Thin and Ultra-Thin Whitetopping (2004)

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.

5Whitetopping overlays have proven to be a successful pave- ment rehabilitation method. The intent of this report is to identify specific cases of their application, along with sug- gested practices to help ensure proper planning, design, con- struction, and maintenance. This synthesis will provide a summary of the state of the practice and a glimpse of the state of the art. DEFINITIONS For the purposes of this synthesis, the first term that requires definition is whitetopping: Whitetopping is a PCC layer constructed atop an existing HMA pavement structure (2). Whitetopping differs from other concrete overlay types, which include • Bonded concrete overlay—a PCC layer constructed directly atop an existing PCC pavement. Specific con- struction techniques are employed to ensure a sound bond between the layers (4). • Unbonded concrete overlay—a PCC layer constructed directly atop an existing PCC pavement but intention- ally separated by a bond breaker, commonly consist- ing of HMA (4). Figure 1 illustrates these overlay types. Within the category of whitetopping, there are additional classifications of conventional whitetopping, TWT, and UTW. These terms were defined in chapter one. One possible source of confusion in the definition of white- topping pertains to the degree of bonding between the new PCC and existing HMA. Although TWT overlays have been reported in the literature dating back to 1918, little is noted about the specific construction techniques used until more recently. For example, no measures were reported as having been taken to enhance the bond between the PCC and HMA layers. To the contrary, previous recommendations for white- topping have included application of a whitewash (e.g., lime slurry or curing compound) to assist in breaking the bond (1). Because conventional concrete pavement design techniques assume the underlying HMA to be a “stiff subbase,” a lack of bond did not have a negative impact on the performance of the overlay (11). In the mid-1980s, a new concept was advanced by research- ers in Europe, where a bond between the PCC and the HMA would be considered in the design and intentionally intro- duced during construction by means of texturing of the HMA before the overlay (12). When a sound bond is assumed to exist, the concrete overlay can be of a thinner design, and thus UTW is possible. In summary, it will be assumed in this report that UTW overlays should always be designed and constructed to maxi- mize the bond between the PCC and HMA. The same assump- tion will usually be made for TWT; however, specific guid- ance will also be given for the design and construction of thicker TWT sections that do not necessarily assume bond- ing. Finally, although conventional whitetopping overlays are not considered in this synthesis, it should be noted that they are commonly designed and constructed as assuming lit- tle or no bonding action. HOT-MIX ASPHALT PAVEMENT REHABILITATION The behavior and performance of HMA pavements varies widely. Depending on the original design and ultimate use of the pavement structure, the HMA can ultimately fail in a num- ber of ways, or the sources of failure may be combined. The most common failure modes of HMA pavements that can lead to the need for rehabilitation include the following: • Rutting—pertains to permanent deformation in the wheel- paths caused by a combination of densification and shear- ing of the various pavement layers. The viscous proper- ties of the asphalt binder (especially at high temperatures) are a significant factor leading to rutting in the HMA layer. Rutting in the underlying layers can be aggravated by the presence of moisture and high strains (12). • Fatigue cracking—as with many engineering materials, repeated loading can damage an HMA, leading to the development of visible cracking. This failure mode is complex and can develop in a number of ways (13). It is commonly aggravated by a weak support system (e.g., owing to a saturated condition) and/or a stiffening of the HMA. • Thermal cracking—at low temperatures, HMA responds in a more elastic and brittle fashion. If the pavement tem- perature lowers rapidly or frequently, transverse crack- ing can form on the pavement surface (13). CHAPTER TWO OVERVIEW AND APPLICATION

