Thin Asphalt Concrete Overlays (2014)

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.

6 chapter two USE OF THIN LIFT OVERLAYS Based on comments by responding agencies, various types of thin overlays are used, and the criteria used for treatment selection is varied. Based on survey responses and supplemen- tal information, it was found that Ohio and California have an extensive evaluation process to aid in selecting the right treatment based on pavement condition. TYPES OF THIN OVERLAYS There are several types of mixes that have been used success- fully in thin overlay construction. Superpave®-type dense- graded mixes, such as 9.5- and 12.5-mm NMAS mixtures, are common fixtures in most highway applications. Even 9.5- and 12.5-mm NMAS stone matrix asphalt (SMA) mixtures are widely viewed as the premium asphalt mixes for asphalt con- struction and resurfacing projects. These mixes are designed for stone-on-stone contact to resist rutting and have a rich mortar to provide long-term durability and resist cracking. They cost more than typical dense mixtures but are very cost-effective because the mixes may last more than 20 years without resurfacing (Newcomb 2009). Another mixture that has gained acceptance within several agencies is the UTBWC, which was originally developed in France. Arkansas uses this mixture exclusively for thin over- lays. The UTBWC is a gap-graded mixture with polymer- modified asphalt binder that is typically placed with a spray paver that applies a polymer-modified emulsion tack coat immediately ahead of the paver auger chamber. Arkansas, Illinois, Kansas, Louisiana, Minnesota, and Vermont indicated that UTBWC mixtures are used routinely as an option for thin overlays. A study at the University of Illinois considered the frac- ture energy test (ASTM D7313-07b) as a means of evalu- ating cracking resistance of UTBWC mixes (Sarfraz et al. 2010). Variables such as tack application rate, air void level, and overlay thickness were accounted for in the experiment, and results from roadway cores were compared with those of plant-compacted samples during construction. The results showed UTBWC mixes had greater resistance to cracking than did typical wearing course mixtures. A dense 4.75-mm NMAS mixture is gaining popularity with many agencies. The fine-graded 4.75-mm mix originally was eliminated from the gradation bands of the Superpave mixture specifications. It was added to the AASHTO speci- fications in 2002 because of the need for small aggregate size mixtures (West et al. 2006). Yet it is being considered for use so widely in thin asphalt overlays that numerous research projects have been conducted to improve the per- formance of the mixture. The mixture has continued to gain acceptance after a couple of research studies by the National Center for Asphalt Technology (NCAT). In 2002, a compari- son of coarse and fine-graded Superpave mixtures (Kandhal and Cooley 2002) that included 14 mixes with 9.5- and 19-mm NMAS showed no significant difference in rutting resistance between the coarse and fine-graded mixes. If the coarser mixes are not more rut resistant, the use of finer mixes becomes more desirable owing to their being more workable, less permeable, less likely to segregate, and their potential to be more economical because they can be placed in thinner layers. The same year, research results were made available that helped establish initial criteria for a Superpave 4.75-mm NMAS mixture (Cooley et al. 2002). The research established parameters for design air void level, voids in mineral aggregate (VMA), and voids filled with asphalt for 50 and 75 gyrations. When SMA technology was introduced to the United States in 1990, many agencies quickly placed experimental projects with 12.5- and 19-mm NMAS mixtures. However, Euro- pean SMA specifications also included finer graded mixes. In 2003, NCAT conducted research using 4.75- and 9.5-mm NMAS SMA mixtures (Cooley and Brown 2003) and mea- sured rutting potential with the asphalt pavement analyzer (APA) rut testing machine. Rut depths were measured after 8,000 load cycles at two test temperatures. The finer mixes were found to be rut resistant (Table 2), could be placed in thinner layers, and were less permeable than coarser mixes, thus making them good candidates for use in thin overlays. Thin overlays are not a new idea to Europe; they have been used in Spain for more than 40 years (Luelmo et al. 1971) as routine road maintenance. Luelmo and colleagues cautioned against placing thin layers during cold weather. Sound advice is also given in a Canadian report that recommended using thin overlays only when the existing pavement and base lay- ers are structurally sound (Cewe 1966). Pavements that are failing or have failed cannot be successfully treated with a thin overlay alone; they must be repaired so that a stable foun- dation is provided before the thin overlay is placed.

