Construction and Maintenance Practices for Permeable Friction Courses (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.

6To assist the researchers in the successful completion of this project, a survey of transportation agencies was conducted during Task 2 of this research. A summary of the survey results are presented here at the beginning of this report; results from the survey are included within the state-of-art presented in subsequent chapters of this report. The term OGFC is gener- ically used within this chapter because OGFC encompasses all of the mix types covered by this survey. Where appropriate, PFC is used to describe aspects specifically related to these OGFC types. The objective of the survey was to obtain as much informa- tion as possible about methods of design (both mix design and structural design), construction, maintenance, safety, rehabil- itation, performance of PFC, and volume of use of PFC. As with any survey, it is seldom possible to get all agencies to respond. Though no response was received from 14 states, many of these states were contacted as the researchers had knowledge of their involvement with PFCs or OGFCs. Where experiences are contained in this report from agencies not completing the survey, that information was obtained from discussion with the agencies through telephone conversations or means other than the survey. The survey was set up as a web-based survey which allowed respondents the opportunity to fill in all the parts at once or different parts at different times instead of having to complete the entire survey before submission. The five parts in the sur- vey consisted of Part 1: General Use and Structural Design; Part 2: Mix Design; Part 3: Construction; Part 4: Maintenance and Rehabilitation; and, Part 5: Performance. Appendix A presents the survey. Because there are differences in the types of PFCs used around the world, a brief introduction that described the differences between PFCs and other OGFCs was provided to respondents. The introduction was intended to help respon- dents answer specific questions related to OGFCs, in general, or PFCs, specifically. The introduction given respondents follows: Open-graded friction courses (OGFCs) are specialty hot mix asphalt (HMA) mixtures that contain an open aggregate grading having a large percentage of coarse aggregates and low percent- ages of mineral matter. OGFCs are designed as a wearing sur- face to improve frictional resistance, reduce splash and spray, improve nighttime visibility, reduce hydroplaning and/or reduce tire/pavement tire noise. Within the overall category of OGFC, there are two predominant OGFCs used within the U.S.: Per- meable Friction Course (PFC) and Asphalt Concrete Friction Course (ACFC). Permeable Friction Courses are OGFC mixes that are specifically designed to have high in-place air void contents, typically in the range of 18 to 22 percent, for removing water from the pavement surface during a rain event. The term Asphalt Concrete Friction Course can be used for OGFC mixes that are not specifically designed for removing water from the pavement surface but rather are utilized to simply improve frictional resistance or to reduce tire/pavement noise. Though ACFCs have relatively high air void contents, generally near 15 percent, they are not designed to remove large volumes of water from the pavement surface like PFCs. The following paragraphs summarize the results of the survey. Each of the five parts is discussed individually. General Use and Structural Design General Use The first question posed to the agencies concerned their use of OGFCs. The researchers not only wanted to know which agencies used OGFCs, but also if the use of OGFCs was lim- ited geographically. Sixty four percent (64 percent) of the U.S. states responded to the survey along with four Canadian provinces, Austria, and Japan. Only 14 states plus British Columbia, Austria and Japan responded that they currently use OGFCs while five states plus Alberta and Yukon stated they did not use OGFCs. Interestingly, of the 32 states respond- ing, 13 states did not respond with a yes or no, but rather related that they either once used OGFCs and stopped (eight states plus Ontario), had a test/trial section planned (three states), C H A P T E R 2 Results of Agency Survey

