A comparison of modified open-graded friction courses with standard open-graded friction courses [Apr. 1997]

GA
TlOO. R4
Si
C6
F7 iC!q7

DEPARTMENTAL RESEARCH GDOT RESEARCH PROJECT NO. 9110
FINAL REPORT .
GEORGIA DEPARTMENT OF TRANSPORTATION

I
r

AC OPEN-G
WITH

PARISON OF M DIFIED ED FRICTION OURSES
ANDARD OPEN-G ED CTION COURSES

I

OFFICE OF MATERIALS & RESEARCH RESEARCH & DEVELOPMENT BRANCH

Georgia Department of Transportation Office of Materials and Research
GDOT Research Project No. 9110 Final Report
Comparison of Standard Open-Graded Friction Courses with Modified Open-Graded
Friction Courses
by
Lanka Santha, P.E. Chief, Pavement Testing Unit
April, 1997
The contents of this report reflect the views of the authors, who are responsible for the facts and the accuracy of the data presented herein. The contents do not necessarily reflect the official view or policy of the Georgia Department of Transportation or the Federal Highway Administration. This report does not constitute a standard, specification, or regulation.

TECHNICAL REPORT STANDARD TITLE PAGE

1. Report No. FHWA-GA-97-9110

2. Government Accession No.

4. Title and Subtitle A Comparison of Modified Open-Graded Friction Courses to Standard Open-Graded Friction Course

7. Author Lanka Santha, P .E. Chief, Pavement Testing Unit
9. Performing Organization Name and Address Office ofMaterials and Research Georgia Department of Transportation 15 Kennedy Drive Forest Park, Georgia 30050

3. Recipient's Catalog No. 5. Report Date
April1997 6. Performing Organization Code 8. Performing Organ. Report No.:
9110 10. Work Unit No.
11. Contract or Grant No.

12. Sponsoring Agency Name and Address Georgia Department of Transportation No. 2 Capitol Square Atlanta, Georgia 30334

13. Type of Report & Period Covered Final1991-1995
14. Sponsoring Agency Code

15. Supplementary Notes Prepared in cooperation with the U.S. Department of Transportation Federal Highway Administration.

16. Abstract A significant percentage of open-graded asphaltic concrete surfaces in Georgia have been damaged by climatic and traffic
related stresses. Open-graded friction course (OGFC) is designed to eliminate the collection of water on pavement surfaces and, at the same time, improve friction properties when surface water is present. The primary objective of this project was to determine if improvements can be made to conventional GDOT open-graded mixes by comparing the performance of these mixes to that of open-graded mixes modified with polymers and other additives.
Six test sections were constructed on 1-75 just south of Atlanta. Each test section was 0.5 mile in length and consisted of conventional GDOT surface mix topped with one of the following mixes: (1) coarse OGFC, (2) coarse OGFC with 16% crumb rubber, (3) coarse OGFC with mineral fiber, (4) coarse OGFC with cellulose fibers, (5) coarse OGFC with styrene butadiene (SB) polymer, and (6) coarse OGFC with SB and cellulose fiber. The conventional GDOT fine OGFC was the control mix.
The test sections were closely monitored for 3-112 years. All six test sections and the control section had acceptable friction levels and smoothness levels. Test sections (1) and (2) experienced high levels of cracking compared to the other test sections and the control, while sections (4) and (5) showed the lowest levels of cracking. Rutting in the control section was significantly higher than rutting in the test sections, and the rutting in the test sections was similar from section to section.
All six test sections showed better permeability than the control section during the study period. In all types of OGFC, permeability decreases sharply over time due to clogging. Clogging of openings was significantly higher in between wheelpaths than in the wheelpaths. Continuous monitoring of the test sections may provide better comparisons between the sections.

17. Key Words Open-graded friction course, polymers, crumb rubber, fibers, permeability

18. Distribution Statement No Restrictions

19. Security Classif. (of this report) Unclassified
Form DOT 1700.7 (8-69)

20. Security Classif. (of this page) 21. No. ofPages Unclassified 57

22. Price

TABLE OF CONTENTS
Section
Executive Summary ........................................................................................................... iii
List of Tables ..................................................................................................................... iv
List ofFigures ......................................................................................................................v
1. INTRODUCTION ............................................................................................................... 1 1.1. Objectives .................................................................................................................... 1
2. BACKGROUND .................................................................................................................2 2.1. Existing Pavement .......................................................................................................2 2.2. Current OGFC Design .................................................................................................2 2.3. Overlay Test Sections ..................................................................................................3 2.3.1. Coarse OGFC (D) ............................................................................................ 5 2.3.2. Coarse OGFC with 16% Crumb Rubber (D16R) ...........................................5 2.3.3. Coarse OGFC with Fibers (DM and DC) ........................................................5 2.3.4. Coarse OGFC with SB Polymer (DP) ............................................................. 6 2.3.5. Coarse OGFC with SB and Cellulose Fibers (DCP) .......................................6
3. LABORATORY MIX DESIGNS..............................................................................................6 3.1. Control Mix (d) .......................................................................................................... 10 3.2. Coarse OGFC (D) ...................................................................................................... 10 3.3. Coarse OGFC with 16% Crumb Rubber (D16R) ...................................................... 10 3.4. Coarse OGFC with Fibers (DM and DC) ................................................................. .12 3.5. Coarse OGFC with SB Polymer (DP) ....................................................................... 12 3.6. Coarse OGFC with SB and Cellulose Fibers (DCP) ................................................. 13
4. MIX PRODUCTION ......................................................................................................... 13 4.1. Plant Production......................................................................................................... 13 4.2. Test Section Construction .......................................................................................... 15 4.2.1. DM ................................................................................................................. 16 4.2.2. DC .................................................................................................................. 16 4.2.3. DCP ................................................................................................................ 16 4.2.4. DP .................................................................................................................. 16 4.2.5. D16R .............................................................................................................. 17 4.3. In-place Permeability ................................................................................................. 17
5. TEST RESULTS FOR CONSTRUCTION SAMPLES .......................................................... 18

6. POST-CONSTRUCTION EVALUATIONS ....................................................................22 6.1. Friction Testing ..........................................................................................................22 6.2. Smoothness Testing ...................................................................................................23 6.3. Visual Distress Survey .............................................................................................. .23 6.4. Rutting Measurements ...............................................................................................24 6.5. Permeability Testing ..................................................................................................25 6.6. Falling Weight Deflectometer (FWD) Testing ..........................................................30 6.7. Laboratory Evaluation of Core Samples ....................................................................31
7. CONCLUSIONS................................................................................................................31 8. BIBLIOGRAPHY 9. APPENDICES
11

EXECUTIVE SUMMARY
An open-graded friction course (OGFC) is a bituminous mixture composed primarily of uniformly graded aggregate designed to have an in-place air void percentage of 15 to 20%. The main function of OGFC is to improve wet weather driving conditions by allowing water to drain away from the roadway surface. This improved surface drainage reduces hydroplaning, reduces splash and spray behind vehicles, and improves pavement friction and surface reflectivity.
The primary objective of this research project was to determine if improvements could be made to conventional GDOT OGFCs by comparing the performance of these mixes to that of open-graded mixes modified with polymers and other additives. Mixes that were designed and evaluated in this project include (1) coarse OGFC, (2) coarse OGFC with 16% crumb rubber, (3) coarse OGFC with mineral fillers, (4) coarse OGFC with cellulose fibers, (5) coarse OGFC with styrene butadiene (SB) polymer, and (6) coarse OGFC with SB and cellulose fibers. GDOT's conventional fine OGFC was the control mix for this study.
The project was constructed on I-75 just south of Atlanta. This section ofl-75 has an ADT of 47,000, with 21% truck traffic. The test sections were placed in the outside northbound lane between mileposts 221.1 and 224.1. The three-mile section was composed of six subsections, corresponding to the six mixes listed above, each 0.5 mile long. The existing pavement was milled to a depth of2-118 in. and inlaid with 1-3/8 in. of recycled dense-graded mix. This layer was then overlaid with% inch of OGFC. None of the mixes presented any significant problems during placement.
The performance of the test sections was evaluated over a period of3-1/2 years. Physical testing was conducted on the roadway, as well as laboratory testing of core samples. The physical testing included friction testing, smoothness testing, falling weight deflectometer (FWD) testing, a visual distress survey, rut measurements, and permeability testing.
The friction levels of all six test sections were acceptable during the evaluation period. No significant difference was noted between the friction values of the control section (fine OGFC) and the test sections after 3-1/2 years of service. Also, the smoothness levels of the six test sections were acceptable during the evaluation period. The test sections containing coarse OGFC and coarse OGFC with 16% crumb rubber experienced high levels of cracking compared with the other sections and the control section. The sections containing coarse OGFC with cellulose fibers and coarse OGFC with styrene butadiene (SB) polymer had the lowest levels of cracking. Rutting in the control section was significantly higher than rutting in the test sections. All of the test sections showed similar degrees of rutting after 3-112 years of service.
All six test sections showed better permeability than the control section. In all types of OGFC, permeability decreases sharply over time due to clogging. Clogging of openings is significantly higher between wheelpaths than in the wheelpaths. Continuous monitoring of the test sections may provide better comparisons between the sections. Since this research study began, GDOT has begun using coarse OGFC modified with SB polymer for overlays.
111

