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.
73 Throughout the history of using open-graded mixes as wearing layers, there have been two predominant performance- related problems: raveling and delamination. These two prob- lems led to moratoriums on the use of OGFCs by many state highway agencies during the 1980s (7, 14). These problems have been noted not only in the United States, but also in Europe (17, 21, 51). This chapter describes the performance of PFCs from around the world. The first section describes the distresses encountered with PFC layers, the second section describes the performance life of PFCs, and the final section discusses performance measures for PFCs. Typical Distresses with PFC The Long Term Pavement Performance Program (LTPP) has identified a number of distresses related to HMA layers (74). Within this document, distresses are categorized accord- ing to the following general distress types: cracking, patching/ potholes, surface deformation, surface defects and miscella- neous distresses. Table 40 lists the distresses defined by LTPP within each of these categories. Of the distresses listed in Table 40, only raveling has been reported as common to PFCs. Huber (7) states that OGFCs typically fail by raveling. Molenaar and Molenaar (38) have described two forms of raveling: short-term and long-term raveling. Short-term raveling is caused by intense shearing forces at the tire/pavement interface that occurs within newly placed PFCs. Pucher et al. (68) state that short-term raveling generally occurs quickly once the flow of traffic begins. Con- ditions that enhance the potential for short-term raveling include placing the PFC at too low of a temperature, incom- plete seating of the aggregates during compaction and drain- down (areas lean in asphalt binder). Long-term raveling was described by Molenaar and Molenaar (38) as being caused by long-term segregation of the asphalt binder from the aggre- gates due to gravity. As the asphalt binder drains from the coarse aggregate structure due to gravity, the aggregates near the surface of the layer are underasphalted. The action of traffic can dislodge the aggregates, resulting in raveling. It should be stated that the long-term draindown of the asphalt binder due to gravity was mostly encountered in PFC mixes that did not include modified asphalt binders. Pucher et al. (68) state that up to a life of 5 to 10 years, PFCs deteriorate slowly. After this time, the rate of deterioration increases. Raveling is the distress most commonly observed due to this increase in degradation. As stated above, delamination is the other distress most commonly associated with PFCs. Delamination of PFC layers could, however, be construed as potholes once the layer has been removed by traffic. The raveling and delamination problems that have plagued OGFC mixes in the past can likely be traced back to mix design, specifically materials selection, and construction problems. Permeable friction courses have an open gradation with a relatively low percentage of material passing the No. 200 (0.075 mm) sieve. Because of the open grading, there is very little aggregate surface area which results in a relatively thick film of asphalt binder coating the aggregates. At typical HMA production/construction temperatures, the heavy film of asphalt binder had a propensity to drain from the aggregate skeleton (7). Because of the draindown issues, a typical remedy was to reduce either the asphalt binder content or the mixing and compaction temperatures during production/construction (4). Reduced asphalt binder contents meant that the OGFC mixes were underasphalted which would increase the poten- tial for raveling. The reduction in temperature increased the viscosity of the asphalt binder which assisted in preventing the asphalt binder from draining from the aggregate skeleton. However, this reduction in temperature also led to the increased potential for raveling and delamination. When production temperatures of the PFC are reduced, all of the internal moisture within the aggregates is not removed. Moisture remaining within the aggregates after plant mixing increases the potential for the asphalt binder stripping from C H A P T E R 1 0 Performance of Permeable Friction Courses
the aggregates leading to the aggregates being raveled out due to traffic (7). Reduced mixing temperatures also resulted in reduced compaction temperatures. Mixture delivered to the roadway that was not at an appropriate compaction temper- ature had difficulty bonding to the tack coat placed on the existing roadway surface. This resulted in an inadequate bond between the OGFC and underlying layer. The lack of an adequate bond increased the potential for delamination problems (4). Evidence that raveling and delamination problems of the past were related to mix design and production/construction practices was provided by Kandhal and Mallick (8). Based upon a 1998 survey, they stated that highway agencies that had experienced good performance with OGFCs were utilizing polymer-modified asphalt binders and relatively high asphalt binder (by using fibers and/or relatively open