used infrequently (one state), or was considering use under special development (one state). Of the 14 states indicating they use OGFCs and British Co- lumbia and Austria, nine states along with British Columbia and Austria described their OGFC mix as a PFC while only three states described their OGFC as an Asphalt Concrete Fric- tion Course (ACFC). Two states, Arizona and Iowa, did not feel that their OGFC description fit into either of these two categories, though no explanation was given. In the case of Arizona, it is suspected that because of their primary use of rubber modified asphalt and small maximum aggregate size gradation they did not feel that (ACFC) adequately fit their description. Interestingly, with the exception of California and Oregon, the use of PFC in the United States was limited to the southeastern states (Alabama, Georgia, Louisiana, North Carolina, South Carolina, Tennessee, and Texas). It is presently the understanding of the researchers that New Jersey has significant research underway using PFC. Whether the OGFC is described as PFC or ACFC, the vol- ume of use per year could be described as low compared to the overall usage of HMA. Of those states that described their OGFCs as PFCs, only Georgia and Texas were among the six states reporting usage greater than 100,000 tons. With five states along with British Columbia and Austria reporting usage less than 20,000 tons, one could assume that their PFC use at present is limited to smaller projects. This was reported for British Columbia. Of interest to the researchers were the types of roadway on which agencies use OGFCs. The respondents were given five roadway types in which to choose: urban freeway, urban arte- rial, urban collector, rural interstate, and rural primary high- way. Again responses were from the 14 states which stated they use OGFCs along with two Canadian provinces (one of which, Ontario, has discontinued use) and Austria. Urban freeway and rural interstate was listed by 75 percent of the respondents while 50 percent of the states and Austria listed rural primary highways. New Mexico, Nevada, and Texas were the only three states that listed OGFCs are used on all five roadway types. Of these states, only Texas classifies their OGFC as a PFC. The respondents were asked to define the factors involved in their selection of OGFC for a roadway. They were given five factors to choose from along with the opportunity to select “other” if the five categories did not meet their selection crite- ria. These five included policy, traffic volume, rainfall, design speed, and geometry. Over 78 percent of the agencies that re- sponded selected “policy” as the primary factor for selecting a roadway to receive OGFC with 50 percent selecting “traffic volume.” Though each factor was selected, it was noticed that as many as four of the five factors were selected by some states. On the other hand, since the majority selected “policy,” one would surmise that each state’s policy probably would include the other factors and the respondent did not feel the need to list them separately. The “other” category was selected by five agencies with five different factors listed, including safety, win- ter maintenance, monitor performance, wet weather accident history, and noise reduction. With the exception of Texas, each state that selected the “other” factor also made a selection of one or more of the researchers’ presented factors. Texas apparently only uses PFC pavements based on wet weather history. This response would parallel the response given for using OGFCs on all types of roadways. Austria indicated all five factors were included in the selection of OGFC. Tied to the factors involved in selection of OGFC for a roadway, the researchers were interested in knowing what limitations the agencies placed on where OGFCs are used. Though some agencies made mention of the number of lanes, traffic volume, etc., the researchers felt that these types of lim- itations were already established based on the roadway type previously selected. Many agencies made mention of the roadway design speed as a limitation, but there again this, for the most part, would have been previously set by the type of roadway. Design speeds mentioned ranged from greater than 40 mph to greater than 55 mph. Of particular interest were statements from California, North Carolina, and Oregon. California stated that OGFCs were not used on unsound pave- ments, in snow/icy areas, or at intersections. North Carolina had a concern with temperature in that a brine solution is re- quired in advance of freeze conditions. In Oregon, OGFCs are not used in curbed or urban routes, and their use is avoided in snow plow zones and areas with active slides. In Austria, OGFC is not utilized on roadways with steep slopes, areas with many curves or areas with many intersections. At least one agency commented on how costs have sig- nificantly increased with the use of polymers and fibers even though their use eliminated the raveling issues experienced with earlier OGFCs. The issue of safety seemed to be in the forefront of agencies’ general use of OGFCs. Comments on considera- tions being given to OGFCs use in areas with low skid numbers, history of frequent wet weather accidents and hydroplaning stand out. However, on the negative side of their use, were comments relative to the need for special maintenance prac- tices to maintain pavement porosity and noise characteristics. Structural Design Respondents to the survey were asked whether any struc- tural value was assigned to the OGFC layer. From the 50 states, four Canadian provinces, Austria, and Japan, responses were given by 26 agencies. Only 27 percent of the agencies respond- ing said that they assigned a structural value to the OGFC pavement layer. Of those assigning a structural value, over 70 percent stated that the structural value was estimated from layer coefficients. Texas uses a resilient modulus of 300 Ksi (2,000 MPa); however, the use of OGFC is limited to pave- ments that are already structurally sound. When asked whether a single lift thickness for all OGFC layers was specified, 17 out 7

of 22 responding agencies responded in the affirmative. One state, California, indicated that the thickness had to be at least 30 mm thick and 1.5 times the maximum aggregate size, so in effect they also established a minimum lift thickness. Many responded that the lift thickness of their OGFC layer was less than 1 in. (25 mm) which may explain the tendency for the large number of agencies not assigning a structural value to the OGFC layer. Mix Design Aggregates The agencies were requested to furnish their current grada- tion requirements for up to three different OGFC mix designs. Tables 1 through 20 present the gradation requirements from those states that presented data in the survey. Only Oregon had a PFC mix design whose maximum aggregate size exceeded 8 Table 1. Design gradation band for Alabama. Table 3. Design gradation band for Connecticut. Table 2. Design gradation bands for California. Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit AL 19 100 100 12.5 100 85 9.5 65 55 4.75 25 10 2.36 10 5 1.18 0.6 0.3 0.15 0.075 4 2 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit CA 19 12.5 9.5 89 78 4.75 37 28 36 29 2.36 18 7 18 7 1.18 0.6 0.3 0.15 0.075 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit CT 19 12.5 100 95 9.5 4.75 35 20 2.36 19 5 1.18 0.6 0.3 0.15 0.075 5 1

12.5 mm. Only California had a mix design whose nominal maximum aggregate size was 4.75 mm. The majority of the gradation requirements were for mixes whose maximum aggregate size was 12.5 mm or less. Respondents were requested to rank seven aggregate char- acteristics in order of importance: abrasion resistance, dura- bility, polish resistance, angularity, shape, cleanliness, and absorption. Graphical plots of the ranking of each character- istic are presented in Figures 1 through 7. Assigning the point values in order of importance (i.e., a value of 1.0 depicting the most important) that had been given to the respondent and the rankings assigned by the respondents, polish resis- tance was considered the most important characteristic of aggregates used in OGFC mixtures with absorption being considered the least important. A close second to polish resistance in importance was durability, followed in descend- ing order of importance by angularity, abrasion resistance, shape, and then cleanliness. Two states assigned point values 9 Table 4. Design gradation band for Delaware. Table 5. Design gradation band for Florida. Table 6. Design gradation bands for Georgia. Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit DE 12.5 100 100 9.5 98 88 4.75 42 25 2.36 15 5 1.18 0.6 0.3 0.15 0.075 5 2 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit FL 19 100 100 12.5 100 85 9.5 75 55 4.75 25 15 2.36 10 5 1.18 0.6 0.3 0.15 0.075 4 2 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit GA 19 100 100 100 100 100 100 12.5 100 100 100 85 100 80 9.5 100 85 75 55 60 35 4.75 40 20 25 15 25 10 2.36 10 5 10 5 10 5 1.18 0.6 0.3 0.15 0.075 4 2 4 2 4 1

