Annotated Literature Review for NCHRP Report 640 (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

165 instances, one year worth of accident reports were compared. The data indicated that wet weather accidents decreased 34 percent when dense-graded surfaces were replaced with porous asphalt. 1.39.2 Materials and Mix Design The only information provided relative to materials and mix design is that the target air voids contents within construction porous asphalt is approximately 20 percent. 1.39.3 Benefits Two benefits mentioned within the paper were decreased noise levels and reduced wet weather accidents (improved wet weather frictional resistance). 1.39.4 Construction Practices No specifics construction practices were given. 1.39.5 Maintenance Practices Research conducted by the authors indicated that there is no significant difference between dense-graded surfaces and porous asphalt surfaces in terms of road surface conditions during a snowfall event. Additionally, there is no significant difference in the salinity concentration between the two surfaces when using anti-icing chemicals. 1.39.6 Rehabilitation Practices No specific rehabilitation practices were given. 1.39.7 Performance The authors indicated that porous asphalt surfaces offer higher skid resistance during rain events, even if the surface is slightly frosted. 1.39.8 Structural Design No specifics on inclusion within structural design were given. 1.39.9 Limitations No specific limitations were given. 1.40 Kandhal, P.S. “Design, Construction and Maintenance of Open-Graded Asphalt Friction Courses.” National Asphalt Pavement Association Information Series 115. May 2002. 1.40.1 General This document provides an overview for materials selection, mix design, construction, pavement structural design, winter maintenance and rehabilitation for OGFCs. Specific chapters are provided on performance benefits, mix design, mix production and placement, pavement design considerations and maintenance and rehabilitation. Kandhal begins the document with a brief history of OGFCs. OGFCs evolved through experimentation with plant mix seal coats. During the 1940’s, the California Department

166 of Highways was experiencing problems with chip seals and seal coats. These problems included bleeding, raveling, loose stone and a relatively short performance life. To combat the problems, the cover aggregate was mixed with a relatively high asphalt binder content in a conventional HMA plant. These plant mix seal coats provided the same frictional benefits as chip seals but were more durable, provided some improvement in ride quality, reduced noise and eliminated loose chips. During the 1970’s, the Federal Highway Administration began a program to improve the overall frictional properties of the roadways within the U.S. The plant mix seal coats were identified as a tool to accomplish this goal. At this time, the term open-graded friction coarse was coined. During this time period, the FHWA published a mix design procedure for OGFC’s. A 1998 survey referenced in the document indicated a number of states had experienced good performance with OGFCs. Of those not experiencing good performance, the primary problems encountered included raveling, delamination and loss of permeability. The survey indicated that the use of polymer-modified binders, relatively high asphalt binder contents (by using fibers) and/or relatively open gradations had alleviated some of these problems. 1.40.2 Benefits of Permeable Asphalt Mixes Benefits of OGFC mixes were categorized as related to safety or environment. Benefits related to safety include improved wet pavement frictional resistance, less potential for hydroplaning, reduced splash and spray, reduced glare and improved visibility of pavement markings. Kandhal cited a number of references describing research conducted in the U.S., Canada and Europe that showed the improved wet pavement frictional resistance of OGFC pavements. Much of the research dealt with comparing the speed gradient (or friction gradient) of OGFCs. A frictional speed gradient can be defined as the rate of decrease of a friction number per unit increase in speed. Research by the Bureau of Public Roads during the late 1960’s showed that the friction gradient of OGFC pavements was considerably lower than for dense-graded pavements. Table 90 presents data from a Pennsylvania Department of Transportation project that also shows a decreased friction gradient for OGFC layers. Similar work in Oregon and Louisiana also shows decreased friction gradients for OGFC compared to dense-graded layers. Table 90: Frictional Data (Pennsylvania) Friction Number Mix Type 30 mph 40 mph Friction Gradient OGFC (gravel) 74 73 0.10 OGFC (dolomite) 71 70 0.10 Dense-graded HMA (gravel) 68 60 0.80 Dense-graded HMA (dolomite) 65 57 0.80