gradations). The combination of modified asphalt binders and fibers helped hold the asphalt binder on the PFCâs aggregate skeleton, min- imizing the potential for draindown. Without the potential for draindown (and relatively high asphalt binder contents), there was no need to lower mixing and compaction temper- atures which minimized the potential for both raveling and delamination. Though raveling and delamination are the two most com- mon distresses listed in the literature, other distresses have been mentioned. Rogge (66) conducted a survey of main- tenance supervisors from the Oregon Department of Trans- portation (ODOT). One question within the survey concerned typical distresses encountered on PFC pavements. Figure 41 illustrates the typical distresses encountered in Oregon on PFC pavements as reported by 78 respondents to the survey. Within Figure 41, the maintenance engineers were requested to rank the various distresses by their frequency using a ranking system of 1 to 4. The higher the ranking, the more frequent the dis- tress encountered. Based upon the survey, tire stud rutting was considered the most common distress. Raveling was the sec- ond highest rated distress (icing problems is considered a win- 74 Cracking Patching and Potholes Surface Deformation Surface Defects Miscellaneous Fatigue Block Edge Longitudinal Reflection Transverse Patch Deterioration Potholes Rutting Shoving Bleeding Polished Agg. Raveling Lane-to-shoulder drop-off Water Bleeding and Pumping Table 40. LTPP defined distresses for HMA Pavements (74). 1.00 1.50 2.00 2.50 3.00 3.50 4.00 Tire Stud Rutting Icing Problems Raveling Gouging/scarring (snow-plow, etc.) Deformation Rutting Clogging Potholes Fat spots/bleeding becomes problem Cracking due to inadequate structure Noisy ride Stripping Bumpy ride Reflective cracking Thermal cracking eraRneeSreveN evisavrePderettacS Figure 41. Results of 2001 survey of ODOT maintenance supervisors (66).
ter maintenance issue). Other distresses that rank closer to scattered than rare included gouging/scarring, deformation rutting, and potholes (clogging is considered a general main- tenance issue). Because of the environment in Oregon, the existence of tire-stud rutting is not unexpected. However, tire-stud rutting should not be considered the same as the traditional rutting seen on typical dense-graded HMA pavements (plastic defor- mation). Tire-stud rutting as described by Rogge (66) is likely raveling within the wheel paths. Studded tires can dislodge aggregate particles in the wheel path giving the appearance of classical rutting. Roggeâs report (66) was the only reference found in the liter- ature that listed rutting as a distress on PFC pavements. Several papers/reports from Europe list resistance to permanent defor- mation as a benefit of PFC pavement layers (5, 2). Permeable friction courses should generally not be associated with plastic deformation rutting. Similar to SMAs, PFCs have a very coarse gradation that results in stone-on-stone contact (40). Because of the stone-on-stone contact, PFCs should not rut due to plas- tic deformation unless there are mix design or construction problems. The only other distress found in the literature was not specif- ically related to OGFC/PFC mixtures; rather, it was the occur- rence of stripping in layers underlying the OGFC/PFC surface. Huber (7) states that open-graded mixes can change the mois- ture balance within a pavement structure. PFCs or OGFCs can create a moist microenvironment at the surface of the under- lying layer. When this exists, the increased humidity created by the moist microenvironment can retard evaporation of water from the underlying layer. This, in essence, traps water within the underlying layer. When PFCs become clogged, the under- lying layer may even become wetter. Therefore, if the HMA mixture underlying the PFC layer contains materials sus- ceptible to moisture, then stripping of the underlying layers may occur. Performance of PFC According to Huber (7), the performance of PFC (or OGFC) pavements in general can be put into one of two cat- egories: performance life and service life. The category of performance life is used to describe the length of time an OGFC pavement maintains its beneficial characteristics. With respect to PFC pavements, these characteristics would include permeability (reduction in potential for hydroplan- ing and splash and spray and improvement in pavement marking visibility) and the ability to reduce tire/pavement noise. Service life describes the length of time that a PFC pavement maintains its frictional properties and smoothness. Structural failure of the PFC also would be included in ser- vice life. Service Life Of the two categories of performance, service life generally will be longer. Service life generally relates to the time that a PFC layer needs to be rehabilitated. The vast majority of reports/papers suggest that PFC pavement layers will have an average service life of about 10 years, though longer periods have been cited. A number of European countries, including the Netherlands (17), Switzerland (21), and Spain (16), indi- cate that the service