10 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit IN 12.5 100 9.5 83 4.75 28 2.36 13 1.18 0.6 0.3 0.15 0.075 4 2 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit KY 19 12.5 100 100 9.5 100 90 4.75 50 25 2.36 15 5 1.18 0.6 0.3 0.15 0.075 5 2 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit LA 19 100 100 12.5 100 100 100 85 9.5 100 90 75 55 4.75 50 25 25 10 2.36 15 5 10 5 1.18 0.6 0.3 0.15 0.075 5 2 4 2 Table 7. Design gradation band for Indiana. Table 8. Design gradation band for Kentucky. Table 9. Design gradation bands for Louisiana. to each characteristic based on what appeared to be what they considered level of importance. When these states were elim- inated from the summary the ranking did not change. Not in- cluding the eliminated states, each characteristic with the ex- ception of the lowest ranking characteristics (cleanliness and absorption) received at least one vote from a respondent who felt it was the most important characteristic. Asphalt Binder Agencies were asked the grade of asphalt binder specified in their OGFCs. Of the 21 states and Austria responding, over 70 percent stated that they specified a PG 76-22 binder. Austria did not provide a performance grade, but did indicate a requirement for polymer modification. Where other grades

11 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit MO 19 100 100 12.5 100 85 9.5 75 55 4.75 25 10 2.36 10 5 1.18 0.6 0.3 0.15 0.075 4 2 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit MS 12.5 100 100 9.5 100 80 4.75 30 15 2.36 20 10 1.18 0.6 0.3 0.15 0.075 5 2 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit NC 19 100 12.5 100 100 100 85 9.5 100 75 100 75 75 55 4.75 45 25 45 25 25 15 2.36 15 5 15 5 10 5 1.18 0.6 0.3 0.15 0.075 3 1 3 1 4 2 Table 10. Design gradation band for Missouri. Table 11. Design gradation band for Mississippi. Table 12. Design gradation bands for North Carolina. different from PG 76-22 were specified, the grade reflected what was typically used in the state with other HMA mixes. For instance, California and Washington specify an AR 4000 or AR 8000. Northern tier states specified a softer grade binder than that used in the south or southwestern states where the climate is much warmer. Almost without exception, states specifying PG 76-22 called for the binder to be polymer mod- ified primarily with some type of elastic polymer (SBS) or rub- ber. Austria also utilizes an SBS polymer. Agencies were asked if other tests on the binder were conducted beyond those nec- essary in grading the binder sample. Though different type tests were specified (e.g., infrared trace, Phase angle, ductility, elastic recovery, etc.), the primary reason for running the test was to ensure they were getting an elastomer type polymer.

12 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit NE 19 100 100 12.5 100 95 9.5 80 40 4.75 35 15 2.36 12 5 1.18 0.6 0.3 0.15 0.075 3 0 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit NV 19 12.5 100 100 9.5 100 90 100 95 4.75 55 35 65 40 2.36 1.18 18 5 22 12 0.6 0.3 0.15 0.075 4 --- 5 --- Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit NY 12.5 100 95 9.5 56 40 4.75 30 20 2.36 14 6 1.18 12 4 0.6 9 3 0.3 0.15 0.075 5 2 Table 13. Design gradation band for Nebraska. Table 14. Design gradation bands for Nevada. Table 15. Design gradation band for New York. Stabilizing Additives When asked whether stabilizing additives were specified to reduce the potential for draindown, over 90 percent re- sponded in the affirmative. Fibers were listed as the predom- inant type stabilizing additive required by roughly 85 per- cent. The percentage of fibers required ranged from 0.2 percent to 0.5 percent, with the average being typically 0.3 percent. Approximately two-thirds of the agencies require a SBR, SBS or SB type polymer modifier while the other third did not specify a specific polymer type. Only one agency required crumb rubber in the asphalt binder while three others allowed its addition as an option. From the responses given, the overwhelming conclusion drawn is that some type of polymer-modified asphalt binder is specified with OGFC mixtures. Mineral or cellulose fibers

13 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit OH 12.5 100 100 9.5 96 85 4.75 45 28 2.36 17 9 1.18 0.6 0.3 0.15 0.075 5 2 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit OR 25 100 99 19 96 85 100 99 12.5 71 55 98 90 9.5 4.75 24 10 32 18 2.36 16 6 15 3 1.18 0.6 0.3 0.15 0.075 6 1 5 1 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit SC 19 100 12.5 100 85 9.5 75 55 4.75 25 15 2.36 10 5 1.18 0.6 0.3 0.15 0.075 4 2 Table 16. Design gradation band for Ohio. Table 17. Design gradation bands for Oregon. Table 18. Design gradation band for South Carolina. also are required to reduce the potential for draindown by most agencies. Mix Design Over 70 percent of the agencies responding stated that they used laboratory compaction during the design of OGFC mixes. There apparently is not a consistency within the agencies on the type of compaction method used with OGFC mix designs. Even with those agencies that have indicated significant use of OGFC mixtures, the method of compaction seems to be split between the Marshall method and the Superpave gyratory compactor. Where the gyratory method is used, the design number of gyrations typically was 50 though one agency

