserted between the bottom lift and the top lift for strain measurement. It was also

realized that the strain gages in the locations of cracks were nearer to the surface of the

top lift. The mixture seemed loose around the gages suggesting some amount of

segregation had occurred during the laying process. Segregation of the mixture could be

observed on the sides of the cores taken at the section at about 1 inch to 2 inches of the

top lift. The mixture had also stripped slightly and the fines pumped to the surface as

shown in Figure 4.33. Figure 4.34 shows the core with the strain-gage induced crack at









the top and mix segregation around the strain gage as well as the joint between the top lift

and the bottom lift.


Figure 4.34. Strain gage induced crack on the top lift of section 3-3B fine-graded
mixture.















CHAPTER 5
CONCLUSIONS AND RECOMMENDATIONS

5.1 Conclusion

Analysis of the results of both the average differential and absolute rut depth

measurement show that the difference in the mean rut depths of the coarse-graded and

fine-graded mixture was statistically insignificant.

From the limited data collected, we can not conclude that the SuperPaveTM

Coarse-graded mixture is either better or worse than the SuperPaveTM fine-graded

mixture.

When the transverse profiles were evaluated, the Area Change Parameter results

show that both the fine-graded and coarse-graded mixture exhibited primarily instability

rutting. The shape of the transverse profile at 100 passes of the HVS wheel shows that

the instability rutting of both mixtures developed relatively early in the service life of the

pavements.

Based on the limited Servopac test results, the plots of the gyratory shear slope

versus the initial vertical failure strain for the fine-graded and coarse-graded mixture

show that they were both "brittle". The fine-graded mixture however had relatively

higher gyratory shear strength and lower strain potential than the coarse-graded mixture.

Examination of the test results of the viscosities of the recovered asphalt from the

cores were very close to another and that its effect on mixture rutting could not be

effectively evaluated









Previous conclusions based on the Interacting Unit Check and their Porosity effects

on mixture performances were verified in this study. The usefulness of the gradation

guideless in explaining differences in pavements and mixtures performance could be used

for the evaluation of mixtures during mix design.

Results of the APA test data did not show any clear differences between the

Fine-graded and coarse graded mixtures and also between APA rut depth and HVS rut

depths.

Results of the IDT test parameters show no statistical significant difference

between the fine-graded and the coarse graded mixtures top-down cracking potential.

The fine-graded mixture used in this study appears to have a better rutting

resistance than an unmodified fine-graded mixture used in a previous study.

5.2 Recommendation

There is need to do more evaluation of the coarse-graded and fine-graded mixtures,

so that a larger statistical would be available for evaluation of the effects of aggregate

gradation on rutting resistance.

Evaluation of different grades of asphalt using the same aggregates source and

gradation should be investigated for the evaluation of instability rutting. Hard grades of

asphalts processed in such as to control the effects of ageing for cracking control are on

the market and provide an alternative for control of instability rutting whilst maintaining

its ductility.
























APPENDIX A

MIX DESIGNS


STATE OF FLORIDA DEPARTMENT OF TRANSPORTATION
STATEMENT OF SOURCE OF MATERIALS AND JOB MIX FORMULA FOR BITUMINOUS CONCRETE

SUBMIT TO THE STATE MATERIALS ENGINEER, CENTRAL BITUMINOUS LABORATORY. 5007 NORTHEAST 39TH AVENUE, GAINESVILLE, FLA 32609


V E Whitehurst & Sons


Address


2230 N W 73rd Place. Gainesville. FL 32653


(352) 573-3816 Fax No (352) 373-3314 E-mail


Howle Moseley


Type MI>


Coarse
SP-12 5


Intended Use of Mi>


Structural


Design Traffic Level


TYPE MATERIAL


D Gyrations @ N des

FDOT
CODE


PRODUCER


PIT NO


DATE SAMPLED


TM-561
1 #78 Stone 43 Junction City Mining, L L C GA-553 12 /22 /2004
TM-561
2 #89 Stone 51 Junction City Mining, L L C GA-553 12 / 22 / 2004
TM-561
3 W-10 Screenings 20 Junction City Mining, L L C GA-553 12 / 22 / 2004
Starvation
4 Local Sand V E Whitehurst & Sons Hill 12 / 22 / 2004