7 is recommended), low severity rutting ≤ 0.50 inches, increase in skid resistance needed, existing pavement in fair to good condition. Not for alligator cracking, not for severe ravel- ing where pavement deterioration exists, and not for rutting > 0.50 inches without correcting rutting first.” In Pennsyl- vania, a fine-graded 9.5-in. NMAS mix is used for over- lay thicknesses of 1.0 to 1.5 in. (25–38 mm) and a 6.3-mm NMAS mix is being used in a pilot program for thicknesses between 0.75 and 1.25 in. (19–32 mm). Most agencies determine when to apply a thin overlay by conducting condition surveys of the existing pavement. As is done in Pennsylvania, the condition surveys generally are performed annually, at least for high traffic or high project classification, or possibly biannually for lower classification roadways. For example, in Illinois thin overlays are allowed only when the existing condition rating is within a certain range. This appears to be a practical approach because if the rating is too low, the structural damage done may not be remedied with a thin overlay. On the other hand, if the condi- tion rating is too high, it may not be cost-effective to place a treatment. Similarly, the Kentucky respondents indicated their agency uses a pavement management system to generate a list of potential candidates and rehabilitation options. When asked what investigation was done to determine when to use thin overlays, a number of agencies responded that no investigation was made (Figure 2). One reason for this response may be that some agencies have decided as a matter of policy what treatments would be used under certain conditions. However, other agencies said they use more than one approach. For example, an agency may take cores from the existing roadway to determine thickness for a structural analysis and may also schedule milling to remove cracking based on observation of the cores. One agency responded that milling was often planned to restore geometric profile and increase surface texture to improve adhesion of the overlay. Open-graded friction course (OGFC) mixtures are also used in thin layer construction. The open structure of the mix causes the OGFC layer to remain cooler than dense- graded mixes. A comparison of the effects of mix type on pavement temperature (Watson et al. 2004) showed that the layer immediately underneath OGFC mix was about 4°F (2°C) cooler than where dense-graded mixes were used. This insulation effect was found to extend the life of jointed con- crete pavements in Arizona by reducing the curling stress in concrete slabs (Belshe et al. 2007). The OGFC layer effec- tively reduces the temperature differential between the top and bottom of the slab. OGFC mixtures have also proven to be useful for reduc- ing roadway noise at the tire-pavement interface. In 1998, Caltrans placed an OGFC layer for the purpose of monitoring noise abatement and started developing a database of various pavement types and noise levels (Rymer and Donavan 2005). Open and dense-graded asphalt mixtures and portland cement concrete (PCC) pavements in both California and Arizona have since been added to the database. The analysis showed that the quietest one-third of the pavements were either OGFC pavements or pavements with crumb rubber included in the mixture. The middle one-third was dense graded with some overlap of OGFC mixtures and some tined PCC pavements. The loudest pavements typically were PCC and large, angular aggregate asphalt mixtures. TREATMENT SELECTION CRITERIA Survey respondents from Pennsylvania provided one of the most descriptive criteria used for selecting thin overlays. The agency recommends that thin overlays be used or lim- ited based on the following conditions: “Low to moderate raveling, low to medium longitudinal cracking not in wheel path, temporary short term fix for longitudinal cracking in wheel path (fatigue), low severity transverse cracks (milling SMA Mix Type Average Rut Depth (mm) at 50°C Average Rut Depth (mm) at 64°C 4.75/2.36 4.2 5.3 4.75/1.18 2.7 5.4 9.5/4.75 2.8 4.4 9.5/2.36 3.5 5.4 12.5/9.5 3.7 4.5 12.5/4.75 4.1 5.4 19/4.75a 1.7 2.6 19/4.75b 1.4 2.2 Source: Cooley and Brown (2003). aDesign gradation—Phase 1. bDesign gradation—Phase 2. TABLE 2 RESULTS OF RUT TESTING ON DESIGNED SMA MIXTURES