life of PFC pavements is approximately 10 years. Similarly, Australia also has indicated 8 to 10 years of service life (19). In the United States, a survey of state highway agencies on OGFCs conducted in 1998 by Kandhal and Mallick (8) showed that 73 percent of the state agencies obtained an estimated average service life of greater than 8 years (Figure 42). Forty-three percent of the state agencies estimated an average service life of greater than 10 years. The major- ity of the agencies indicating an average service life greater than 10 years were utilizing OGFC mixes that would clas- sify as PFCs. No specific literature was found that presented a research approach that followed the frictional properties or smooth- ness of a PFC layer until the end of the service life. Survey results depicted in Figure 42 likely reflect more of an issue with smoothness than friction. Smoothness would be affected by raveling problems associated with PFC pavements and ravel- ing was cited by the vast majority of papers/reports reviewed as the primary performance problem with OGFC layers. Addi- tionally, delamination, which also has been labeled as a major problem with OGFC layers (6), also would negatively affect smoothness. Figure 41 showed distresses observed in Oregon on OGFC pavement layers as well as the frequency in which those dis- tresses are encountered (66). Based on this figure, most of the distresses that had a frequency closer to scattered than rare would affect smoothness. As tire-stud rutting, raveling, gouging/scarring, deformation rutting and potholes increase, smoothness would decrease. As stated previously, Pucher et al. (68) indicated that up to 5 to 10 years, PFCs deteriorate slowly. After this time, the rate of deterioration increases. Raveling is the distress most commonly observed due to this increase in degradation. Similar to smoothness, no specific literature was found that followed the frictional properties of a PFC pavement layer from construction till the end of the service life. The literature does suggest that the frictional characteristics of PFC layers are relatively low (but acceptable) immediately after construction (10, 21, 51). PFCs are intentionally designed to include a rel- atively high asphalt binder content. After production and placement, aggregates within the PFC layer will be coated with a thick film of asphalt binder. This thick film of asphalt binder prevents a vehicle tire from adhering to the aggregates 75
(microtexture) at the surface of the layer (51). Greibe (24) stated that when the wheels lock during a braking action, the friction created between the tire and pavement surface begins to melt the asphalt binder coating the aggregates which hinders friction. This is only true when wheels are locked. When an anti-lock braking system is used, the baking distance on porous asphalt is similar to that of dense-graded HMA. Some litera- ture indicates that it can take 3 to 6 months for the asphalt binder film to wear from the aggregates at the surface of the layer (24, 51). However, a research study in Georgia indicated that the asphalt binder layer wore off within 2 weeks (10). Table 41 illustrates the results of skid trailer friction testing conducted on six OGFC test sections over a 3.5 year time period just south of Atlanta, Georgia (10). All of the sections listed within Table 41, except the âStandard OGFCâ section, are considered PFC sections. Within this table, the first friction tests were conducted the day after construction. These mea- surements were all relatively low compared to the subsequent test dates. The data clearly shows that once the asphalt binder film has worn from the aggregates, friction will increase. There are two primary reasons for the good frictional prop- erties of an PFC layer: permeability and macrotexture. Because of the high percentage of air voids associated with PFC layers, water will readily drain from the pavement surface into the interstitial voids of the PFC layer. Water that drains into the PFC layer is not available to be trapped between the vehicle tire and pavement surface in the form of water films, thus improv- ing wet weather friction (15, 19, 20). Because of the open grad- ing of PFC mixtures, these mix types result in a relatively high amount of macrotexture (13, 19, 75). Table 42 presents macro- texture measurements from a research study conducted in Indiana that compares the texture of a porous friction test section to other types of HMA. Results shown in Table 42, expressed as mean profile depth (MPD), were obtained using 76 17 10 30 33 10 0 5 10 15 20 25 30 35 <6 6-8 8-10 10-12 >12 Estimated Average Service Life Pe rc en ta ge o f S ta te s Figure 42. Reported estimated average service lives for OGFC layers (8). Friction Number (ASTM E274) Test Section Designation 10/27/92 11/11/92 4/12/93 2/6/96 Std. OGFC (d) 42 53 52 50 Coarse OGFC (D) 41 50 52 51 D + Mineral Fibers (DM) 39 50 53 49 D + Cellulose Fibers (DC) 37 47 53 49 DC + SB Polymer (DCP) 35 46 52 50 D + SB Polymer (DP) 32 47 51 51 D + 16% Crum Rubber (D16R) 37 48 53 51 Table 41. Average friction test results for six PFC test sections (10).