LIST OF TABLES
Table No. and Title 1. Laboratory Test Results for Research Mixes ..............................................................................9 2. Laboratory Tests on Virgin AC-20s, Crumb Rubber Blended AC-20s, and Thin-film
Oven Residue ....................................................................................................................... 11 3. Laboratory Tests on Virgin AC-20s, StyrelfBlended AC-20s, and Thin-film
Oven Residue ....................................................................................................................... 12 4. Laboratory Test Results for Truck Samples of Research Mixes .............................................. 19 5. Laboratory Test Results for Roadway Core Samples of Research and Control Mixes ............21 6. Average Friction Test Results in Test Sections ........................................................................22 7. Average Smoothness Values in Test Sections ..........................................................................23 8. Results of Distress Survey on Test Sections .............................................................................24 9. Average Rutting in Test Sections .............................................................................................25 10. Permeability Results in Test Sections by Wheelpath ............................................................ .26 11. Average Permeability Results in Test Sections ......................................................................26 12. Average Normalized Deflections in Test Sections .................................................................30
lV

LIST OF FIGURES
Figure No. and Title 1. Test Section Layout ....................................................................................................................4 2. Schellenberg Drainage Test ........................................................................................................8 3. Permeabilities of Open-Graded Friction Course Mixes (1993) ............................................... .28 4. Permeabilities of Open-Graded Friction Course Mixes (1996) ................................................29
v

COMPARISON OF STANDARD OPEN-GRADED FRICTION COURSES WITH MODIFIED OPEN-GRADED FRICTION COURSES

1. INTRODUCTION An open-graded friction course (OGFC) is a bituminous mixture composed primarily ofuni-
formly graded aggregate designed to have an in-place air voids percentage of 15 to 20%. The main function of OGFC is to improve wet weather driving conditions by allowing water to drain away from the roadway surface. This improved surface drainage reduces hydroplaning and spray behind vehicles, and it improves pavement friction and surface reflectivity.
This research project was initiated to determine if improvements made to GDOT' s conventional OGFC would alleviate the negative properties of this mix type and still ensure the superior wet-weather performance of the mix.

1.1. Objectives The primary objective of this research project is to determine the improvements made to the existing GDOT open-graded mixes by comparing the performance of these mixes to that of opengraded mixes modified with polymers and other additives. Mixes that were designed and evaluated in this project are shown below.

Mix Code d D
Dl6R OM DC DP DCP

Mix Description Standard OGFC Coarse OGFC Coarse OGFC with 16% crumb rubber Coarse OGFC with mineral fibers Coarse OGFC with cellulose fibers Coarse OGFC with styrene butadiene (SB) polymer Coarse OGFC with SB and cellulose fibers

These mixes were compared to one another and to a control section of standard OGFC. In each design and corresponding test section, performance parameters such as raveling, cracking, smoothness, friction performance, and permeability were assessed.
2. BACKGROUND 2.1. Existing Pavement This project was constructed on I-75 just south of Atlanta. This section ofl-75 handles high
traffic levels nearly year round arising from (1) transit to and from Florida and (2) local rush hour commuter traffic. The average daily traffic (ADT) level in the test section is 47,000, with 21% truck traffic. The test sections were placed in the outside northbound lane from MP 221.1 to 224.1. The three-mile test section was comprised of six subsections that were each 0.5 mile long.
The existing roadway consisted of approximately 5.5 in. of asphaltic concrete over 9 in. of portland cement concrete (PCC) jointed at 30' intervals. Most of the existing pavement had an OGFC surface. The existing pavement was extensively cracked both transversely and longitudinally. Transverse reflective cracks extended the full width of the lane and into the shoulder. Longitudinal shoulder joint cracking was extensive also. The Maysmeter roughness index was 46 in./mile within the test section limits, and rutting depths averaged 114 in. per wheelpath within the section limits.
2.2. Current OGFC Design In order to alleviate raveling problems associated with the standard OGFC, fog coats were applied to the OGFC in certain projects to help hold the aggregate in place. Applying the fog coat to the surface fills voids and reduces the permeability of the pavement. In response to these
2

problems, GDOT changed its specifications in the summer of 1992 to require a larger nominal size aggregate for better permeability. The specifications also require the addition of a thermoplastic modifier and mineral fibers to the mix to (1) help hold the aggregate in place and reduce raveling, and (2) reinforce film thickness and reduce or eliminate draindown. The gradation requirements of the standard OGFC and the coarser OGFC used in the project are shown below.

Gradation % Passing 3/4" Sieve % Passing 112" Sieve % Passing 3/8" Sieve %Passing No.4 Sieve % Passing No. 8 Sieve % Passing No. 200 Sieve

Standard OGFC
100 85-100 20-40 5-10
2-4

Coarse OGFC 100
90-100 65-85 15-25 5-10 2-4

All of the mixes used in the test section had gradations corresponding to the coarse OGFC. Since this project was conducted under a supplemental agreement on a previously awarded project in which the standard OGFC was being placed, all test sections were compared with this finer mix. Although GDOT use of this mix is now limited, the mix is useful for comparison, since it was the standard GDOT OGFC for many years.

2.3. Overlay Test Sections In the test section, 2-1/8 in. ofthe existing pavement was milled, and 1-3/8 in. of a densegraded surface mix ("E" mix) was then inlaid. The 3/4 in. of OGFC was placed over the "E" mix. A total of six different overlays were constructed as shown in Figure 1 on page 4. The control section, which was more finely graded, was placed at 5/8 in., and the underlying "E" mix was placed at 1-112 in. to provide the same total thickness as the test sections. Brief descriptions of the mixes used in the test sections follow Figure 1.

3

224.1
~

l l 0T.5ypMicialel

221.1
r

d D16R DP DCP DC OM

D

d

-1>

d: Fine OGFC Control

D: Coarse OGFC

D16R: Coarse OGFC/ Crumb Rubber

DC: Coarse OGFC w/ C~iiulose Fibers

DP: Coarse OGFC w/SB Polymer

OM: Coarse OGFC w/ Mineral Fiber

DCP: Coarse OGFC w/SB & Cellulose Fibers

Figure 1 - Test Section Layout

2.3.1. Coarse OGFC (D) This coarse OGFC, as well as that in other subsections, consisted of the coarsergradation mentioned previously. The purpose of using the coarser gradation was to provide more drainage capability than the standard "d" mix. This test section differed from the "d" mix in gradation and layer thickness only.
2.3.2. Coarse OGFC with 16% Crumb Rubber (D16R) In an effort to reduce the amount of solid waste, namely tires, generated each year in the U.S., Congress issued mandates in the Intermodal Surface Transportation Efficiency Act (ISTEA) of 1991 that would withhold federal funding from states that did not incorporate a certain percentage of crumb tire rubber into their asphalt mixes. Because this legislation was pending at the time this research study was proposed, a test section containing such crumb rubber was placed. This legislation is no longer in effect. The crumb rubber thickens the asphalt cement (AC) and reduces draindown, thereby allowing more AC to be added to the mix and improving aggregate retention. Tire rubber also contains hydrocarbons and carbon black, which act as antioxidants. The rubber for this section was -80 sieve, and it is marketed as "Ultrafine GR-80" from Rouse Rubber Industries of Vicksburg, MS. The rubber was used at a dosage rate of 16% by weight of total AC.
2.3.3. Coarse OGFC with Fibers (DM and DC) Mineral and cellulose fibers are added to OGFC mixes primarily to reinforce and stiffen the asphalt cement film, thus reducing draindown potential. Test sections utilizing cellulose and mineral fibers, respectively, were placed to compare the performance of the two types of fiber.
5

The mix designated as DC contained stabilized cellulose fibers, CF 35100, obtained from Custom Fibers International of Valencia, CA. These fibers were added at a rate of0.3% of the total mix weight. The mix designated as DM was stabilized with Mineral Fibers obtained from Fiberand Corp. of South Miami, FL, and these fibers were marketed as Inorphil. The Inorphil fibers were added at a rate of0.5% ofthe total mix.
2.3.4. Coarse OGFC with SB Polymer (DP) This test section contained AC modified with 3% styrene butadiene (SB) polymer. The SB, known as Styrelf, was obtained from Elf Aquitaine Asphalt of Raleigh, NC. Styrelf is combined with neat AC to produce a viscous blend that will resist draindown and, at the same time, improve the cold-weather performance of the mix.
2.3.5. Coarse OGFC with SB and Cellulose Fibers (DCP) This test section is similar to the new Georgia OGFC specifications (see Appendices), except that cellulose fibers were used in lieu of mineral fibers. This test section contains SB polymer to aid aggregate retention via the elasticity of SB, and cellulose fibers to improve cement binding.
3. LABORATORY MIX DESIGNS The selection of optimum AC contents for each mix consisted of three different steps that are
described in test method GDT-114, which is shown in Appendix A. The most recent specifications for the fiber and polymer additives are shown in Appendix B. The base AC for all test mixes and the control mix was Shell AC-20 Special, designated AC-20s. AC-20s is defined as viscosity grade AC-20 having a penetration at 77 F not exceeding 80 dmm, and an initial viscosity at 77 F not less than 2.3 million poises per AASHTO T-202. Both hydrated lime and
6

liquid anti-strip additive were required for each mix to prevent stripping. All aggregates used in this project were from the Florida Rock quarry at Mt. View, Georgia. This quarry produces granite-gneiss with an L.A. abrasion value of 42%.
Table 1 on page 9 summarizes all pertinent design information for each research mix. The optimum AC contents listed are obtained from GDT-114 and engineering judgment as well. In addition to the tests described in GDT-114, two additional tests were performed, the Schellenberg draindown test and the Cantabro wear test. These tests were performed for information only.
In the Schellenberg test, approximately 1 kg of asphalt mix is placed into a preweighed glass beaker. The beaker is then placed in a 350 F oven for two hours. Next, the mixture is poured from the beaker, at which time the beaker is reweighed to determine the amount of binder that drained out of the mix. The binder that drained out of the mix will remain in the beaker when the mix is poured out. As a rule, if the amount of drainage is less than 0.3% of the original weight at optimum AC content, the design is considered acceptable (see Figure 2 on page 8). This test was used to compare the draindown resistance of the mixes.
Results of Schellenberg testing on laboratory mixes at 6.5% AC are shown in Table 1 on page 9. These results indicate that the coarse mix and the coarse polymer mix may experience draindown problems.
7