8 also included, so the project can compare the cost-benefit of each treatment to the option of doing nothing (Hunley 2013). Performance curves may also be developed based on the time it takes for each section to deteriorate to the same condition level that existed before the treatment. One of the sections is a UTBWC mixture placed with a spray paver at 0.75 in. (19 mm) thick. The same thickness is used for comparison on seven test sections with various 4.75-mm NMAS mixes. The 4.75-mm mix sections (Fig- ure 3) include variations in surface preparation, asphalt binder grade (including a highly modified binder), and use of reclaimed asphalt pavement (RAP) and recycled asphalt shingles (RAS). New Jersey has compared several mixes used in thin lift surface courses and found that OGFC mixes provide the most benefit for the least cost (Bennert et al. 2005). OGFC mixes performed extremely well at noise reduction, wet friction numbers, ride quality, and cost-effectiveness. The 12.5-mm Superpave mix, which is the standard mix used for overlays, is typically placed a minimum of 2 in. (50 mm) thick, result- ing in a construction cost twice as expensive as the OGFC. Survey respondents were also asked when they would not recommend using thin overlays. Not surprisingly, the largest area of response was that such overlays should not be used when there is existing cracking, especially at a medium to high severity level (Figure 4). As a general rule, underlying In supplemental information provided during the survey, Ohio respondents described using a decision tree to deter- mine if a project is a candidate for a thin overlay. The deci- sion matrix uses a combination of traffic, pavement condition ratings, and structural deducts to determine if an overlay or other treatment is needed. There is a decision tree for use with primary routes (more than four lanes), and another for use with the general system (two lanes) of roadways. Thin overlays are considered to be cost-effective if the existing pavement condition rating (PCR) score is between 70 and 90 for Ohio’s Priority System, and between 65 and 80 for its General System pavements (Chou et al. 2008). An example based on a portion of Ohio’s General System decision tree for flexible pavements is given in Appendix D. A web-based training course (No. 131110) available through the National Highway Institute discusses pavement preservation and in Module 11 describes how to choose various treatments depending on the type of distress being addressed (FHWA-National Highway Institute). The guidelines for the course were based on a Technical Advisory Guide published by the Caltrans Office of Pavement Preservation. A research project being conducted by NCAT in coop- eration with several sponsors will evaluate 25 test sections of a variety of preservation treatment options. The project, which was constructed in 2012, is unique in that all of the test sections are on the same roadway with the same traffic and structural conditions. Control sections with no treatment are FIGURE 2 Investigation done to determine when to use thin overlays. (Source: Survey responses.) Secon 18 19 20 21 22 23 24 25 Surface 4.75/PG 67-22 4.75/PG 67-22 4.75/PG 76-22 4.75/PG 76-22 UTBWC 4.75 50% RAP 4.75 5% Shingles 4.75 PG 88-22 Subsurface Fibermat Exisng Full-Depth Reclamaon Exisng Exisng Exisng Exisng Exisng FIGURE 3 The 4.75 mm test sections on Alabama Road 159 in Lee County. (Source: Cooley et al. 2002.)

9 cracks propagate upward at a rate of about 1 in. (25 mm) per year. Thus, if asphalt layers are being placed in thin lifts, the cracking can be expected to reflect through to the surface in a short time. Both rutting and cracking distresses may not be resolved easily with a thin overlay. Rutting, for example, may extend to about 4 in. (100 mm) into the structure. This occurs as the asphalt binder in the pavement begins to soften near the sur- face from exposure to high ambient temperatures and radia- tion from the sun. In some cases rutting may be a reflection of unstable base or subgrade underneath the structure. In those cases, a more serious (and costly) approach, including reha- bilitation, may be needed. Some current research, such as the high polymer modified asphalt section (Section 25) on Lee County Road 159 may provide useful information about how such a mix can help withstand reflective cracking even in thin layers. Georgia has placed two test sections on the current (2012) research cycle of the NCAT Test Track that will evalu- ate the effectiveness of different methods for trying to retard reflective cracking. Saw cuts were made to simulate structural block cracking (Figure 5) and covered with surface treatment (chip seal) variations before a thin overlay was placed. FIGURE 4 Conditions where thin overlays would not be recommended. (Source: Survey responses.) FIGURE 5 Saw-cut sections at NCAT Test Track to simulate structural cracking. (Source: Buzz Powell, NCAT.)