167 Research in Virginia, France and Canada showed that OGFC layers reduced wet weather crashes. In Virginia, wet weather accidents were reduced from 39 percent of all accidents on State Route 23 to 17 percent of all accidents. On the A7 motorway in France, the number of accidents fell from 52 (1979 to 1985) to none (1985 to 1989) after OGFC was placed on a section of roadway. In Canada, the placement of OGFC on a section of roadway reduced the number of wet weather accidents by 54 percent and the total number of accidents by 20 percent. During a rain event, water infiltrates into the OGFC layer. Because the water infiltrates into the OGFC layer, a continuous film of water will not be available to cause hydroplaning. Kandhal indicates that hydroplaning also may not occur during prolonged, heavy rainfalls because the pressure developed under a vehicle’s tire will dissipate through the porous structure of the OGFC. Another benefit of the OGFC related to its ability to drain water is the reduction of splash and spray. The use of OGFC almost eliminates splash and spray because water does not pool on the pavement surface. Kandhal references a UK research study that showed a 90 to 95 percent reduction in the amount of splash and spray on OGFC surfaces when compared to dense-graded surfaces. Due to the lack of water pooling on the pavement surface, drivers do not see the glare caused by the headlights of oncoming vehicles. The reduction in glare contributes to better visibility and helps result in reduced driver fatigue. The final benefit related to safety mentioned by Kandhal was improved visibility of pavement markings. Kandhal states that pavement markings on OGFC surfaces have high nighttime visibility, especially during wet weather. The lone benefit mentioned by Kandhal of OGFC surfaces related to the environment is reduction of tire/pavement noise. Kandhal cited numerous research studies that showed OGFC layers reduce tire/pavement noise approximately 3 dB(A) compared to dense- graded HMA. To put a 3 dB(A) reduction in noise into perspective, this reduction can also be achieved by reducing the volume of traffic by half. 1.40.3 Materials and Mix Design Kandhal presented a mix design system that was based upon research at the National Center for Asphalt Technology. The mix design system includes four primary steps which include: 1) materials selection; 2) selection of moisture susceptibility; 3) selection of optimum asphalt binder content; and 4) evaluation of moisture susceptibility. Materials needed for selection include coarse and fine aggregates, asphalt binders, and stabilizing additives. Kandhal states that aggregate requirements for OGFC can be similar to those for stone matrix asphalt (SMA). Aggregates should be tough, durable, angular and cubicle. Table 91 summarizes Kandhal’s recommendations for coarse aggregate property requirements for OGFC. For fine aggregates, the uncompacted void content (fine aggregate angularity) was recommended with a minimum value being 45

168 percent. Asphalt binders should be selected for the project location considering the environment, traffic and expected functional performance for the layer. Kandhal recommends the use of high stiffness binders, generally two grades stiffer than the local climate requires, for hot climates and cold climates having freeze/thaw cycles, medium to high traffic volumes and OGFC mixes with more than 20 percent air voids. The addition of fiber is also desirable to reduce the potential for draindown and to allow relatively high asphalt binder contents. Table 91: Coarse Aggregate Requirements for OGFC Property Recommended Criteria Los Angeles Abrasion less than 30 percent Fractured Faces 90 percent with two or more faces minimum Flat and Elongated 5:1 – 5 percent maximum 3:1 – 20 percent maximum The second step in the mix design system entails developing a design gradation. Similar to SMA, OGFC should have an aggregate skeleton that has stone-on-stone contact; therefore, the voids in coarse aggregate (VCA) of the gradation is determined (Figure 16). A single gradation band was recommended (Table 92). Similar to the Superpave mix design system, Kandhal suggests developing three trial blends using the selected aggregates. For each of the trial blends, the aggregates are mixed with 6 to 6.5 percent asphalt binder using 50 gyrations of a Superpave gyratory compactor. The air voids and VCA of the compacted trial blends are then evaluated and the blend having stone-on- stone contact with high air voids is selected as the design gradation.

169 Figure A: VCADRC is obtained from the Dry Rodded Unit Weight of just the coarse aggregate. Figure B: VCAMIX is calculated to include everything in the mix except the coarse aggregate. Figure C: VMA includes everything in the mix except the aggregate (both coarse and fine). For the VCAMIX and VMA calculations, asphalt absorbed into the aggregate is considered part of the aggregate. Air Coarse Aggregate Figure A 30-40% of Volume Dry Rodded Unit Weight VCADRC Air Coarse Aggregate Figure B VCAMIX Air Aggregate Blend Figure C VMA Asphalt Fine Aggregate Effective Asphalt Content Asphalt Effective Asphalt Content Fine plus Coarse Aggregate Figure 16: Voids in Coarse Aggregate Concept for Ensuring Stone-On-Stone Contact Table 92: Recommended Gradation for OGFC Sieve, mm Percent Passing 19 100 12.5 85-100 9.5 35-60 4.75 10-25 2.36 5-10 0.075 2-4 Kandhal recommends four properties in the selection of design asphalt binder content: air voids, Cantabro abrasion loss on unaged OGFC samples, Cantabro abrasion loss on aged OGFC samples and draindown. Optimum asphalt binder content should provide at least 18 percent air voids, though Kandhal indicates higher values are desirable. The Cantabro Abrasion test evaluates the resistance of compacted OGFC specimens to abrasion loss. The test method entails compacting an OGFC specimen to the design compactive effort, allowing the specimen to cool to room temperature, weighing the