5 PG 67-22

6

PERCENTAGE BY WEIGHT TOTAL AGGREGATE PASSING SIEVES
Blend 29% 35% 28% 8% JOB MIX CONTROL RESTRICTED
Number 1 2 3 4 5 6 FORMULA POINTS ZONE
3/4" 190mm 100 100 100 100 100 100
i 1/2" 125mm 94 100 100 100 98 90 -100
N 3/8" 95mm 66 100 100 100 90 90
- No 4 475mm 18 37 98 100 54
O No 8 236mm 6 7 71 100 32 28 58 391 391
No 16 118mm 3 3 45 100 23 256 31 6
LU No 30 600pm 3 2 30 93 17 191 231
> No 50 300pm 2 2 20 49 11
L No 100 150pm 2 1 12 10 5
- No 200 75pm 15 10 76 36 45 2 10
o GSB 2809 2799 2770 2626 2779
The mix properties of the Job Mix Formula have been conditionally verified, pending successful final verification during production at the assigned plant, the
mix design is approved subject to F D O T specifications
No 200 reflects aggregate changes expected during production

LD 04-2543A (TL-D)


Director. State Materials Office


-r76c2@1


A-i Gradation and sources of aggregate for the coarse mixture


Contractor

Phone No


Submitted By











HOT MIX DESIGN DATA SHEET

LD 04-2543A (TL-D)


P ,,F P P P 1 1


4.5 2.485 2.589 4.0 14.6 73 44 1.0 87.0 97.2

4.6 2.488 2.583 3.7 14.6 75 4.5 1.0 87.3 97.5

4.8 2.495 2.577 3.2 14.5 78 4.7 1.0 87.7 98.0

_______________E_____7____][7][Z [_______________1 ____


TotalBinderContent 4.5 %







Lab. Density 155.1 Lbs/Ft 2485 Kg/mi
VMA 14.6 %
961

959
40 45
% Asphalt


Total Binder Content 4.5 %

Lab. Density 155.1 Lbs/Ft2 2485 Kg/m3


VMA 14.6 %


147 78
-- -14 --; ------ --- ----- -8 ----------

J46 _- 76
: - -v -- u- -
-- 145 ------ -- --5 -- ----- -

144 73

143 72
50 40 45 50 40 45
% Asphalt % Asphalt


FAA 46 % Mixing Temperature 310 F 154 C

%G @ Ndes 96.0 Compaction Temperature 310 F 154 C
Ar-Maz AdHere LOF 65-00 (M-0014)
NCAT Oven -0.23 Additives Antistrip 0.75 %
Calibration Factor
Be Added)/(-To Be Subtracted)


A-2. Hot mix design data sheet for coarse-graded mixture













STATE OF FLORIDA DEPARTMENT OF TRANSPORTATION
STATEMENT OF SOURCE OF MATERIALS AND JOB MIX FORMULA FOR BITUMINOUS CONCRETE

SUBMIT TO THE STATE MATERIALS ENGINEER. CENTRAL BITUMINOUS LABORATORY. 5007 NORTHEAST 39TH AVENUE. GAINESVILLE. FLA 32609


V E Whitehurst & Sons


Address


2230 N W 73rd Place Gainesville FL 32653


(352) 573-3816 Fax No (352) 373-3314 E-mail


Howle Moseley


Type Mix


Fine
SP-12 5


Intended Use of Mix


Structural


Design Traffic Level


TYPE MATERIAL


D Gyrations @ N des

FDOT
CODE


PRODUCER


PIT NO


DATE SAMPLED


TM-561
1 #78 Stone 43 Junction City Mining, L L C GA-553 12/22/2004
TM-561
2 #89 Stone 51 Junction City Mining, L L C GA-553 12/22 /2004
TM-561
3 W-10 Screenings 20 Junction City Mining, L L C GA-553 12/22 /2004
Starvation
4 Local Sand V E Whitehurst & Sons Hill 12 / 22 / 2004