14 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit TN 19 100 100 12.5 100 85 9.5 60 35 4.75 25 10 2.36 10 5 1.18 0.6 0.3 0.15 0.075 4 2 Percent Passing Gradation 1 Gradation 2 Gradation 3 Sieve Size, mm Upper Limit Lower Limit Upper Limit Lower Limit Upper Limit Lower Limit TX 19 100 100 100 100 12.5 80 100 95 100 9.5 60 35 80 50 4.75 20 1 8 0 2.36 10 1 4 0 1.18 0.6 0.3 0.15 0.075 4 1 4 0 Table 19. Design gradation band for Tennessee. Table 20. Design gradation bands for Texas. utilizes 20 gyrations. The number of blows most often spec- ified when the Marshall hammer was used as the compaction method was 50. During the mix design procedure, all respondents, except for four, indicated that draindown testing was included as part of mix design. Three of the four agencies that indicated they did not include draindown testing had previously stated in their survey that they had stopped using OGFC pavements. Approximately 65 percent of the agencies using a drain- down test stated they used the draindown basket method. The remainder used a glass plate or other method. Where the bas- ket was specified, the requirement called for either 0.2 percent or 0.3 percent maximum draindown allowed. Only three agencies responding indicated that they included permeability testing within the mix design procedure. Appar- ently with the high design voids required for PFCs, most agencies do not consider it necessary to perform any type per- meability test; however, approximately one half of the agencies stated that they specified other laboratory tests during mix design. The majority of these tests consisted of the Cantabro Abrasion Loss test or some other type aggregate abrasion test. Some sort of moisture susceptibility testing was the next most often mentioned requirement with a boil test men- tioned by at least three agencies. Only five agencies required a freeze-thaw cycle with their tensile strength ratio test for moisture susceptibility with only one cycle required. Two agencies, neither of which had a moisture susceptibility test, either called for an anti-strip additive or required lime to be added to OGFC mixtures. Austria indicated they utilized loaded wheel testing. Construction Production Nearly two-thirds of the agencies reporting indicated that they specified OGFC by Standard Specifications, indicating that OGFC was part of their standard paving operations. The other third used Special Provisions. The majority of the agencies indicated that any typical plant used for HMA production also was used to produce OGFC. The plants typically being used were drum, batch or both. A batch plant was indicated as being the type of plant exclusively

15 0 1 2 3 4 5 6 7 Assigned Number 0 1 2 3 4 5 Fr eq ue n cy Ranking Abrasion Resistance Note: A lower number means a higher ranking 0 1 2 3 4 5 6 7 Assigned Number 0 2 4 6 8 Fr eq ue n cy Ranking Durability Note: A lower number means a higher ranking 0 1 2 3 4 5 6 7 Assigned Number 0 4 8 12 Fr e qu en cy Ranking Polish Resistance Note: A lower number means a higher ranking 0 1 2 3 4 5 6 7 Assigned Number 0 2 4 6 Fr eq u en cy Ranking Angularity Note: A lower number means a higher ranking Figure 1. Respondent ranking for aggregate properties – abrasion resistance. Figure 3. Respondent ranking for aggregate properties – polish resistance. Figure 4. Respondent ranking for aggregate properties – angularity. Figure 2. Respondent ranking for aggregate properties – durability.

used in only one state, Connecticut. A drum plant was exclu- sively indicated as the type used in Louisiana, Nebraska, and South Carolina. Mixing time in the plant for OGFC mixtures was the same for dense-graded mixtures by all agencies except for California, North Carolina, and Ohio. Only three agencies reported having a maximum and/or minimum temperature requirement for mixing in the plant. Responses to what the maximum and/or minimum mixing temperature specified varied among the agencies, but most stated that it was dependent upon the type/ grade asphalt binder that was used. The agencies were equally split on having a maximum silo storage time for OGFC mixtures. Those that had a time limit allowed OGFC to be kept in a silo from 1 hour to 12 hours. Agencies indicating earlier in the survey that they specified a PFC instead of other OGFC types required a shorter time held in the silo. Typically for these agencies the time held was no more than 2 hours. Transportation Only two agencies reported that they limited haul dis- tances. Ohio set its limit at a mileage distance (50 miles) with a minimum mixture temperature requirement (not pro- vided). South Carolina limited the haul distance by time (no more than 1 hour). Though the large majority of the agen- 16 0 1 2 3 4 5 6 7 Assigned Number 0 1 2 3 4 5 Fr eq ue nc y Ranking Shape Note: A lower number means a higher ranking 0 1 2 3 4 5 6 7 Assigned Number 0 2 4 6 8 Fr eq u en cy Ranking Cleanliness Note: A lower number means a higher ranking 0 1 2 3 4 5 6 7 Assigned Number 0 4 8 12 Fr eq ue nc y Ranking Absorption Note: A lower number means a higher ranking Figure 5. Respondent ranking for aggregate properties – particle shape. Figure 6. Respondent ranking for aggregate properties – cleanliness. Figure 7. Respondent ranking for aggregate properties – absorption.