800 ml Beaker

Approx.
1000 9 Asphalt Mixture

I II\\

g.., . ~

r

... r

r ~r

... r 4

r

.. .... .

r ~ r

<0.2% Good

, "'w ..

.... , 4 ,

.......

"

~
""

'.::~",'

......... ,r " q

,. r ",. ,,,

0.2-0.3% Still Acceptable

;~~~" :~r.-,4 4tl. . ....~..~ . r:..\..,..-.~ ~
>0.~%
6ad

Figure 2. - Schellenberg Drainage Test

8

The Cantabro wear test is a measure of how effectively asphalt cement holds aggregate particles in place. Marshall specimens of a particular mix are placed in a standard L.A. abrasion machine and undergo 300 revolutions. However, the steel spheres that are used in L.A. abrasion testing are not used in this test. The percent wear is determined from the known weight of the specimen before and after testing. This test provides a good indication of the AC binding ability in a particular design as well as the aggregate abrasion resistance. Generally, where a test loss of30% or less occurs, the design is considered acceptable. None of the research mixes had losses approaching 30%. All testing was performed using an AC content of 6.5% The Cantabro test results are shown in Table I below.

Sample Type
Sieve Size 3/4" 1/2" 3/8" No.4 No.8
No. 200
o/o AC
Cantabro (%Wear) Drainage (%Loss) Film Thickness (u)

Table 1: Laboratory Test Results for Research Mixes.

Coarse OGFC
(D)
100 99 75 18 8 2
6.0

D+16% Rubber (D16R)

D +Mineral

D+

Fibers

Cellulose

(DM)

Fibers (DC)

D+SB Polymer
(DP)

Total Percent Aggregate Passing by Weight

100

100

100

100

99

99

99

99

75

75

75

75

18

18

18

18

8

8

8

8

2

2

2

2

Percent Asphalt Cement of Total Mix

6.6

6.3

6.4

6.2

Miscellaneous Test Data

13.5 0.37 34.07

8.6 0.05 36.90

5.7 0.06 35.92

5.8 0.06 36.54

8.6 0.34 35.30

DC+SB Polymer (DCP)
100 99 75 18 8 2
6.4
8.2 0.04 36.54

9

The rubber selected for use in this project was -80 sieve rubber, which is marketed as "Ultrafine GF-80" from Rouse Rubber Industries, Inc. of Vicksburg, Mississippi. When selecting rubber as an AC additive, GDOT inferred that a gradation of this rubber with high surface area would more easily "digest" into the neat AC and produce a more homogeneous mix.
The D 16R was designed to contain 16% rubber by weight of total AC. The crumb rubber was blended into the virgin AC by high-speed mixing. Mixing in the 340 op AC continued for 35 minutes, the duration recommended by the rubber manufacturer. Results of laboratory tests performed on the virgin AC, the AR blend, and thin-film oven (TFO) residue are shown in Table 2.

Table 2: Results of Laboratory Tests on Virgin AC-20s, Crumb Rubber Blended AC-20s, and Thin-film Oven Residue.

Test
Viscosity@ 140 F (poises) Viscosity@ 275 F (eSt) Penetration@ 77 F, 100 g, 5 sec (dmm) Softening point, ring and ball (F) Elastic recovery from 10 em, 5 em/min., 77 F (%) Force ductility@ 10-cm ext., 5 em/min. 77 F (lb.) Thin-film Oven Residue Viscosity@ 140 F (poises) Force ductility@ 10-cm ext., 5 em/min., 39 F (lb.) Ductility, 5 em/min., 39 F (em)

AC-20s
1793 352 68 122 18 0.04
4569 3.87 12

AC-20s+ 16% Rubber 17,648 10,195 40 138 57 0.43
70,794 15.3 20

The data in Table 2 indicates that the AR blend was much more viscous than the base or virgin AC and also showed greater elasticity. This greater elasticity may contribute to better aggregate retention, and the ductility testing indicates good low-temperature performance. Asphalt rubber may be more crack resistant at low temperatures than conventional mixes, as shown by the greater TFO ductility.

11

Due to the increased viscosity of the asphalt rubber, the mixing and compaction temperatures in GDT-114 were increased to 325 F. Draindown tests and VMA determinations indicate that mixes containing rubber can tolerate much higher total binder contents, possibly resulting in greater film thickness and better aggregate retention.

3.4. Coarse OGFC with Fibers (DM and DC) For each mix type, fibers were added to the hot aggregate and lime, then dry mixed for a few seconds to ensure good dispersion before the addition of AC. Because of the fibers' affinity for AC, temperatures were increased to 275 F to enhance aggregate coating. These mixes demonstrated the best aggregate retention values based on Cantabro testing as shown in Table 1.

3.5. Coarse OGFC with SB Polymer (DP) The Styrelf and AC-20s were preblended by the polymer supplier. Laboratory tests on the virgin AC and Styrelfmodified binder are shown in Table 3.

Table 3: Laboratory Tests on Virgin AC-20s, Styrelf Blended AC-20s, and Thin-film Oven Residue.

Test
Viscosity@ 140 op (poises) Viscosity@ 275 op (eSt) Penetration@ 77 F, 100 g, 5 sec (dmm) Softening point, ring and ball (F) Elastic recovery from 10 em, 5 em/min., 77 op (%) Force ductility@ 10-cm ext., 5 em/min., 77 F (lb.) Thin-film Oven Residue Viscosity@ 140 op (poises) Force ductility@ 10-cm ext., 5 em/min., 39 F (lb.) Ductility, 5 em/min., 39 F (em)

AC-20s 1793 352 68 122 18 0.04
4569 3.87 12

AC-20s with Styrelf 9,739 1,095 55 138 77 0.18
29,665 11.5 31

12

The data in Table 3 indicates that the SB blend was much more viscous than the base AC, and it also exhibited much greater elasticity. The ability of the SB blend to recover from elongation was almost four times better, as shown in the elastic recovery test. Thin-film aging had very similar effects on the base AC and SB blend, but the SB remained almost three times more ductile at 39 F. This may make the SB blend more resistant to cold weather cracking.
Due to the greater viscosity of the SB blend, temperatures in the GDT-114 tests were increased to 300 F. Results in Table 1 show that the DP mix is more susceptible to drainage (Schellenberg testing) at higher temperatures compared to the fiber stabilized mixes.
3.6. Coarse OGFC with SB and Cellulose Fibers (DCP) This mix is designed to combine the mix enhancement properties of both the DC and DP mixes. Again, temperatures were increased to 300 F. Preliminary results, other than binder tests, indicate that the DCP mix should perform similarly to the DC mix.
4. MIX PRODUCTION 4.1. Plant Production C.W. Matthews Company of Marietta, Georgia was the prime contractor on this resurfacing
project. Asphalt mix was produced from their Astec "double barrel" drum plant at Mt. View, Georgia. This plant is a drum-in-a-drum, counter flow drum mixing plant. A silo was used to store the mix and distribute material into the haul trucks. To prevent AC drainage from the OGFC, a maximum of 50 tons of mix was stored at any given time. Mix was produced at a rate of 150 tons/hour, while standard dense-graded mixes are produced at 200 tons/hour.
Two modifications were made to the asphalt plant in order to produce the research mixes. Modifications were made so that the fibers and asphalt rubber could be introduced into the plant.
13