5 PG 67-22

6

PERCENTAGE BY WEIGHT TOTAL AGGREGATE PASSING SIEVES
Blend 28% 12% 50% 10% JOB MIX CONTROL RESTRICTED
Number 1 2 3 4 5 6 FORMULA POINTS ZONE
3/4" 190mm 100 100 100 100 100 100
U 1/2" 125mm 94 100 100 100 98 90 -100
N 3/8" 95mm 66 100 100 100 90 90
- No 4 475mm 18 37 98 100 68
co No 8 236mm 6 7 71 100 48 28 58 391 391
No 16 1 18mm 3 3 45 100 34 256 31 6
L No 30 600pm 3 2 30 93 25 191 231
> No 50 300pm 2 2 20 49 16
L No 100 150pm 2 1 12 10 8
- No 200 75pm 15 10 76 36 49 2 10
co GsB 2809 2799 2 770 2626 2769
The mix properties of the Job Mix Formula have been conditionally verified, pending successful final verification during production at the assigned plant, the
mix design is approved subject to F D 0 T specifications
No 200 reflects aggregate changes expected during production

LD 04-2544A (TL-D)


Director, State Materials Office


-7766' /


A-3. Gradation and sources of aggregate for the fine-graded mixture


Contractor

Phone No


Submitted By











HOT MIX DESIGN DATA SHEET

LD 04-2544A (TL-D)


F, :,1 I. l .F P P If ,. ", ..' I1 ",,C ,.,11






__ I__ _I I I I I II
[I I ". II 1 1


1 46
% Asphalt


a6
sphalt


6
asphalt


Total Binder Content 4.6 %

Lab. Density 154.4 Lbs/Ft3 2475 Kg/mn

VMA 14.7 %


FAA 46 %

%Gmm@Nde 96.0

NCAT Oven -0.28
Calibration Factor
Be Added)(-To Be Subtracted)


Mixing Temperature 310 F 154 C

Compaction Temperature 310 F 154 C
Ar-Maz AdHere LOF 65-00 (M-0014)
Additives Antistrip 0.75 %


A-4. Hot mix data sheet for the fine-graded mixture


I



















APPENDIX B
RECOVERED VISCOSITIES


VISCOSITY OF RECOVERED ASPHALT SECTION: 3A LIFT:BOTTOM
MIXTURE TYPE:FINE SPECIMEN ID: A
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
Cp RPM % D/cm2 SEC-1 C
1275000 1.2 15.3 5712 0.71 60.2
1188095 4.2 49.9 16966 2.14 60.2
1132307 6.5 73.6 25024 3.6 60.2
1164285 4.2 48.9 16626 2.14 60.2
1233333 1.2 14.8 5032 0.71 60.2
AVERAGE 1198604



VISCOSITY OF RECOVERED ASPHALT SECTION: 3A LIFT:BOTTOM
MIXTURE TYPE:FINE SPECIMEN ID: A3
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
Cp RPM % D/cm2 SEC-1 C
800000 2.1 16.8 5202 0.4 60.2
752380 6.3 47.4 16116 1.43 60.2
715094 10.6 75.8 25772 2.21 60.2
738095 6.3 46.6 15810 1.43 60.2
785714 2.1 16.5 5610 0.41 60.2
AVERAGE 758256.6



VISCOSITY OF RECOVERED ASPHALT SECTION: 3A LIFT:BOTTOM
MIXTURE TYPE:FINE SPECIMEN ID: B
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
Cp RPM % D/cm2 SEC-1 C
2228571 0.7 15.6 5304 0.21 60.2
1983333 3 59.5 20230 1.02 60.2
1866666 3.6 67.2 22848 1.22 60.2
1866666 3 56 19040 1.02 60.2
1985714 0.7 13.9 4726 0.24 60.2
AVERAGE 1986190
B-1. Viscosity of recovered asphalt section 3A bottom lift












VISCOSITY OF RECOVERED ASPHALT SECTION: 3B LIFT:TOP
MIXTURE TYPE:FINE SPECIMEN ID: I
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
Cp RPM % D/cm2 SEC-1 C
2883333 0.6 17.3 5882 0.2 60.2
2727778 1.8 49.1 16694 0.61 60.2
2541935 3.1 78.8 26792 1.05 60.2
2622222 1.8 47.2 16048 0.61 60.2
2783333 0.6 16.7 5678 0.2 60.2
AVERAGE 2711720.2