cies (80 percent) did not require insulated trucks, nearly two-thirds did specify that the trucks had to be tarped. Only two agencies require that the haul trucks must be heated. One of these agencies had stated previously in the survey that they used OGFC infrequently while the other had dis- continued use. Regardless of the restrictions or limitations placed on the haul distance and truck, all but two of the re- porting agencies responded that they had a minimum mix temperature when the truck reached the paving site. This minimum temperature varied from agency to agency. Some reported a fixed minimum temperature with none below 225°F. The temperature went up from there to as high as 300°F. Other agencies set the minimum temperature at 20 to 30°F below the job mix formula or the target compaction temperature. Release agents are allowed by all agencies reporting except for one, Nebraska. Most indicated that the types allowed had to be non-petroleum (no fuel oil or diesel). Also, most agencies indicated that they had an approved/qualified products list of acceptable agents. Laydown/Compaction Less than 30 percent of the reporting agencies require a material transfer vehicle (MTV) in the paving train for plac- ing OGFC. Only one agency, Connecticut, stated they did not require a tack coat to be placed prior to placing the OGFC. The type and application rate of tack coat varied among the agencies. The types of tack coat used included different types of emulsions that included RS-1, RS-2, CRS-2, CRS-2P, SS-1, SS-1h, CSS-1, and CSS-1h. Though the ma- jority of the agencies using tack called for some type of emulsion, some used performance-graded binder such as PG64-22. Other agencies did not report the specific type of tack used. The application rate ranged from a low of 0.04 gal- lons per square yard to an amount as high as 0.2 gallons per square yard. The use of a calibrated distributor truck to apply the tack was equally split with half of the agencies stating that they did not require one while the other half reported that they did require one. When asked whether other techniques to ensure an impermeable underlying layer other than a tack coat were used, only two agencies re- ported that they did. One state, Connecticut, stated that it only placed the OGFC on a dense-graded HMA. Texas stated that it sometimes required an underseal if the existing pave- ment was susceptible to water intrusion; however, it usually did not require the underseal if they were paving on a new layer of hot mix. There was only one agency that did not have a minimum specified air and/or surface temperature for placing OGFC. Once again, the minimum temperature varied from agency to agency. The lowest minimum temperature reported was 50°F and went up from there to as high as 70°F. In nearly every instance, the temperature reported was ambient air temper- ature with agencies reporting that this temperature had to be rising. Some agencies specifically stated that the temperature was a surface pavement temperature. At least three agencies indicated that the minimum temperature was dependent on the type binder used in the OGFC with the higher perfor- mance grades (i.e., PG76-22 versus PG64-28) requiring a higher minimum temperature. When asked how they specified compaction, all except one reported that they used a method specification. This one state apparently read the question as “Do you . . . ?” rather than “How do you . . . ?” as they answered with a “no” indi- cating that they did not specify compaction. This response was verified by their answer to the next question on whether a certain type of roller was specified for compaction. Since they responded that they did not specify a certain type of roller, this was followed by their stating the contractor was required to make two passes with a steel wheel roller only. Therefore, 100 percent of the agencies reported using a method specification while requiring a certain type roller for compaction with no specific density requirement. Without exception, the roller specified was a steel wheel roller though at least one agency used the term of non-pneumatic. Some agencies went as far as to set a minimum weight for the roller. Others set the number of passes. One agency required a minimum of two double-drum steel wheel rollers for com- paction. Without exception, when specifically addressed by an agency, the steel wheel roller had to be operated in the static mode. No agency indicated that vibrating was allowed or required with some going as far as stating that if the con- tractor used a vibratory roller it had to be operated in the static mode. The practice of tacking of the vertical face of longitudinal joints was not generally used. Only one agency indicated that they allowed tacking of the vertical face of longitudinal joints. Quality Control/Quality Assurance Gradation and binder content were the two primary tests required for quality control/quality assurance (QC/QA) of OGFC mixtures. Both of these tests were included in all the state’s QC/QA programs except for two agencies and each agency included one or the other test. There were as many as 12 other tests mentioned by some of the agencies as part of their QC/QA testing. Some of these included modified Lottman test (for stripping), permeability, temperature, draindown, boil test (for stripping), smoothness, moisture content (mix), percent air voids, gyratory density and Rice specific gravity. Of these, only draindown, air voids, and Rice gravity were included by more than one agency. 17