Cellulose and mineral fibers were blown into the dryer drum at the same location as the hydrated lime. Both fibers were blown in by a fiber blowing machine (Krendle Model 8000 series). This machine fluffs the fibers, crowds the material to a consistent density by use of a series of paddles and augers, and then augers the fibers at a specified rate into an air jet that blows them into the drum. The machine was calibrated by attaching the fiber supply line to a preweighed sealed container and allowing the machine to operate for one minute. At the end of one minute, the container was weighed to see if the desired amount had been delivered to the container. This procedure was repeated, and adjustments were made until a feed rate of 15 lb./minute was achieved for the cellulose fibers and 25 lb./minute for the mineral fibers.
To determine where the fibers would be introduced into the drum, either at the lime injection point or the AC injection point, fibers were introduced into trial mixes at both points. Fibers were introduced into 100 tons of "E" mix at the lime injection point as well as at the AC injection point. Samples ofbaghouse fines were taken during each injection period to determine if fibers may be caught into the plant's airstream. No appreciable amount of fibers was detected in any ofthe samples. The tensile strength results for mixes produced using both types of injection provided no definite conclusions about the proper injection point for the fibers. Tensile strength results were slightly higher for the mix where fiber was injected with the lime, and GDOT decided that all fiber mix would be produced with the fibers being introduced with the hydrated lime. This injection point allowed dry mixing of the fibers with the aggregate before the AC was introduced.
The asphalt rubber was mixed in a batch type blending unit leased from Novophalt America, Inc. The rubber and AC were mixed by mechanical agitation via shaft driven paddles that "fold" the rubber into the hot AC. Approximately 8400 lb. of AC and 1600 lb. of rubber were blended
14

at 350 op for about one hour. The rubber was supplied in 50-lb. bags, and thus 32 bags were added to the hot AC in the Novophalt unit for each batch.
The AR was gravity fed from the Novophalt unit into the Brodie pump at the plant's AC storage tank. The screen in the pump was removed to prevent possible straining of the rubber out of the blend and to avoid clogging the screen. The rate at which the AR was pumped into the drum mixer was set by computer, based on the specific gravity of the blend. Since the specific gravity of the AR is slightly higher than that of neat AC (1.056 vs. 1.031 ), the pump was adjusted accordingly to achieve the desired AR content in the job mix formula (JMF).
The Styrelf modified AC was delivered to the asphalt plant by tanker truck and pumped into the plant. Numerous samples of each test mix were taken from trucks at the plant for laboratory analysis. Cores were taken from the roadway also. Laboratory tests run on the mixes included (1) Abson extractions for gradation, binder content, viscosity (140 F), and penetration, and (2) Marshall specimens for density, maximum gravity, and air voids, and for Cantabro wear testing.
4.2. Test Section Construction The existing pavement was milled to a depth of2-1/8 in. and inlayed with 1-3/8 in. of recycled "E" mix. The "E" mix was tacked with AC-20s at approximately 0.068 gaL/sq. yd. All test mixes were placed with a Blaw Knox PF-200 paver using dual inboard skis. Compaction and finishing were accomplished with a Dynapac CC-50 and CC-42, respectively. As with all OGFC mixes, pneumatic rollers were not used, and vibrator rollers were used in the static mode only. Both the control (d) and the regular coarse (D) mixes were placed with no problems. None of the mixes presented any significant problems during laydown and compaction. The following paragraphs contain observations made during placement of the mixes containing additives:
15

4.2.1. Coarse OGFC with Mineral Fibers (DM) During laydown, this mix behaved as if it were too cold, even though temperatures fell within JMf limits. During the lab design process, GDOT observed that the mix could probably be placed at a higher temperature without the risk of draindown. The JMf temperature was set at 275 f, and truck temperatures taken at the plant averaged 278 f.
4.2.2. Coarse OGFC with Cellulose Fibers (DC) This mix, which contained cellulose fibers, had a dull, flat appearance. Problems were encountered when the mix stuck to the roller drums and was picked up. Keeping the rollers back to allow the mix to cool before rolling seemed to help this problem, but adding soap or releasing agent to the drum watering system is likely to be the best alternative. Mix temperatures at the plant averaged 289 f, and this range fell within the tolerance of the JMf temperature, 275 f.
4.2.3. Coarse OGFC with SB Polymer and Cellulose Fibers (DCP) This mix contained SB polymer and cellulose fibers, and it was difficult to lay so that it matched the adjoining, previously placed lane. The thickness depended on paver speed, and any increase or decrease in speed would cause the mat to get thinner and thicker respectively. The JMf was set at 300 f, and the temperature at the plant averaged 302 f.
4.2.4. Coarse OGFC with SB Polymer (DP) This mix contained Styrelfpolymer, and it was the most difficult mix to lay. Paver speed changes greatly affected the thickness of the mix, as with the DCP mix. Cores taken from this section had thickness variations of up to 0.4 in. AC draindown apparently occurred in four loads, indicating that the JMf placement temperature (300 f) was probably set too high. Mix
16

temperature averaged 312 op at the plant. The difficulty in laying this mix and the DCP mix is reflected in the relatively high Maysmeter values of these mixes compared with those of the other mixes.
4.2.5. Coarse OGFC with 16% Crumb Rubber (Dl6R) This mix, which contained 16% crumb rubber, produced only minor laying or compaction problems. Some pulling of the mat occurred for approximately the first 75 feet but disappeared shortly afterward. The paver screed may have been cold, since this mix was the first which was placed that day. The JMF placement temperature was set at 325 F, and the mix at the plant had an average temperature of333 F.
4.3 In-place Permeability Each test mix was tested for in-place permeability using a falling head permeameter. This device is placed directly on the pavement and delivers a known amount of water to a confined area. The apparatus used to measure permeability, referred to as a permeameter, consists of a circular base plate, a grease gun, and a cylindrical plastic water container. The base plate has two rubber 0-rings fitted at the edge of the plate, and the distance between the two 0-rings is 1/2 in. When the permeameter is in use, the 0-rings are seated on the pavement, and grease is pumped between the 0-rings, creating an impermeable seal. The plastic cylinder is then filled with water and allowed to flow through the base-plate into the pavement. The time required for the water level to drop between two known points is recorded. The volume of water between these two points is 1600 ml. The following equation determines the permeability obtained using this modified falling
17

head permeameter: K =(aLlAT) Log10 (H/H1), where K =Permeability in centimeters per second (1 em/sec= 2835 ft./day), a = Cross sectional area of stand pipe in square centimeters, L = Height of sample in centimeters, A= Surface area of water flow in square centimeters, T =Time in seconds to drop water level from H0 to H,,
Ha = Head at the beginning of test in centimeters,
H, =Head at the end oftest in centimeters.
5. TEST RESULTS FOR CONSTRUCTION SAMPLES The samples listed in Table 4 (p. 19) were taken from haul trucks at the plant and placed in one-
gallon cans. Extractions were performed to determine gradation and AC content using biodegradable solvents. To ensure an accurate determination of AC content, the entire contents of each can was extracted. The inside of the can was thoroughly washed out with solvent to remove any cement that may have drained out of the mix. AC contents obtained from the extractions of the D 16R mix averaged 1.2% less than the AC content in the JMF. This discrepancy occurs because rubber particles, which are insoluble in the solvent, remain as aggregate and thus increase the washed aggregate weight. A correction factor was derived to account for this discrepancy by testing samples of the mix made with a known AC content.
18

Sample Type
Sieve Size 3/4" 1/2" 3/8" No.4 No.8
No. 200

JMF
100 99 75 18 8 2

%Binder
Max. Gravity
Density (pet)
%Voids
Viscosity (Ps)
Penetra. (dmm) Cantabro (%Wear)

EXTR/JMF
-----
--
--

Table 4: Laboratory Test Results for Truck Samples of Research Mixes.

Coarse OGFC
(D)
100 98.3 70.0 21.0 8.7 3.6
5.85/6.00 2.484 136.1 12.2 5,890
39
10.3

D +Mineral Fibers (DM)
100 98.9 76.2 23.9 9.0 3.1
6.22/6.30 2.445 135.2 11.4 5,757

D+ Cellulose Fibers (DC)

DC+SB Polymer (DCP)

D+SB Polymer
(DP)

Total Percent Aggregate Passing by Weight

100

100

100

96.7

97.0

99.1

64.0

68.6

69.9

19.0

19.1

23.1

7.7

7.8

8.4

2.8

2.4

3.1

Miscellaneous Test Data

6.16/6.40

6.14/6.40

6.25/6.20

2.429

2.424

2.476

134.1

134.7

132.7

11.5

10.9

14.1

4,298

24,308

28,832

41

44

37

34

8.1

14.7

7.0

15.9

D+l6% Rubber (Dl6R)
100 96.3 60.3 15.7 7.4 2.6
6.41/6.60 2.451 134.5 12.0 4,722
46
7.6

JMF (d)
--
100 94 27 8 2
EXTR/JMF
---
--
--
--
--

Standard OGFC(d)
100 98.6 82 25.4 8.4 2.9
6.16/6.20 2.469 133.7 13.3 4,000
45
8.0

The results in Table 4 indicate that the AC recovered from the polymer mixes was much stiffer than the AC recovered from the other mixes, as reflected in the 140 op viscosity test. The viscosity of the AC recovered from the D16R mix is similar to that recovered from the mixes without modifying agents. This similarity reinforces GDOT's assumption that when this mix was extracted, the insoluble rubber particles were left behind with the aggregate. The oils which were absorbed by the rubber may have dissolved, reverting back to the original unmodified AC.
Table 5 on page 21 shows the results of tests performed on roadway core samples and the permeability of test sections just after construction. Viscosities obtained from Abson recoveries of the D16R cores were lower than expected. The viscosity of the D16R exceeded the viscosities of the fiber stabilized mixes, but it was still about halfthat of the recovered DP and DCP. This viscosity difference also shows that the rubber particles are not being extracted with the AC.
As with the mixes prepared in the lab, the DP mix performed least favorably in the Cantabro wear test, but the polymer modified mix combined with cellulose fibers (DCP) performed best. No difference was noted in the permeabilities of the more finely graded "d" mix and the coarser D mix. All of the other test mixes had better permeabilities than these two mixes.
As mentioned earlier, the polymer mix, DP, was difficult to place because of the sensitivity of screed forces to paver speed, resulting in varying placement depth. Core results verified these varying placement depths, indicating that the DP mix thickness varied as much as 0.4 in., the most of any of the mixes. The DCP mix may have derived some stabilization from the fibers, since the mix thickness varied by only 0.17 in. After placement of the test sections, Maysmeter smoothness and surface friction testing was conducted on each mix. Low smoothness values and high friction values are desirable.
20

Table 5: Laboratory Tests Results for Roadway Core Samples of Research and Control Mixes.