a circular texture meter and show that porous friction courses have significantly more surface texture than dense-graded HMA layers and markedly more surface texture than SMA lay- ers. McDaniel and Thornton (13) also used the results of fric- tion testing with the dynamic friction tester to determine the IFI for the three mix types shown in Table 42. The IFI utilizes the results of friction measurements along with MPD data to provide a harmonized frictional characteristic measure inde- pendent of the equipment used. Results, shown in Table 43, indicate that the PFC had the highest IFI followed by the SMA and dense-graded HMA, respectively. These results show the significant influence of surface texture on the IFI. Because of the significant amount of macrotexture pro- duced within PFC pavement surfaces, PFC layers will maintain adequate frictional characteristics even after becoming clogged (15). The macrotexture will allow water films to be dissipated under tires during rain events. Performance Life Similar to smoothness and friction, no specific references were identified that followed the permeability and noise- reducing characteristics of PFC layers over time. Generally, the performance life will be shorter than the service life. This will especially be true in areas that do not employ a general main- tenance program for cleaning flogged PFC layers. Isenring et al. (15) listed a number of causes for reduction in permeabil- ity within PFC layers. First, dust and debris can fill the void structure causing the layer to become clogged. Secondly, slight densification of the layer under traffic will reduce permeability from initial values. Other factors that can lead to reduced per- meability include environment (amount of rain) and type of traffic volume. Isenring et al. (15) state that permeability will generally be maintained within wheelpaths. Wheelpaths will maintain permeability longer because of the cleaning pressure suction action caused by tires traveling over the layer. Van Heystraecten and Moraux (22) also reported that clogging potential is reduced with intense traffic. Isenring et al. (15) state that some PFC layers will maintain permeability for more than 5 years without maintenance while some will become almost impermeable within one year. As stated in Chapter 8, in order to maintain permeability through proper maintenance, main- tenance should take place while the layer is still permeable (15). Isenring et al. (15) listed a number of favorable conditions for maintaining permeability including: areas with reduced amounts of dirt and debris; good drainage (daylighted edge and sufficient cross slope in underlying layer); high air void contents within the PFC; and the cleaning action of rapid and intense traffic. Additionally, they stated that larger maximum aggregate size gradations maintained permeability longer than smaller maximum aggregate size gradations. Ruiz et al. (16) reported less clogging in PFC mixes having more than 20 per- cent air voids. British Columbia indicated that no clogging or reduction of permeability had been observed (53). Isenring et al. (15) conducted a number of noise measure- ments to compare PFC and dense-grade surfaces. They eval- uated sound absorption, tire/pavement noise (using a trailer) and wayside measurements. For sound absorption, their research showed that PFC layers that are in good functional condition (permeability has been maintained) are capable of absorbing sound. Layers of PFC thicker than 2 in. (50 mm) had the potential for absorbing more sound. Isenring et al. (15) also showed a relationship between permeability and sound absorption. As permeability increased, sound absorp- tion also increased. However, the surface texture (macro- texture) seemed to be more important than permeability. Several pavements exhibiting low permeability values (clogged) still had the ability to absorb sound. When measuring the tire/pavement noise, Isenring et al. (15) found that PFCs in good functional condition had lower noise levels than typical dense-graded layers. At speeds above 30 to 35 mph (50 to 60 km/hr) the difference between the two wearing layers became larger. Testing with the noise trailer also resulted in a relationship between permeability and noise levels. As permeability increased, noise levels generally decreased. Also, PFCs having a smaller maximum aggregate size generally resulted in lower noise levels than coarser gradations. McDaniel and Thornton (13) also used a noise trailer to show a 4 to 5 dB(A) reduction in noise levels when comparing PFC to SMA and dense-graded layers. 77 Table 42. Results of surface texture measurements using circular texture meter (13). Mix Type Mean Profile Depth, mm (Standard Deviation) )31.0(73.1esruoCnoitcirFsuoroP )41.0(71.1tlahpsAxirtaMenotS )50.0(03.0AMHdedarG-esneD Average Dynamic Friction Tester (DFT) Number (Standard Deviation) Mix 20 kph 40 kph 60 kph International Friction Index (F60) PFC 0.51 (0.03) 0.45 (0.03) 0.42 (0.03) 0.36 SMA 0.37 (0.01) 0.31 (0.01) 0.29 (0.01) 0.28 HMA 0.52 (0.01) 0.47 (0.01) 0.44 (0.01) 0.19 Table 43. IFI data (13).