VISCOSITY OF RECOVERED ASPHALT SECTION: 3B LIFT:TOP
MIXTURE TYPE:FINE SPECIMEN ID: 13
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
cP RPM % D/cm2 SEC-1 C
2750000 0.6 16.5 5610 0.2 60.2
2500000 2.1 52.5 17850 0.7 60.2
2375758 3.3 78.4 26656 1.12 60.2
2404762 2.1 50.5 17170 0.71 60.2
2600000 0.6 15.6 5304 0.2 60.2
AVERAGE 2526104



VISCOSITY OF RECOVERED ASPHALT SECTION: 3B LIFT:TOP
MIXTURE TYPE:FINE SPECIMEN ID:J
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
cP RPM % D/cm2 SEC-1 C
3260000 0.5 16.3 5542 0.17 60.2
2870588 1.7 48.8 16592 0.58 60.2
2765517 2.9 80.2 27268 0.99 60.2
2805882 1.7 47.7 16218 0.58 60.2
3100000 0.5 15.5 5270 0.17 60.2
AVERAGE 2960397.4


B-2. Viscosity of recovered asphalt section 3B top lift












VISCOSITY OF RECOVERED ASPHALT SECTION: 3C LIFT:TOP
MIXTURE TYPE:FINE SPECIMEN ID: E
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
Cp RPM % D/cm2 SEC-1 C
4500000 0.5 22.5 7650 0.17 60.1
4166667 1.2 50 17000 0.41 60.2
3915000 2 78.3 26622 0.68 60.2
4008333 1.2 48.1 16354 0.41 60.1
4220000 0.5 21.1 7174 0.17 60.2
AVERAGE 4162000



VISCOSITY OF RECOVERED ASPHALT SECTION: 3C LIFT:TOP
MIXTURE TYPE:FINE SPECIMEN ID: E3
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
cP RPM % D/cm2 SEC-1 C
4900000 0.3 14.7 4998 0.1 60.2
4433333 1.2 53.2 18088 0.41 60.2
4210000 2.2 84.2 28628 0.68 60.2
4375000 1.2 52.5 17850 0.41 60.2
4800000 0.3 14.4 4896 0.1 60.2
AVERAGE 4543666.6



VISCOSITY OF RECOVERED ASPHALT SECTION: 3C LIFT:TOP
MIXTURE TYPE:FINE SPECIMEN ID: F
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
cP RPM % D/cm2 SEC-1 C
3260000 0.5 16.3 5542 0.17 60.2
2870588 1.7 48.8 16592 0.58 60.2
2765517 2.9 80.2 27268 0.99 60.2
2805882 1.7 47.7 16218 0.58 60.2
3100000 0.5 15.5 5270 0.17 60.2
AVERAGE 2960397.4


B-3. Viscosity of recovered asphalt section 3C top lift












VISCOSITY OF RECOVERED ASPHALT SECTION: 5B LIFT:TOP
MIXTURE TYPE:COARSE SPECIMEN ID:K
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
Cp RPM % D/cm2 SEC-1 C
3060000 0.5 15.3 5202 0.17 60.2
2818750 1.6 45.1 15334 0.54 60.2
2680769 2.6 69.7 23698 0.88 60.2
2800000 1.6 44.8 15232 0.54 60.2
2980000 0.5 14.9 5066 0.17 60.2
AVERAGE 2867903.8



VISCOSITY OF RECOVERED ASPHALT SECTION:5B LIFT:TOP
MIXTURE TYPE:COARSE SPECIMEN ID: K3
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
cP RPM % D/cm2 SEC-1 C
2700000 0.6 16.2 5508 0.2 60.2
2479167 2.4 59.5 20230 0.82 60.2
2400000 3.4 81.6 27744 1.16 60.2
2462500 2.4 59.1 20094 0.82 60.2
2716667 0.6 16.3 5542 0.2 60.2
AVERAGE 2551666.8