Sample No.

JMF

Sieve Size

%"

100

Coarse OGFC(D)
100

D +Mineral Fibers (DM)
100

D+

D+SB+

D+SB

Cellulose Cell. Fibers Polymer

Fibers (DC)

(DCP)

(DP)

Total Percent Aggregate Passing by Weight

100

100

100

D+16% Rubber (D16R)
100

JMF (d)
--

Standard I OGFC (d)
i
'
100

~,

99

99.3

98.6

99.2

97.6

99.3

99.2

100

98.7

3/8"

75

77.3

77.2

75.5

73.1

76.5

76.7

94

86.0

No.4

18

28.1

28.3

28.0

26.9

27.8

28.0

27

37.8

No.8

8

13.1

13.6

13.7

13.0

13.1

13.1

8

15.5

No. 200

2

3.8

4.1

3.5

3.9

3.8

3.4

2

4.1

Miscellaneous Test Data

%Binder EXTR/JMF 5.51/6.00

5.87/6.30

6.18/6.40

5.27/6.40

5.85/6.20

5.69/6.60 EXTRIJMF 6.09/6.20

N......

Max. Gravity

--

2.484

2.445

2.429

2.424

2.484

2.445

--

2.429 !

Density (pet)

--

%Voids

--

Viscosity

--

127.5

126.3

126.7

127.1

127.3

125.4

--

17.8

17.2

16.4

16.0

17.6

18.1

--

4,985

5,164

6,134

19,033

35,589

10,035

--

122.7 I
20.4 i
5,176

(Ps)

Perm.

--

152

270

235

234

277

222

--

155

(ft./day)

Layer (in.)

--

0.73

0.81

0.89

0.83

0.96

0.86

--

0.73

-

6. POST-CONSTRUCTION EVALUATIONS The pavement test sections were monitored for a period of 3-1/2 years to evaluate their per-
formance. The evaluations included both physical testing on the roadway and laboratory testing of core samples. The physical testing on the roadway included friction testing, smoothness testing, falling weight deflectometer (FWD) testing, a visual distress survey, rut measurements, and permeability testing.

6.1. Friction Testing Friction testing was performed on the test sections one day, two weeks, six months, and 3-112 years after construction. Friction testing followed ASTM-E274 procedures. Table 6 presents the results of these tests.

Table 6: Average Friction Test Results in Test Sections.

Test Section
Standard OGFC (d) Coarse OGFC (D) D + Mineral Fibers
(DM) D + Cellulose Fibers
(DC) D +Cellulose Fibers+
SB Polymer (DCP) D + SB Polymer (DP)
D+16%Crumb Rubber (D16R)

10/27/92 42 41 39
37
35
32 37

Friction Number

11/11/92

4/12/93

53

52

50

52

50

53

47

53

46

52

47

51

48

53

2/6/96 50 51 49
49
50
51 51

Friction values obtained on the day after construction (10/27/92) were naturally lower. Friction tests run two weeks later showed friction values had increased to an average of 49, with 46 for the DCP section being the lowest value and 53 for the standard "d" section being the highest

22

value. Friction values recorded six months after construction (4/1993) and 3-112 years after construction (2/1996) were satisfactory for all surface types. No significant differences occurred in the friction values between surface types over 3-112 years of service. Friction values typically ranged from 49 to 51 after 3-1/2 years.

6.2. Smoothness Testing Smoothness testing was performed on test sections 3-1/2 years after construction (5/1996). The new laser profiler was used in the testing. Table 7 contains the smoothness testing results.

Table 7: Average Smoothness Values in Test Sections.

Test Section
Standard OGFC (d) Coarse OGFC (D) D + Mineral Fibers (DM) D +Cellulose Fibers (DC) D +Cellulose Fibers + SB
Polymer (DCP) D + SB Polymer (DP) D + 16 % Crumb Rubber
(D16R)

Smoothness Value, 5/96 (HRI, mm/km) 427 547 564 551 543
585 642

The acceptable roughness value for GDOT OGFC is 750 mm/km, and smoothness test results show that all test sections have acceptable smoothness values after 3-1/2 years of service.

6.3. Visual Distress Survey Following the construction of test sections, annual visual distress surveys were conducted to determine the existence of surface distress in the sections. Visual distress surveys were performed according to the GDOT PACES rating method. After 3-1/2 years, visual distress found in the test sections consisted of reflective cracking and some longitudinal cracking. In order to
23

compare the distresses in the test sections, 0.2 mile portions of each test section were surveyed, and the results of this survey are given in Table 8.

Table 8: Results of Distress Survey on Test Sections.

Test Section

Description of Distresses

Standard OGFC (d) Coarse OGFC
(D) D +Mineral Fibers (DM) D + Cellulose Fibers (DC) DC+ SB Polymer(DCP)
D+SB Polymer (DP)
D+ 16% Crumb Rubber
(D16R)

Low severity reflective cracks
Medium severity reflective cracks & some longitudinal cracks at right edge of right lane Low severity reflective cracks
Low severity reflective cracks
Low severity reflective cracks
Low severity reflective cracks
Low-medium severity reflective cracks & some longitudinal cracks at right edge of right lane

Number of Reflective Cracks in 0.2 Mile 7 26 10 3 12 4 20

As shown in Table 8, Test Section DC had the lowest distress, three low severity cracks in 0.2 mile. Test Section DP also had a low level of distress. Test Section D experienced the highest level of distress. This section had 26 medium severity retlective cracks in 0.2 mile. Test Section D16R also experienced a high level of distress. Both D and D16R test sections had some longitudinal cracks at the edge of the right lane.

6.4. Rutting Measurements Rut measurements were taken in each test section several times during the 3-112 year evaluation period. The rutting was measured with a string line at intervals of 100 feet and recorded in both the left wheel path (LWP) and the right wheel path (RWP). Table 9 contains the average rut

24

measurements for each test section in 1993, 1994, and 1996. As indicated in Table 9, test section "d" (standard OGFC) had the highest rutting in the 3-1/2 years after construction. Rut depths in all of the other test sections had similar magnitudes.

Test Section
Standard OGFC (d)
Coarse OGFC (D)
D +Mineral Fibers (DM) D + Cellulose Fibers (DC)
DC+SB Polymer (DCP) D + SB Polymer
(DP) D + 16% Crumb Rubber (D 16R)

Table 9: Average Rutting in Test Sections (inches).

11/93

LWP

RWP

0.09

0.06

0.17

0.14

0.16

0.14

0.11

0.05

0.13

0.03

0.07

0.04

9/94

LWP

RWP

0.14

0.17

0.14

0.19

0.13

0.17

0.05

0.14

0.03

0.12

0.04

0.14

2/96

LWP

RWP

0.25

0.28

0.16

0.10

0.07

0.14

0.15

0.17

0.12

0.06

0.18

0.08

0.16

0.17

6.5. Permeability Testing The respective permeabilities of the test sections were monitored over the past 3-112 years. In each test section, permeability was measured across the lane in at least three locations: in the inner wheelpath, between the wheelpaths, and in the outer wheelpath. Table 10 shows the respective permeability measurements in these three locations for each test section. From these measurements, the average permeability was then obtained for each test section, as shown in Table 11.

25

Test
d D DM DC DCP DP D16R

Table 10: Permeability Results in Test Sections by Wheelpath.

OWP
46.1 51.9 98.2 63.5 46.2 51.0 62.8

1993
MWP
12.4 6.1 46.1 36.3 75.5 127.1 56.6

Permeability(ft./day)_

IWP 7.2 35.1 55.7 84.1 80.3 108.7 48.2

OWP
13.0 56.4 23.8 4.8 31.5 12.5 38.8

1996
MWP
3.4 3.7 5.1 17.9 8.8 7.7 3.4

IWP
7.1 26.7 9.6 62.4 15.6 12.8 34.6

Table 11: Average Permeability Results in Test Sections.