6Other common structural and material failures include block cracking (the result of oxidation and shrinkage of the HMA), stripping (separation of the asphalt binder from the aggregate owing to the presence of moisture), and bleeding (excess asphalt binder forced from the matrix to the outside of the HMA layer). In addition to undergoing structural and material failures, an HMA pavement may experience a loss of function. These failure modes include degradation of ride quality (pavement smoothness), loss of skid resistance, and an increase in noise. By far the most common rehabilitation method for existing HMA pavements is an HMA overlay. Depending on a number of factors, the HMA overlay may be preceded by a milling operation (14). HMA overlays commonly “correct” functional failures, and depending on the particulars of the project, they can be used to address some structural failures (15). Whitetopping overlays provide the industry with an alter- native to HMA overlays. Both UTW and TWT are intended to correct structural and functional distress in an existing HMA pavement at a cost that is comparable to that for an HMA over- lay, especially if life-cycle costs (LCC) are a consideration (1). FUNDAMENTAL BEHAVIOR UTW and TWT overlays provide a unique pavement structure that is fundamentally different from other pavement types. UTW and, in most cases, TWT overlays are designed and con- structed with consideration of a sound bond between the PCC and HMA materials (16). The result is a composite structure that distributes traffic and environmental loading differently than more conventional PCC or HMA pavement structures. As Figure 2 illustrates, the stress distribution in a bonded systems versus that of an unbonded system can be significantly different. As a result of the composite action, the stresses in the top (PCC) layer are significantly lower in the bonded than those in the unbonded case. Furthermore, because much of the slab is in compression, and because concrete is much stronger in compression than in tension, the design for the slab can be thinner for a bonded case than for an unbonded case (2,17–19). Although a fully bonded system would be ideal, it has been shown that a partial bond is usually realized as a result of a number of factors (20,21). In such a case, the stresses will lie somewhere in between the two extreme cases, as illus- trated in Figure 2. HISTORICAL PERSPECTIVE According to NCHRP Synthesis of Highway Practice 99, the first recorded use of whitetopping in the United States was in Terre Haute, Indiana (3). On this project, constructed in 1918, a 3- to 4-in. jointed reinforced concrete overlay was put in place. Between that time and 1992, approximately 200 white- topping projects had been documented; of which 158 have been jointed plain concrete pavement, 14 continuously rein- forced concrete pavement, 10 FRC, and 7 jointed reinforced concrete pavement (4). Since 1992, the ACPA has been track- ing the use of UTW in the United States and it has documented more than 300 projects during this 10-year period (7). Fig- ure 3 displays the historical use of UTW projects in the United States according to the ACPA statistics. The average project, as noted in this figure, is approximately 2500 m2 (3,000 yd2), and ranges from 170 to 58 500 m2 (200 to 70,000 yd2). It should also be noted that although not as closely tracked as for UTW, the trend for TWT use is believed to be similar, based on earlier historical usage, along with individual cita- tions found in the literature (4,22). The use of UTW and TWT projects has not been limited to this country. The first examples of modern UTW con- struction were cited in Europe, specifically in Belgium and Whitetopping Bonded Concrete Overlay Unbonded Concrete Overlay FIGURE 1 Concrete overlay types. Unbonded BondedLo ad Lo ad Compres sion Compression Compres sionTen sion Tension Tension Reduced Stress FIGURE 2 Effect of composite action on UTW and TWT behavior under flexural loading.