VISCOSITY OF RECOVERED ASPHALT SECTION:5B LIFT:TOP
MIXTURE TYPE:COARSE SPECIMEN ID:L
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
cP RPM % D/cm2 SEC-1 C
2716667 0.6 16.3 5542 0.2 60.2
2519048 2.1 52.9 17986 0.71 60.2
2402941 3.4 81.7 27778 1.16 60.2
2461905 2.1 51.7 17578 0.71 60.2
2633333 0.6 15.8 5372 0.2 60.2
AVERAGE_


B-4. Viscosity of recovered asphalt section 5B top lift












VISCOSITY OF RECOVERED ASPHALT SECTION: 5A LIFT:TOP
MIXTURE TYPE:COARSE SPECIMEN ID:H
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
Cp RPM % D/cm2 SEC-1 C
725000 2.0 14.5 4930 0.68 60.2
651429 7.0 45.6 15504 2.38 60.2
602308 13.0 78.3 26622 4.42 60.2
620000 7.0 43.4 14756 2.38 60.2
680000 2.0 13.6 4624 0.68 60.2
AVERAGE 655747.4



VISCOSITY OF RECOVERED ASPHALT SECTION:5A LIFT:TOP
MIXTURE TYPE:COARSE SPECIMEN ID: G3
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
cP RPM % D/cm2 SEC-1 C
3080000 0.5 15.4 5236 0.17 60.2
2850000 1.6 45.6 15504 0.54 60.2
2703448 2.9 78.4 26656 0.99 60.2
2843750 1.6 45.5 15470 0.54 60.2
3060000 0.5 15.3 5202 0.17 60.2
AVERAGE 2907439.6


B-5. Viscosity of recovered asphalt section 5A top lift












VISCOSITY OF RECOVERED ASPHALT SECTION:5C LIFT:BOTTOM
MIXTURE TYPE:COARSE SPECIMEN ID:C3
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
cP RPM % D/cm2 SEC-1 oC
1454545 1.1 16 5440 0.37 60.3
1322857 3.5 46.3 15742 1.19 60.2
1252727 5.5 68.9 23426 1.87 60.3
1254285 3.5 4.9 14926 1.19 60.2
1381818 1.1 15.2 5168 0.37 60.2
AVERAGE 1333246.4



VISCOSITY OF RECOVERED ASPHALT SECTION:5C LIFT:BOTTOM
MIXTURE TYPE:COARSE SPECIMEN ID:D1
VISCOSITY SPEED TORQUE SHEAR STRESS SHEAR RATE TEMPERATURE
cP RPM % D/cm2 SEC-1 oC
1266666 1.2 15.2 5168 0.41 60.3
1167500 4 46.7 15878 1.36 60.2
1081428 7 75.7 25738 2.38 60.3
1125000 4 45 15300 1.36 60.3
1200000 1.2 14.4 4896 0.41 60.2
AVERAGE_


B-6. Viscosity of recovered asphalt section 5C bottom lift

















APPENDIX C
AREA CHANGE PARAMETER


3-2A Location


Fl :=


X := FI Y0:= (Fl)()


span := 0.10

0 := less (Xo,Y,span)

fi(x) := interp (X X,Y0,x)

tx:= min(Xo),nMin(X0) + .1..ima(xo)

.47
Fl fit(x) dx
Fl1:= fit(x x Fli = -734.581
60 1

span := 0.10

28k := less (X8k,Yskspan)
fitl(x) := interp (zsk, X8k Y8kk,

txl:= nmin(Xs8 ,nin(X8k) + .1.. max(Xgk)

f.47
Flf:= fitl(x) dx F f = -732.944
60


C-1. Area-change parameter section 2A


0 1 2 3
0 60 12.29 60 12.29
1 59.5 12.29 59.5 12.29
2 59.1 12.3 59.1 12.29
3 58.6 12.31 58.6 12.31
4 58.1 12.31 58.1 12.31
5 57.6 12.31 57.6 12.32
6 57.2 12.31 57.2 12.32
7 56.7 12.32 56.7 12.33
8 56.2 12.33 56.2 12.34
9 55.7 12.33 55.7 12.35


X8k:= Fl Y8k:= (F1)