Test Section
Std. OGFC (d) Coarse OGFC (D)
D +Mineral Fibers (DM) D + Cellulose Fibers (DC) DC+ SB Polymer
(DCP) D + SB Polymer
(DP) D+ 16%Crumb Rubber (D16R)

Average Permeability (ft./day)

1992
146 142 254

1993
21.9 31.0 66.7

1996
7.8 28.9 12.9

222

61.3

28.4

220

67.4

18.6

262

95.6

11.0

210

55.9

25.6

Permeability in 1996 as % of Permeability
in 1992
5% 20% 5%
13%
8%
4%
12%

The permeability of all the test sections has substantially decreased with time. After 3-1/2 years of service, the lowest permeability was found in the control section (d), which represented fine OGFC. The permeability in the test sections after 3-1/2 years as a percentage of the initial permeability ranged from 4% to 20%.
The decrease in average permeability is mainly due to clogging in the asphalt layer. Figures 3 and 4 on pp. 28-29 show, for the years 1993 and 1996 respectively, the permeabilities of each

26

test section in the middle of the wheelpaths, and in the outer and inner wheelpaths. Figure 4 indicates that after 3-112 years of service, the permeability of six out of seven test sections was considerably higher at the middle of wheelpaths than in the wheelpaths themselves. This difference in the permeability occurring in the wheelpaths and in the middle of the wheelpaths strongly supports the assumption that greasy substances dropped from traffic are the main cause for the clogging of openings in the surface course of the pavement. Also, the squeezing action of the tires in the wheelpaths may reduce the clogging and increase the permeability therein.
27

Figure 3: Permeabilities of Open-Graded Friction Course Mixes (1993).

140

120

-~ 100

~

N 00

"-.0...

-..:=
'....-c...."

80

~ 60

-a ~

~
~

40

Outer Wheelpath IB Mid Wheelpath DInner Wheelpath

20

0
d

D

DM

DC

DCP

DP

Open-Graded Friction Course Mix

D16R

Figure 4: Permeabilities of Open-Graded Friction Course Mixes (1996).

70

60

50
-.".'.t.:.:.).
"4:! 40
--......c..

~ ~

30

a ~

r.

~
~

20

10

0
d

D

DM

DC

DCP

DP

Open-Graded Friction Course Mix

D16R

Outer Wbeelpath Ill Mid Wheelpath DInner Wheelpath

6.6. Falling Weight Deflectometer (FWD) Testing Deflection data was collected from test sections using the FWD and compared to identify differences. Table 12 shows the average normalized deflections recorded in FWD sensor numbers one and seven in each test section. Since the loads recorded in FWD testing vary, Table 12 is based on deflections normalized to 9000 lb. In general, FWD sensor one deflection indicates the structural integrity of the upper layers of the pavement, while sensor seven gives a rough indication of subgrade conditions.

Table 12: Average Normalized Deflections in Test Sections.

Test Section
Coarse OGFC (D) v D +Mineral Fibers (DM) D + Cellulose Fibers (DC) D +Cellulose Fibers+ SB
Polymer (DCP) D + SB Polymer
(DP) D+16%Crumb Rubber (D16R)

ND1 (mils)

1994

1996

4.07

3.98

5.14

4.51

5.66

5.63

4.39

4.96

4.98

5.0

6.72

6.82

ND7 (mils)

1994

1996

1.15

1.24

1.16

1.21

1.52

1.59

1.45

1.69

1.38

1.56

1.47

1.37

Normalized deflections show that all six test sections are in good structural condition. Normalized deflection one (ND1) in the crumb rubber test section (D16R) is significantly higher than the other ND 1 values in both the 1994 and 1996 deflections. Sufficient evidence is not available to explain the higher ND1 values of the D16R test section. The 16% crumb rubber in the surface asphalt layer may contribute to the higher deflections at sensor one. A visual distress survey identified fairly high levels of reflective cracking in the D16R section. This cracking may

30

indicate a weaker surface layer, which could cause higher NDI values. However, the distress survey identified high reflective cracks in Section D, which recorded the lowest ND 1 values. Firm conclusions, therefore, cannot be made using the FWD deflection results.
6.7. Laboratory Evaluation of Core Samples During the study period, core samples were taken from test sections for laboratory analysis. Results of these analyses are shown in Tables AI to A7 in the Appendix C. Laboratory analysis of core samples showed asphalt viscosity increasing in all test sections from the time of construction to the last evaluation. This trend in the asphalt viscosity is normal. Another trend observed was the reduction in the asphalt content of the pavements in the same time period. This reduction was found in most of the test sections. The cause of this reduction may be the wearing of the friction course, which caused some asphalt to be lost in the surface course.
7. CONCLUSIONS AND RECOMMENDATIONS The friction levels of all six test sections were acceptable. No significant difference was
noted between the friction values of the control section (fine OGFC) and the test sections after 31/2 years of service. Also, the smoothness levels of the six test sections were acceptable.
The coarse OGFC (D) test section and the coarse OGFC with 16% crumb rubber (D16R) test section experienced high levels of cracking compared with the other test sections and the control section. The sections containing coarse OGFC with cellulose fibers and coarse OGFC with SB polymer had the lowest levels of cracking. Rutting in the control section (fine OGFC) was significantly higher than rutting in the test sections. All test sections showed similar degrees of rutting after three and one-half years of service.
31

All six test sections showed better permeability than the control section. In all types of OGFC, permeability decreases sharply over time due to clogging. Clogging of openings is significantly higher in between wheelpaths than in the wheelpaths themselves, possibly due to greasy substances dropped between the wheelpaths from moving traffic, or from the squeezing action of tires in the wheelpaths upon surface water.
Based on the results of this study, the following actions are recommended: 1) The use of coarse OGFC should be continued and the use of fine OGFC discontinued. 2) Both fiber and polymer should be included in coarse OGFC. Neither additive appears to perform significantly better or worse than the other in coarse OGFC. 3) Test sections may need to be continuously monitored to improve comparisons between them. 4) The use of crumb rubber in coarse OGFC is not recommended.
32

BIBLIOGRAPHY
Amirkhanian, S.N., "A Feasibility Study of the Use of Waste Tires in Asphaltic Concrete Mixtures," Report SC-92-04, May, 1992.
Decoene, Y., "Contribution of Cellulose Fibers to the Performance ofPorous Asphalts," Transportation Research Record 1265, National Research Council, Washington, DC, (1990).
"NCHRP Synthesis of Highway Practice 180, Performance Characteristics of Open-Graded Friction Courses," Transportation Research Board, National Research Council, Washington, DC, (1992).
Perez-Jimenez, F.E. and J. Gordillo, "Optimization ofPorous Mixes Through the Use of Special Binders," Transportation Research Record 1265, National Research Council, Washington, DC, (1990).
Schellenbert, K. and W. Von der Weppen, Verfahren zur Bestimmung der HomogenitatsStabilitat von Splittmastixasphalt. "Bitmen," No. 1, Jan. 1986.
Taylor, Donald, "Fundamentals of Soil Mechanics," John Wiley and Sons, Inc., New York, N.Y.

APPENDICES

APPENDIX A:
TEST METHOD GDT-114

111-114 Pa9e l of 5

GDT-114

~

METHOD OF TEST FOR

DETERMINING OPTIMUM ASPHALT CONTENT FC

OPEN GRADED BITUMINOUS PAVING MIXTURES

The purpose of this test method is to determine the optimum asphalt content for an open-graded bituminous mix.

TEST METHOD

A. SCOPE:

This method of selecting the asphalt content consists of three steps. The first is to conduct a measurement of the surface capacity (KCl of the predominant aggregate size fraction (material retained on No. 4 sieve).
Surface capacity includes absorption, superficial area, and surface roughness, a11 of which affect asphalt cement requirements. The second step is to mold nine samples by using a modified AST~D-1559. The third
step is to prepare samples for visual evaluation in pyrex pie pans.

B. APPARATUS:

The apparatus required shall consist of the following:

l. 13 metal pie pans
2. oven capable of maintaining 121 5C (250 9F) temperature 3. oven capable of maintaining 60f3C (140f5F) temperature 4. beakers, glass, 500 ml 5. balance~ 5000 gr. capacity 0.1 gr. accuracy 6. glass funnels, top diameter 88 mm (3 1/2 in.), height 114 mm l4 1/2 in.)
orifice 13 mm (1/2 in.) ~ith a piece of 2 mm CNo. 10) sieve exposed to the bottom of opening. Cork stopper to fit outlet. 7. oil, S.A.E. No. 10 lubricating
B. timer
9. Marshall desian equipment as specified in AASHTO: T-245 10. 4 twelve inch- diameter py~ex pie pans

c. STEP :1.

SURFACE CAPACITY (KC)

1. Determine the surface capacity of the aggregate fraction that is retained on a No. 4 sieve in accordance ~ith the following procedure.

1.1 Quarter out 105 gr. of aggregate representative of the material passing the three-eight inch sieve and retained on the No. 4 sieve.
1.2 Dry aggregate in 250~~F oven to a constant weight and allo~ to cool.

1.3 weight out 100.0 gr. and place in glass funnel.
1.4 Completely immerse the aggregate in S.A.E. No. 10 oil for five
minutes by plugging funnel outlet ~ith cork stopper.

June, 1989 1.5 Drain for -two minutes.

lll-114
Page 2 of 5
.,

l.6 Place funnel containing sample in the 140Cf oven for 15 ~inutes for aoditional draining.

1.7 four sample from funnel 1nto a tared ~an; cool and rewei9h sample
to nearest 0.1 gram. Subtract original ~i~ht and record difference as percent oil retained lbased on 100 gr. of dry a.lgregate).

l.B Use the attached chart for determination of MKcu.

a. If the apparent specific gravity for the fraction is greater

than 2.70 or less than 2.60 apply correction to oil retained.

using the following formula:



.-

Oil Retained Corrected (~)=Oil Retained (~) X Apparent Sp. Gr. of Coarse Aog.