170 specimen to the nearest 0.1 gram, and then placing the specimen into a Los Angeles abrasion machine without a charge of steel spheres. The Los Angeles abrasion machine is then operated for 300 revolutions at a rate of 30 to 33 rpm at a temperature of 25˚ C. After the 300 revolutions, the specimen is removed and again weighed to the nearest 0.1 gram and the percent mass loss determined. As stated above, there are requirements for both aged and unaged OGFC samples. For the aged condition, samples are placed in a forced draft oven at a temperature of 85˚C for 120 hours. The final property evaluated during the design of OGFC is draindown potential. For this testing, the draindown basket developed at the National Center for Asphalt Technology is used with a maximum allowable percentage of draindown being 0.3 percent. Kandhal states that laboratory permeability testing is optional. If utilized, permeability vales greater than 100 m/day are recommended. Optimum asphalt binder content is selected as a binder content in which all four properties are met. Generally, Cantabro abrasion loss will define a minimum asphalt binder content and draindown/air voids will define a maximum asphalt binder content. The final step in the mix design system is to evaluate the designed OGFC mix at the selected optimum asphalt binder content for moisture susceptibility. For this, AASHTO T283 is utilized with the following modifications: 1) specimens should be compacted to 50 gyrations with no air void requirement; 2) apply a partial vacuum of 26 inches of Hg for 10 minutes to saturate the specimens; and 3) keep the specimens submerged in water during freeze/thaw cycles. Kandhal recommends 5 freeze/thaw cycles. The retained tensile strength after 5 freeze/thaw cycles should be at least 80 percent. 1.40.4 Construction Practices Production and placement of OGFC is similar to dense-graded HMA; however, Kandhal provides some differences. The main modification required at the production facility is the addition of a method to incorporate fibers into the production process. Depending upon the type of plant (batch or drum), the addition of fiber can differ. In the case of batch plants, the fiber is added at the pugmill. Bales of loose fibers have been added manually into the pugmill and also the fibers have been blown into the pugmill from a fluffing device. Pelletized fibers can also be placed directly into the pugmill either manually or from a feeder. Within drum plants, fibers (loose or pelletized) are generally blown continuously into the drum. The fiber line is generally placed about 0.3m (1ft) upstream of the asphalt binder line. This is done so that the fibers are captured by the asphalt binder and do not end up in the dust collection system. During mixing, it is good practice to increase the mixing time to ensure that the fibers are fully dispersed within the mix. When using a batch plant, the dry mixing time should also be increased. Care must also be taken on the screen decks of batch plants. OGFC gradations consist of a large percentage of single size aggregates; therefore, override of the screen decks and hot bins can occur. In order to reduce draindown potential, OGFC should not be stored in surge bins or silos.

171 OGFC is transported to the project using normal haul trucks. However, because of the relatively high asphalt binder content and the use of polymer modified binders, several precautions are needed. First, truck beds will need a heavy coat of asphalt release agent to prevent the OGFC from sticking to the truck bed. The heavy coat should not be able to pool in the truck bed; however, pools of release agent will cool the OGFC and cause cold lumps. In order to maintain mixture heat, trucks should be tarped. Without tarping the truck, there is an increased potential for crusting of the mix during transportation. This potential is further increased during long haul distances. OGFC should only be placed on an impermeable pavement layer. Kandhal indicates that newly placed dense-graded HMA layers may be permeable to water; therefore, he recommended a uniform tack coat at an adequate application rate to fill and seal the surface. Kandhal references a FHWA recommendation of applying a 50 percent diluted slow-setting emulsion tack coat at a rate of 0.05 to 0.10 gallon per square yard. A slow setting emulsion was suggested because it is more likely to penetrate the surface voids more effectively. Kandhal states that the use of a material transfer device between the haul truck and paver is optional, but highly recommended. This type of equipment remixes any cold lumps that may result from transportation. A conventional paver is generally used to place OGFC mixes. A hot screed is important to prevent pulling in the mix. A conventional steel wheel roller is used to compact OGFC. Pneumatic tire rollers are not recommended because they tend to cause pick up. Vibratory rollers are also not recommended because they can break down aggregate in the compacted mix. Kandhal does state that vibratory rollers may be used at transverse joints. Because of the open aggregate grading and the use of polymer modified asphalt binders, OGFC mixes are generally harsh; therefore, handwork is often difficult. 1.40.5 Maintenance Practices General maintenance activities revolve around removing debris from the OGFC layer that builds up over time. Kandhal mentions three methods of cleaning an OGFC pavement; 1) cleaning with a fire hose; 2) cleaning with a high pressure cleaner; 3) cleaning with a specially made cleaning vehicle. Kandhal references a research project in Switzerland that suggested that cleaning with the high pressure cleaner was found most effective of the three methods mentioned above. Kandhal states that there has been little problem with winter maintenance in Europe. An OGFC layer does have different thermal properties than dense-graded HMA layers as the temperature will tend to be about 2˚C lower. Therefore, frost and ice will accumulate sooner and last longer on an OGFC surface. Because of this, Kandhal states that long