General Construction Issues Agencies were asked what materials were used for mark- ings placed on OGFC pavements. Apparently the agencies did not consider the need to specify anything different for markings on OGFC pavements than what they were cur- rently using on dense-graded HMA. Many of the respondents specifically stated that typical pavement markings were used. The most frequently listed markings included waterborne paint and thermoplastic. Some of the northern tier agencies mentioned epoxy. One agency, North Carolina, has a con- cern about the effects that thermoplastic may have on the ability of OGFC to properly drain. Though Massachusetts did not respond to the survey, research by the team found that the chief engineer of Massachusetts signed an engineer- ing directive dated June 16, 2005 that prohibited the use of thermoplastic markings and required the use of epoxy mark- ings on all OGFC pavements. Further, the directive required the use of slotted in pavement reflectorized pavement mark- ers in place of snowplowable raised reflectorized pavement markers. Agencies were next asked if rumble strips were constructed at the OGFC pavement edge. Again the results were nearly equally split with slightly more states constructing rumble strips at the pavement edge. Finally, agencies were asked if they had experienced any dis- tresses related to pavement markings and rumble strips. Only three agencies responded affirmatively. One agency response dealt with pavement markings. Another centered on rumble strips while the third mentioned both. Connecticut indicated that thermoplastic line stripes caused raveling in the OGFC. Nebraska stated that pavement markings may restrict drainage and that grinding in rumble strips pulls out too much aggregate. Texas, on the other hand, stated that rolled in rumble strips did not work as well as grinding in rumble strips. With nearly a 50 percent split on the use of rumble strips, the researchers felt that further investigation was warranted on rumble strip usage. The following is background information on rumble strips along with comments from telephone conversations with individuals from South Carolina and New York on the subject. Rumble strips in the United States are mainly installed on highway shoulders as a countermeasure against run-off-the- road accidents. There are four types of shoulder rumble strips: milled, rolled, formed, and raised. They differ primarily in how they are installed, their shape and size, and the amount of noise and vibration produced. Rolled and milled rumble strips are used most often on asphalt highway shoulders. The first rumble strips were installed in 1955 in New Jersey and they were called “singing shoulders.” They were wavy bumps installed at the concrete paved shoulder of a bridge. In the 1960s, many states adopted rumble strips of various designs. Rolled rumble strips were developed in the 1970s; the Pennsylvania Turnpike Commission created milled rum- ble strips in 1990 (1). The new interest in rumble strips seems to be at least partly the result of a FHWA notice is- sued in 1986 (see U.S. Department of Transportation Federal Highway Administration, Notice – Shoulder Texture Treat- ments for Safety, N 7560.9 April 28, 1986). Highway agen- cies often use all types of shoulder rumble strips, depending on the need and the material. In the United States, a method of milling paved roads was developed in the 1990s and it spread rapidly. This method employs a groove pattern that can be installed intermittently or continuously. The groove pattern, depth, width, shape, and spacing also may change with the road agency. The milled rumble strip method seems to be the preferred choice of most states. Further informa- tion on rumble strips can be found in NCHRP Synthesis Report No. 191 published by the Transportation Research Board (2). The purpose here is not to debate the use or nonuse of rumble strips or when used what specific design criterion should be employed by the highway agency; however, there may be some concern as to whether rumble strips can be used with PFCs. Contacts were made to a number of agencies about their practice on installing rumble strips. For those states that daylight the PFC just outside the pavement edge strip, the installation of rumble strips in PFC was not an issue. Other states, such as South Carolina, have recently begun to daylight their PFC pavements at eight feet (8’) off the edge line. Con- versation with Tim Linberg in South Carolina’s Construction Office on January 5, 2006 revealed that their policy was to use milled rumble strips at two feet from the edge stripe. Many early concerns and problems with rumble strips when states first began using them involved curves. States, such as Oregon, experienced problems with the rumble strips, but these issues or problems seemed to center more on the installation pro- cedure and where the rumble strips were located and not that the rumble strips should have been placed or were erroneously placed in the OGFC. One of the early issues involved in using rumble strips with OGFC included premature deterioration of the asphalt in the rumble strip area under winter main- tenance activities. This early concern centered primarily on snow plows with chains operating in the rumble strips. It was determined that this problem could be overcome by place- ment of the rumble strip and by going to milled rumble strips instead of rolled. It was felt that the lack of compaction of the asphalt of the rolled rumble strips attributed to the asphalt failure around the rumble strip. A discussion of the issues that Oregon experienced was reviewed (2). Further discussion on rumble strip usage with OGFC pavements was conducted on January 5, 2006 by telephone with Mr. Emmett McDevitt. Mr. McDevitt is a Transportation Safety Engineer for the FHWA New York Division Office. Mr. McDevitt did not know 18

of any issues that would prevent the use of rumble strips with PFCs. Further, he knew of no failures that could be attrib- uted to their use. It should be pointed out that there has been only a modest amount of PFCs of the type addressed in this research (designed to have air voids of at least 18 percent) constructed in the United States. The majority of the construction has been in the warmer regions of the country and therefore all issues associated with winter maintenance may not have come to light. Typically milled shoulder rumble strips are placed at least 18 in. (457 mm) or greater from the edge line with approxi- mately a 0.5 in. (13 mm) groove depth and on 12 in. (305 mm) center to center groove spacing. PFC mixtures are typically constructed in thicker layers than conventional OGFC (1.0 to 2.0 in. as opposed to 1.0 in. or less). For PFC pavements placed at these thicknesses, the opinion is that milled rumble strips can be constructed with no adverse problems different from those encountered or expected where rumble strips are placed on other type pavements. However, in colder climates where agencies use sand and salt, the need to perform mainte- nance that includes cleaning of the PFC pavements is an issue that will need to be addressed. Maintenance and Rehabilitation General Maintenance (Non-Winter Related) When asked to give their most common general mainte- nance issues for OGFC, the responses could be placed in one of two categories: unclog/clogging and raveling/delamina- tion, with the latter being listed only slightly more than the first. If OGFC pavements have a tendency to clog, then un- clogging them would be a concern. Austria indicated that they utilize special equipment to unclog OGFC pavements. However, when asked if any regularly scheduled mainte- nance activities were scheduled for OGFC pavements, not one agency stated that they did. Further, when asked if they employed maintenance activities to unclog OGFC pave- ments, not one agency responded that they did. Only one agency responded affirmative to the question whether any field test was used to identify when general maintenance ac- tivities were required. Washington responded that they used a high-speed van, and these vans likely are used to identify surface distresses. Patching OGFC pavements was performed by just over 50 percent of the responding states. Without exception, those agencies that patch OGFC pavements use a dense-graded HMA. Texas indicated that for small patches they used a pro- prietary OGFC patching mix, but for larger patches used the dense-graded HMA. Georgia stated that on interstates with significant patching being required, they just replaced with OGFC, implying milling and complete replacement. Winter Maintenance Only 22 percent of the agencies completing the survey stated they used any type of weather prediction system for winter maintenance activities. Nevada stated that they used a RWIS system for snow plow and anti-icing activities. Weather condi- tions that would trigger winter maintenance most frequently were snow, freezing rain or sleet, and frost. Interestingly, nearly one-third of the agencies responding to the survey passed on this question. Calcium chloride and sodium chloride were the two most often mentioned ice control chemicals used by the states. Brine along with sand or aggregate screenings also was listed. One- third of those responding stated that they employed anti-icing methodologies. To get a feeling of the spread rates used for controlling ice on roadways, respondents also were asked to give their spread rates for both dense-graded mixtures and OGFC mixtures. For the most part these questions were not answered. Most of the agencies gave no response at all or stated that the rates were unknown. Austria indicated that 20 to 50 percent more chem- icals are required for OGFC pavements than dense-graded. Others gave a rate for dense-graded, but did not give any rates for OGFC. Of the two agencies that did give responses to both questions, Kentucky and Oregon, the answers varied extremely. Kentucky gave values for dense-graded pavements, and then stated that the same spread rate was used for OGFC. Oregon on the other hand stated that the rate varied accord- ing to conditions. They indicated that a single stream nozzle was used on dense-graded surfaces and a triple stream nozzle was used on OGFC, implying triple the rate. One other agency did respond that the rate would be higher for OGFC. Rehabilitation Raveling was the answer given most often for the type of problem that triggered rehabilitation. This answer showed up in different forms from the respondents, but essentially meant the same thing. In lieu of the term raveling, some responded with pot holes, delamination, and loss of ride quality. The loss of permeability/noise characteristics was mentioned by one agency. Without exception, when problems arose that called for rehabilitation, agencies milled and replaced the pavement with a new surface. Only four agencies stated that they were aware of any known or perceived problem with maintenance/rehabilitation techniques. Georgia questioned the placing of OGFC on milled surfaces and stated they were looking at micro-milling for those instances. On the same line, Texas stated that some en- gineers were concerned about how to mill and replace PFC once it reaches the end of its design. One northern tier agency stated that PFC pavements were more expensive to maintain. 19