2.65

b. Start at the bottom of chart ~ith the corrected percent of oil retained; follow a straightedge vertically up~ard to
intersection with diagonal line; hold point, and follo~ the straightedge horizontally to the left. The value obtained will be the surface content for the retained fraction and is
known as KC.

1.9 Determine the required asphalt content from the following formula.
= Percent Asphalt 2.0 (KC) + 3.5

lno correction need be applied for viscosity)

D. STEP 2

MODIFIED MARSHALL DESIGN

2. The folowing procedure is suggested for the Marshall design:

2.1 Heat aggregate to 27~F; heat molds to 300F. and heat A.C. to proper mixing temperature specified in Mixture Control Temperature Chart.

2.2 Mix aggregate with asphalt at three asphalt contents in 0.5~ intervals nearest the optimum asphalt content established in step 1. Three specimens should be compacted at the nearest 0.5~ interval and three specimens each at both 0.5~ above and belOH the interval.

2.3 After mixing, return to oven if necessary, and when at 250 F.
compact using 25 blo~s on each side. (Use oil on base plate to prevent sticking).

2.4 When compacted, cool to room temperature before removin9 from mold.

June l9B9

--
Page 3 of 5

2.5 Bulk specific gravity

a. Determine the density of a regular shaped specimen of compacted ~ixture from its dry mass (in grams) and its
volume in cubic centimeters obtained from its dimensions for
height and radius. Convert the density to bulk specific gravity by dividing by 0.99707 gr/cm 3 , the density of "ater at 25C (77f).

Formula: Bulk Sp. Gr. =

w

= Weicht (gms.) x 0.0048417

Height {in.)

w= Weight of specimen in grams
1r : 3.1416
T' = Radius in centimeters
= h Height in centimeters
0.99707 = Density of "ater at 25C (77F)

2.6 Determine stability and flow after 1 hour in 77F ~ater bath.
2.7 Calculate percent voids VMA~ and voids filled ~ith a~phalt based on aggregate bulk specific gravity.
2.8 Plot VMA curve versus A.C. content.
2.9 Select the asphalt content at the lowest point on the VMA curve
.::E=-=..-~S:....T..:.-:::E::..:P:...--....:3=--___;P~Y....:R-=E::..::X...___.-B OWL MET H 0 0
3. The follo~ing procedure is to be used for visual evaluation:
3.1 Batch up 990 9ms. of aggregate in each of four pans and heat a9gregate to 2509F.; heat A.C. to temperature specified in Step 2.1 above
3.2 Add 10 grams hydrated lime into mixing bowl of heated aggregate; mix thoroughly to coat aggregate.
3.3 Add A.C. at 5.5~ of total ~eight and thoroughly mix. Pour mixture into clean, clear glass (Pyrex) pans. Repeat the process for the remaining pans except that the amount of A.C. in the total mix is increased in 0.5~ increments (such as 5.5~, 6.~, 6.5~, 7.0S).
3.4 Place samples back in 25trF. oven for one hour; remove and let cool to room temperature.
3. 5 Vi sua11 y observe the amount of H quid aspha1t on the bot tom of each pan.

.-- ...

....
J1.1ne, 1989

111-114 Page 4 of 5

3.6 Sele'ct t.he asphalt content at a level where ample bonding is evident between film coating and glass pan without having
excessive drainage.

F. SELECT OPTIMUM A.C.

A. Determine the optimum A.C. content for the mix by averaging the selected asphalt contents from the three methods (surface capacitv. modified Marshall and {Pyrex B~ll. ~ny previous field experience
should also be considered in establishi119 the optimum A.C.

G. STEP 5

BOIL TEST

Perform Boil Test according to GHD-56 except that the re9ular batch of open-graded mix shall be used (at optimum asphalt content determined in
section F above). If the sample treated with hydrated lime fails to maintain 95\ coating~ a sample shall be t.ested in which O.SS liQuid antistrip additive has been used t.o treat t.he asphalt cement 1n addition t.o
treatment of aggregate with hydrated lime. In some cases a specific brand of A.C. and/or liquid additive may need to be used to prevent stripping.

June. 1989

DETERMINATION OF KC FACTOR

lll-114 Page 5.,of 5

.o~:~~~~-~~~~~~-~~~~~~-~-~sr-=~~-~--~-@.---~-~-~-~-~-~-~-~~~=-~~-~_~,~~~z'~~/~~-~~,~~~-;

[

- -- -:--.:-::.= ..:::.::=:.:-:-:-_ ... - --- -

--_ ~:- .;.,:--.=y ~: .----:.:

.6 ;-

-

--

..:-.::- --

==:::!

--:-:__: :-.::~:: -:7=~--~:----~:-- 7'-

1

::J

~

---~---=-=i- ---=:---.....:.:..:_. . : / .:..::-_~_-::: _ -- ---~

.. _,.

.8
b
.4

--- ~~_.

I
=-7..--

.,

- Y' --

7-

'- :=j

. ----,- -,

.0

;=l

.8

--r--

-----"------,--

.6

.4
;..___
.2

/
-----;;--- ~

j

-:

!

.; 'T ~:..::.t7'
--/

I

I

-~~--
:
i

i~~

r---

.. .--'--/- ~-+
..,~::-~

:~ -j -/-'-'__;_1-~

. v

/,

I I

~---+--~--r-~~--+--,-~,r~-+~~~-+~-~~-~~~------T-----+----~--~---~--+-----~-----~----1----

/

I

/

I i

i

! I/

1.5

I I

' I; ! : Ii

' :I

;

~
!

' I

I !I I'

!

I

! I I I

2

I I I I I I I ! I i l I I ! I ! ; II ! ! ;t : I lj: :''' . I

I

I

I I iII
I I l II 3

i

I
!

! I

I '

i i; I
'
I 1I I

i: i I
I! ~ I

l ~ i: iiI T,;!

.;:! 'll1 ,,

4

s

6 7

I
I

8

I I
'I
9

I
1(

,
: OIL RETAINED (Corrected for Aggregate Sp. Gr.)

APPENDIXB:
FIBER AND POLYMER SPECIFICATIONS

September 14, 1992

GEORGIA DEPARTMENT OF TRANSPORTATION STATE OF GEORGIA

SUPPLEMENTAL SPECIFICATION

Modification of Standard Specification SECTION 819 - FIBER STABILIZING ADDITIVES

819.01 DESCRIPTION: This Section covers the general requirements for fiber stabilizing additives which are incorporated into asphaltic concrete mixtures. These fibers are generally used to stabilize the aspha1t fi 1m surrounding z.ggregate particles in order to reduce drain-down of the asphalt cement. A fiber stabilizer such as cellulose or mineral fiber shall be utilized.

819.02 MATERIALS: The selected fiber shall meet the properties described below. Dosage rates given are typical ranges but the actual dosage rate used shall be approved by the Office of Materials and Research.

A. Cellulose Fibers: Cellulose fibers shall be added at a dosage rate between 0.2% and 0.4% by weight of the total mix as approved by the Engineer. Fiber properties shall be as follows:

1. Fiber length:

0.25 inch (maximum)

2. Sieve Analysis:

a. Alpine Sieve Method

Passing No. 100 sieve: 60 - 80%

b. Ro-Tap Sieve Method

Passing No. 20 sieve: 80 - 95%

Passing No. 40 sieve: 45 - 85%

Passing No. 100 sieve: 5 - 40%

3. Ash Content:

18% non-volatiles (5%)

4. pH:

7.5 (:tl.O)

5. Oil Absorption:

5.0 (1.0}

(times fiber weight)

6. Moisture Content:

5.0% (maximum)

B. Cellulose Pe11ets: Cellulose pellets shall consist of a 50/50

blend of cellulose fiber and asphalt cement and shall be added at

a dosage rate between 0.4% and 0.8% by weight of the total mix.

The cellulose used shall comply with requirements of Section

819.02.A.

.

1. Pellet size: 2. Asphalt:

1/4 cubic inch (maximum) 25 - 80 pen.

C. Mineral Fibers: Mineral fibers shall be made from virgin basalt or diabase which is to be treated with a cationic sizing agent to enhance disbursement of the fiber as we11 as increase adhesion of the fiber surface to the bitumen. The fiber shall be added at a dosage rate between 0.3% to 0.5% by weight of the total mix as approved by the Engineer.