Isenring et al. (15) also reported on wayside measure- ments. The evaluations were conducted where a comparison in noise levels between PFC and dense-graded layers could be made. For single vehicle cars, a level in noise reduction of between 1 and 5 dB(A) was observed when testing PFCs. Noise levels for a traffic stream showed reductions between 0 and 3.5 dB(A). Brousseaud et al. (28) used wayside measure- ments to show a 3 to 5 dB(A) reduction in noise levels when using the Statistical Pass-By Method. The Danish government has an initiative to reduce the number of dwellings exposed to a noise level of 65 dB(A) by two-thirds (31). The two-layer PFC system was identified as potentially the most effective means of achieving this goal. According to Dutch experience (31), two-layer PFC systems have good noise-reducing characteristics compared to dense- graded layers. The reason for this is the structure of the sys- tem which contains a large number of interconnected voids. Tires rolling on the surface result in air pumping as the tire pushes air into the layer and then the air is sucked out as the tire passes. This pumping action generates a high-frequency noise. On PFCs, the pumping is reduced because the air is pumped into the interconnected voids of the layer. Similar to the work of Isenring et al. (15), the Dutch state that PFC layers also reduce noise levels by absorbing some of the noise emitted by vehicles (31). On dense-graded layers, noise emitted towards the pavement is reflected to the sur- roundings; however, on PFC some of this noise is absorbed by the pavement through the interconnected void structure. Huber (7) cited a number of references that indicated PFCs maintain their sound attenuation for five years or more as long as their design air voids are above 18 percent. A number of references reported on how the noise-reducing ability of PFCs compared to dense-graded surfaces; however, how the noise levels were measured was not reported. British Columbia reported a reduction in noise levels of 5 dB(A) (53). Iwata et al. (20) reported a reduction of 3 dB(A). Kandhal (4) cited numerous references that also stated PFCs resulted in a 3 dB(A) reduction in noise levels when using PFCs. Graf and Simond (29) reported an average reduction of 6 dB(A) after installation of a PFC layer in Switzerland. Some limited work by Graf and Simond showed that PFCs maintained the ability to reduce noise levels for up to 9 years. Performance Measures The preceding sections on service life and performance life highlighted the various performance measures encountered in the literature. Three performance measures were men- tioned more than others: noise levels, friction, and perme- ability. These three measures represent the reasons that PFC are a desirable HMA for pavement surfaces. The literature was explicit that PFCs improve wet weather friction and reduce noise levels. Improved wet weather friction can be directly related to the ability of PFCs to remove water from the pavement surface (permeability). Also, Isenring et al. (15) reported relationships between noise levels and permeability. Within Chapter 4, a number of benefits were discussed when utilizing PFCs. The vast majority of these benefits are directly related to permeability. Therefore, permeability is likely the most important performance measure for PFCs. Several pieces of equipment for determining the perme- ability characteristics of PFC layers were described. Without exception, the various equipment utilized a falling-head con- cept. Isenring et al. (15) provided the most complete descrip- tion of a permeability device utilized for PFCs called an IVT Permeameter. It was made of a plexiglass cylinder (or single standpipe) having air interior diameter of 7.5 in. (190 mm) and a height of 10 in. (250 mm). The plexiglass cylinder con- tained five engraved markings that were 0.8 in. (20 mm) apart with the âzeroâ marking being a height of 10 in. above the pavement surface. Special putty was used to seal the flow of water under the cylinder through the surface texture. An index of permeability is expressed as the amount of time required for water to travel between the âzeroâ mark and 3.2 in. (80 mm) mark. Isenring et al. (15) stated that this test is a single point estimator of permeability and that many loca- tions should be conducted in order to adequately character- ize the properties of the pavement. 78