Oregon stated that capping of OGFC pavements with dense- graded HMA or chip seals might cause damage and reduced performance. They stated that they had some inlay projects exhibit isolated shoving spots due to failure of the old layer underneath the OGFC. They felt that the cold planning broke the seal of the pavement mat and the new OGFC pavement allowed water to infiltrate and damage the underlying layer. Performance In the final part of the survey, respondents were asked about the performance aspects of OGFC pavements. The first ques- tions dealt with the estimated average service life of OGFC pave- ments. Responses could not be grouped based on geographical location. Responses ranged from less than six years to as high as 15 years. Not including those that responded less than six years, each respondent gave a two to three year range. Most of the respondents gave the service life for OGFC pavements as eight to 10 years. This was also the range where the average of all respondents fell. When asked about the common distresses in OGFC pave- ments, responses were virtually the same as those given for the first question under Part 4, Section C, on those problems that triggered rehabilitation activities. The most common response was raveling or delamination. Cracking also was mentioned as a distress. Respondents were asked to rank seven performance charac- teristics related to OGFCs in terms of their importance. These performance characteristics were improved wet weather fric- tion, reduced splash/spray, smoothness, noise reduction, reduced hydroplaning, improved nighttime visibility, and improved wet weather visibility. Assigning the point values in order of importance that had been given to the respondent and the rankings assigned by the respondents, improved wet weather friction was considered the most important perfor- mance characteristic of OGFC pavements with reduced splash/ spray and reduced hydroplaning tied for second most impor- tant (Figures 8 through 12). Even though noise reduction is one of the leading characteristics of PFC pavements in Europe, noise reduction along with smoothness was considered the least important by the respondents. Improved wet weather visibility fell in the middle of the rankings. Each of the per- formance characteristics received a number one ranking from at least three or more of the respondents. Only improved wet weather friction and reduced splash/spray did not get a last place vote in the rankings. Obviously, each of the perfor- mance characteristics of OGFC pavements is important and beneficial. Typically, agencies use the same tests/equipment for measuring performance quality characteristics on OGFC pave- ments as they do on dense-graded pavements. This is particu- larly true for field tests with friction testing being by nearly all 20 0 1 2 3 4 5 6 7 Assigned Number 0 2 4 6 8 Fr eq u en cy Ranking Improved Wet Weather Friction Note: A lower number means a higher ranking 0 1 2 3 4 5 6 7 Assigned Number 0 2 4 6 8 10 Fr eq u en c y Ranking Reduced Splash/Spray Note: A lower number means a higher ranking Figure 8. Respondent rankings of performance characteristics – wet weather friction. Figure 9. Respondent rankings of performance characteristics – reduced splash/spray.

21 0 1 2 3 4 5 6 7 Assigned Number 0 2 4 6 Fr eq u en cy Ranking Noise Reduction Note: A lower number means a higher ranking 0 1 2 3 4 5 6 7 Assigned Number 0 2 4 6 8 Fr e qu e n c y Ranking Reduced Hydroplaning Note: A lower number means a higher ranking Figure 10. Respondent rankings of performance characteristics – noise reduction. Figure 11. Respondent rankings of performance characteristics – reduced hydroplaning. respondents. Only one agency stated that they ran a field water permeability test to determine if the pavement was clogged by fines. Two agencies indicated that they perform noise testing of some type. Only two agencies stated that they had conducted life cycle cost analysis for OGFC pavements. Oregon stated that though life cycle costs were used to determine pavement type selection and rehabilitation options, wearing course selection was based on other factors besides just economic considerations. Survey Summary In general, results from the survey can be considered a suc- cess. A significant amount of information was obtained within the five parts of the survey. This section summarizes results from the survey. The majority of agencies within the United States have used OGFCs. Some agencies have ceased using these mix types because of performance problems in the past. Currently, most of the states within the United States do not utilize OGFCs. Those states that do utilize OGFCs are generally located in warmer weather climates; however, some agencies within the colder climates are successfully using OGFCs. Most of the agencies reporting the use of OGFC state that their mix type would classify as a PFC using the definitions provided 0 1 2 3 4 5 6 7 Assigned Number 0 2 4 6 8 Fr eq ue n cy Ranking Improved Wet-Weather Visibility Note: A lower number means a higher ranking Figure 12. Respondent rankings of performance characteristics – improved wet-weather visibility.