1. Size Analysis: Maximum Fiber length: 0.25 inches Average Fiber thickness: 0.0002 inches (maximum)
2. Shot content (ASTM C612l Passing No. 60 sieve: 90 - 100% Passing No. 230 sieve: 65 - 100%

Materials and Research

September 14, 1992 DEPARTMENT OF TRANSPORTATION
State of Georgia SUPPLEMENTAL SPECIFICATION Modification of the Standard Specification SECTION 820 - ASPHALT CEMENT
Retain this section as written and add the following: 820.02 POLYMER MODIFIED ASPHALT CEMENT: When the asphalt cement for any particular asphaltic concrete mixture on this project is to be modified with a thermoplastic polymer, the asphalt cement to be modified must comply with requirements in 820.01. Sufficient polymer shall be added to the asphalt cement to produce a modified binder that complies with the following requirements. The composite shall be thoroughly blended at the supply facility prior to being loaded onto the transport vehicle or at the asphalt plant prior to being injected into the mix. All blending procedures and operations must be approved by the Office of Materials and Research.
A. TEST REQUIREMENTS FOR POLYMER MODIFIED ASPHALT CEMENT 1. Polymer Modified Asphalt Cement: Viscosity at 140F, poises, minimum ------------------5000 Viscosity at 275F, eSt, maximum ---------------------2500 Penetration at 77F, 100 g, 5 sec, dmm ---------------40-75 Flash )oint, CDC, F, minimum ------------------------450 Ash Content, maximum : -------------------------------1.0 Softening Point, Ring and Ball, F, minimum ----------135 Force Ductility, Pounds @10 em extension, 5 em/min, 77F, minimum ------------------0.3 Elastic Recovery from 10 em, 5 em/min,
S7e7paFr,atmioinnimtuemst,---R-in-g---&--B--a-ll--d-i-f-f-e-r-e-n--c-e-,-------------50S F. maximum -----------------------------------------4
2. Properties after Thin-Film oven treatment: Viscosity Ratio, 140F, Maximum ----------------------2.5 Duc.tility at 39.2F ,cm,5 em/min, minimum--------------10

DEPARTMENT OF TRANSPORTATION STATE OF GEORGIA SPECIAL PROVISION
Modification of Standard Specific~tions SECTION 828- HOT MIX ASPHALTIC CONCRETE MIXTURES

828.02 OPEN - GRADED SURFACE MIXTURE: Delete Grading Requirements as shown for Asphaltic Concrete 110'' and substitute the following:

Sieve Size

Percent Passing

3/4" 1/2" 3/8" No.4 No.8 No.200

100 90-100 65-85 15-25 5-10 2-4

The rema1n1ng Job Mix Formula and Design Limit requirements for this mixture will be as shown in this sub-Section.

For this project the asphalt cement shall be modified with a thermoplastic polymer at the dosage rate recommended by the supplier. The mixture shall also be treated with mineral fibers at the rate of 0.5~ of the total weight of the mixture. The source of polymer and fiber shall be approved by the Office of Materials and Research.

The spread rate for Asphaltic Concrete 11D" Mix for this project shall be 75 lbs./s.y.

Materials and Research

APPENDIXC:
LABORATORY TEST RESULTS FOR ROADWAY CORE SAMPLES

Table Al: Laboratory Test Results for Roadway Core Samples on Coarse OGFC (D).

Sample No. Date Sampled
Sieve Size 3/4" 112" 3/8" No.4 No.8
No. 16 No. 50 No. 100 No. 200
%Binder
Max. Gravity Density (pet)
%Voids 140 f Vis. (Ps)

JMF
--
100 99 75 18 8
--
---
2
6.00
-----

D

10/19/92

11123/94

2/20/96

Total Percent Aggre ate Passing by Weight

100

100

100

98.3

98.0

98.8

70.0

75.4

76.2

21.0

31.2

28.8

8.7

14.7

14.6

--

11.8

11.7

--

8.5

8.4

--

6.3

6.2

3.6

4.2

4.1

Percent Bitumen of Total Mix

5.85

4.83

4.26

Miscellaneous Test Data

2.484

2.478

136.1

133.2

12.2

12.6

2.485
---

5,890

80,577

111,398

Table A2: Laboratory Test Results for Roadway Core Samples on Coarse OGFC with Mineral Fibers (DM).

Sample No. Date Sampled
Sieve Size 3/4" 112" 3/8" No.4 No.8
No. 16 No. 50 No. 100 No. 200
%Binder
Max. Gravity Density (pet)
%Voids 140 f Vis. (Ps)

JMF
--
100 99 75 18 8
--
---
2
6.3
---
---

DM

10/19/92

11123/94

Total Percent Aggregate Passing by Wei~ht

100

100

98.6

99.0

77.2

78.4

28.3

33.0

13.6

16.7

--

13.3

--

9.7

--

6.1

4.1

4.3

Percent Bitumen of Total Mix

5.87

5.87

Miscellaneous Test Data

2.445

2.441

135.2

133.2

11.4

12.6

5,757

46,144

2/20/96
100 97.4 75.4 27.3 13.4 10.7 8.0 6.3 4.3
4.81
2.469
958,842

Table A3: Laboratory Test Results for Roadway Core Samples on Coarse OGFC with Cellulose Fibers (DC).

Sample No. Date Sampled
Sieve Size 3/4" 112" 3/8" No.4 No.8
No. 16 No. 50 No. 100 No. 200
%Binder
Max. Gravity Density (pet)
%Voids 140 op Vis. (Ps)

JMF
--
100 99 75 18 8
---
--
2
6.4
--
----

DC

10/19/92

11123/94

Total Percent Aggregate Passing by Wei!!ht

100

100

99.2

97.7

75.5

76.7

28.0

30.8

13.7

15.1

--

11.9

--

8.9

--

6.8

3.5

4.7

Percent Bitumen of Total Mix

6.18

5.77

Miscellaneous Test Data

2.429

2.44

126.7

133.1

16.4

12.6

6,134

417,617

2/20/96
100 98.6 75.5 25.9 12.4 9.8 7.3 5.6 3.9
5.24
2.429 126.7 16.4 97,924

Table A4: Laboratory Test Results for Roadway Core Samples on Coarse OGFC with Cellulose Fibers and SB Polymer (DCP).

Sample No. Date Sampled
Sieve Size 1" 3/4" 1/2" 3/8"
No.4 No.8 No. 16 No. 50 No. 100 No. 200
%Binder
Max. Gravity Density (pet)
%Voids 140 op Vis. (Ps)

JMF
--
--
100 99 75 18 8
----
2
6.4
-----

DCP

10/19/92

11/23/94

Total Percent Aggregate Passing by Wei!!ht

--

--

100

100

97.6

97.8

73.1

76.7

26.9

33.2

13.0

17.7

--

14.4

--

10.0

--

6.9

3.9

4.3

Percent Bitumen of Total Mix

5.27

5.82

Miscellaneous Test Data

2.424

2.425

127.1

131.7

16.0

13.0

19,033

122,856

2/20/96
--
100 97.4 76.3 31.5 16.4 13.2 9.7 6.8 4.5
5.59
2.421
121,000

Table AS: Laboratory Test Results for Roadway Core Samples on Coarse OGFC with SB Polymer (DP).

Sample No. Date Sampled
Sieve Size 3/4" 112" 3/8" No.4 No.8
No. 16 No. 50 No. 100 No. 200
%Binder
Max. Gravity Density (pet)
%Voids 140 op Vis. (Ps)

JMF
--
100 99 75 18 8
----
2
6.2
--
--
--
--

DP

10/19/92

11123/94

2/20/96

Total Percent Aeere2ate Passing by Weight

100

100

100

99.3

98.4

99.5

76.5

79.8

78.8

27.8

34.9

32.0

13.1

17.8

15.3

--

13.9

12.2

--

9.7

9.0

--

6.9

6.5

3.8

4.5

4.3

Percent Bitumen of Total Mix

5.85

6.21

5.40

Miscellaneeus Test Data

2.484

2.471

127.3

134.6

17.6

12.7

35,589

35,057

2.433
--
--
231,660

Table A6: Laboratory Test Results for Roadway Core Samples on
Coarse OGFC + 16% Crumb Rubber (D16R).

Sample No. Date Sampled
Sieve Size 3/4" 1/2" 3/8" No.4 No.8
No. 16 No. 50 No. 100 No. 200
%Binder
Max. Gravity Density (pet)
%Voids 140 op Vis. (Ps)

JMF
--
100 99 75 18 8
--
---
2
6.6
-----

D16R

10/19/92

11123/94

Total Percent Aeere2ate Passingby Weil!ht

100

100

99.2

98.1

76.7

80.0

28.0

34.3

13.1

19.4

--

15.3

--

10.1

--

6.8

3.4

4.3

Percent Bitumen of Total Mix

5.69

5.80

Miscellaneous Test Data

2.445

2.437

125.4

134.3

18.1

11.7

10,035

41,603

2/20/96
100 99.5 78.4 32.4 12.8 14.0 9.5 6.5 4.5
5.10
2.454
---
143,455

Table A7: Laboratory Test Results for Roadway Core Samples on Standard OGFC (d)

Sample No. Date Sampled
Sieve Size 3/4" 1/2'' 3/8" No.4 No.8
No. 16 No. 50 No. 100 No. 200
%Binder
Max. Gravity Density (pet)
%Voids 140 op Vis. (Ps)

JMF
-
--
100

d

10/19/92

11123/94

Total Percent Ae:e:regate Passing by Weight

100

100

98.7

98.6

2/20/96
100 99.1

94

86.0

86.5

85.1

27

37.8

41.2

39.1

8

15.5

18.1

16.7

--

--

13.8

11.2

--

--

9.7

7.2

--

--

6.9

4.6

2

4.1

4.9

2.4

Percent Bitumen of Total Mix

6.2

6.09

5.61

5.17

Miscellaneous Test Data

--

2.429

2.462

2.463

--

122.7

132.4

--

--

20.4

13.8

--

--

5,176

618,295

105,555