in the survey instructions. OGFCs are typically used on high- speed highways with the majority of usage being on interstate type pavements. The decision for using OGFC mixes as a wear- ing surface is generally a policy decision. Selection of OGFC mixes is likely due to the safety benefits realized when using these mix types. This was confirmed within the performance characteristics section of the survey when the respondents indicated that improved wet weather friction was considered the most important performance trail of OGFCs. Most agencies do not assign structural values to layers of OGFC. For the most part, agencies utilize a single lift thick- ness for constructed OGFC layers. Most commonly the lift thickness is less than 1 in. (25 mm); however, thicker lifts are sometimes specified in the United States and internationally for PFC layers. Gradations used for OGFCs vary from agency to agency. The maximum aggregate size (mas) of OGFC mixes are gen- erally 12.5 mm. Agencies reporting the use of PFCs generally specify gradations coarser than those reported as ACFCs. With respect to aggregate characteristics, the survey respon- dents identified polish resistance as most important, followed closely by aggregate durability. Other aggregate properties deemed important were angularity and abrasion resistance. Most agencies specify that ground tire rubber be used in OGFCs. Both elastomeric polymers and ground tire rubber are used in OGFCs. For states that reported the use of PFCs, polymers are generally specified. Another material common to PFCs is fiber stabilizers. Roughly 85 percent of the re- sponding agencies indicated that fibers are required. Mineral and cellulose fibers are the most commonly specified fiber types. Fiber dosage rates ranged from 0.2 to 0.5 percent for the two fiber types. During the design of OGFC mixtures, 70 percent of respon- dents stated that a laboratory compactive effort is utilized. This would indicate that the remaining 30 percent use a recipe method of designing OGFCs. The compaction methods men- tioned were the Marshall hammer and Superpave gyratory compactor. When using the Marshall hammer, 50 blows per face was the most common compactive effort. For the Super- pave gyratory compactor, 50 gyrations was the prevalent design compactive effort. Most of the agencies reporting the use of PFCs utilized the Superpave gyratory compactor. A number of other tests were identified as being used during mix designs and include: draindown, moisture susceptibility, Cantabro Abrasion Loss, and permeability. Nearly two-thirds of the agencies specify OGFC by Stan- dard Specification, indicating that OGFC is part of their stan- dard paving operations. Both batch and drum plants have been used to successfully produce OGFC mixtures. None of the responding states required an increased mixing time for OGFCs. Roughly half of the responding agencies do limit silo storage times with limits ranging from 1 to 12 hours. However, 22 agencies reporting the use the PFCs generally specify a maxi- mum of 2 hours storage time. Haul distances are generally not specified; rather, most agencies specify a minimum mix tem- perature when the haul trucks reach the paving site. This min- imum temperature is typically based upon the grade (type) of asphalt binder used in the mix. Release agents are allowed for truck beds by all agencies. The type and rate of tack coats vary by agency. Both emul- sions and neat asphalt binder have been used successfully as tack coats. Only one agency did not have a minimum spec- ified air and/or surface temperature for placing OGFC. As would be expected, the specified minimum temperature var- ied by climatic region. One constant related to the specified minimum temperatures was that the temperature must be above the minimum and rising. Compaction of OGFC layers is conducted using static steel wheel rollers. This was consis- tent with all responding agencies. Some agencies do specify the number of passes during compaction. Two QC/QA tests are relatively constant among the respond- ing agencies: asphalt content and gradation. Other QC/QA tests mentioned by the agencies included tensile strength ratio, per- meability, temperature, draindown, air void content, and Rice specific gravity. Another construction issue addressed in the survey was pavement markings. Unfortunately, a definitive method (or material) for use with OGFCs was not found. The methods appeared to be agency specific. Likewise, a definitive method of constructing rumble strips was not identified by the survey. Primary maintenance activities conducted by the respond- ing agencies were related to winter maintenance. Respondents acknowledged the need for general (non-winter) maintenance related to clogged OGFCs; however, no agencies currently uti- lize a routine maintenance program for evaluating the perme- ability of OGFC layers. For general maintenance, most agencies do monitor the existence and manifestation of raveling, crack- ing, and delamination. For winter maintenance, the primary piece of information observed from the survey was that more deicing chemicals are needed for OGFC layers than for dense-graded layers. Calcium chloride and sodium chloride were identified as the two most common ice control chemicals used for OGFCs. One-third of the respondents did state that they used anti- icing methodologies prior to winter events. For small distressed areas, the responding agencies indi- cated that patching with dense-graded mix was the typical method of rehabilitation. When distressed areas are larger, the typical strategy is to remove the existing OGFC by milling and replace with a new OGFC. One state did question placement of OGFC over a milled surface, likely because of the grooves left in the existing pavement. This agency is investigating the need for micro-milling to provide a smoother surface for plac- ing OGFC.

The final portion of the survey was on performance. Most of the respondents indicate that the service life of OGFCs was 8 to 10 years. The most common distresses encountered with OGFCs are raveling and delamination. Cracking also was mentioned as a distress. Respondents were asked to rank seven performance characteristics related to OGFCs. Of these performance characteristics, wet weather friction was identi- fied as the most important performance characteristic. Reduced splash/spray and reduced hydroplaning tied for the second most important performance characteristic. Interestingly, noise reduction and smoothness were considered the least impor- tant. This overwhelmingly indicates that the safety benefits related to OGFCs are the primary reason for selecting OGFC layers. 23