7Sweden (12,23–25). Other countries reporting recent proj- ects include Canada (26), Mexico (27), Brazil (28,29), the Republic of (South) Korea (30), Japan (31–33), France (34–36), Austria (37), and The Netherlands (38). During the mid-1980s, a concept termed “fast-track” con- crete paving became more widely known (39). This method involves the use of high early strength concrete, thicker slab design, and a number of other construction techniques, allow- ing concrete pavements to be opened to traffic in as little as 12 h after placement (39,40). This method is of particular benefit on UTW and TWT projects, because it demonstrates that these overlay types can be cost competitive and still con- structed with minimal disruption to the traveling public. EXAMPLES OF USE The application of UTW and TWT overlays has been wide- spread. Of those respondents to the synthesis survey, 40% have worked with these overlays within the last year, with two-thirds of the projects in the UTW category. As men- tioned previously, this type of rehabilitation has been used for more than 80 years, although it has only more recently been advanced as a viable solution in modern HMA pave- ment rehabilitation. From the literature, it is evident that both UTW and TWT overlays have been successfully used to rehabilitate existing HMA pavements nationwide. In Tennessee, it is reported that “the use of ultra-thin whitetopping . . . has moved beyond its experimental status to become a viable alternative to HMA in certain applications” (41). The Florida Department of Trans- portation (DOT) reports, “With the success of the UTW research in Gainesville, the FDOT implemented the first UTW project in the rehabilitation of pavements at the Ellaville Truck Weight-Station on I-10” (42). The following is a summary of a number of additional projects cited in the literature. More detail on some of these case studies can be found in Appendix B of this synthesis or in the references cited. Kentucky The first high-profile UTW project in the United States took place in Louisville, Kentucky, in 1991 (6,43). It was a land- mark in UTW history, demonstrating that “ultra-thin white- topping 50 to 90 mm (2 to 3.5 in.) thick can carry traffic loads typical of many low-volume roads, residential streets, and parking lots” (10). This project was constructed of FRC pan- els, which were instrumented to provide data on the effect of heavy wheel loads. The pavement section served as an access road to a dis- posal facility, which was selected to provide performance data on the common distresses and failure modes of UTW overlays that are subjected to heavy loads similar to those of garbage trucks. Because the large number of trucks per week (400 to 600 trucks per day, five and one-half days per week) is 20 to 100 times greater than on lower-volume roads, this project was considered an accelerated test. Some of the significant findings from the Louisville UTW exper- iment included • The bond between the UTW and the existing HMA pave- ment significantly reduces the stresses in the concrete section. This allows the section to perform as a compos- ite section. • Corner cracking was the predominant distress. • Joint spacing has a significant effect on the rate of cor- ner cracking. Virginia The Accelerated Loading Facility (ALF) at the FHWA Turner–Fairbank Highway Research Center was recently used to test UTW over existing HMA pavements (44,45). Eight lanes of UTW were constructed with different thicknesses [64 to 89 mm (2.5 to 3.5 in.)], joint spacings, and concrete mix designs (with and without synthetic fibers). Design, construction, and performance data were collected for the UTW overlays placed at the facility. Specific data included concrete temperatures, slump, unit weight, and air content. Concrete properties, including compressive strength, elastic modulus, and flexural strength were also tested from numerous cylinders and beams cast during placement. Bond strengths between the PCC and HMA were also tested using a procedure developed by the Iowa DOT (46), and stiffnesses of the various pavement layers were estimated by falling weight deflectometer (FWD) testing (2). Data related to distress development and modes of fail- ure have been analyzed as part of a recently completed proj- ect (2,47–49). From this analysis, it was shown that a major contributor to the failure in the UTW was the resilient versus permanent deformation characteristics of the support layers below the UTW overlays. Specifically, it was found that the 0 50 100 150 200 250 300 350 N o. o f P ro jec ts 19 89 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 Year FIGURE 3 UTW construction in the United States.

8more viscous the HMA layer, the more quickly and severely the UTW would deteriorate. The ALF applies approximately 50,000 loads per week, testing 24 h a day. After several months of testing, 0.7 to 3.5 million 80-kN (18-kip) equivalent single-axle loads (ESALs) were applied. The result was that six of the eight lanes had no significant loss in ride quality (49). The two lanes that did exhibit significant distress were constructed on the softest HMA sections. Iowa An 11-km (7-mi.) whitetopping overlay project on Iowa’s Route 21 between Victor and Belle Plaine consisted of four overlay types representing both UTW and TWT (162). Thick- nesses on the project ranged from 50 to 200 mm (2 to 8 in.), and some of the concrete was fiber reinforced. This project yielded performance data on whitetopping with a real highway application. As with other whitetopping research projects, design data, construction techniques, instru- mentation, and performance data are all available. Given the large number of combinations of thickness, joint spacing, con- crete mix, and surface preparation, this experiment continues to be one of the most referenced pertaining to the demonstra- tion of the applicability of this rehabilitation alternative. Minnesota In 1997, the Minnesota DOT (Mn/DOT) constructed two UTW overlay projects. The first was on US-169 in Elk River, where a 75-mm (3-in.) UTW was placed (50). A second proj- ect included both UTW and TWT ranging from 75 to 150 mm (3 to 6 in.), constructed on I-94 at the Mn/ROAD test facil- ity (51,52). These sections were heavily instrumented, which allowed measurement of the static and dynamic response of the pavement under various applied loadings and environ- mental conditions. The goal of these projects was to deter- mine design features that can optimize the life of the overlay. The results of these projects have better enabled interested parties to understand the behavior and performance of UTW overlays under very controlled conditions. Missouri Both UTW [89 mm (3.5 in.)] and conventional [250 mm (10 in.)] whitetopping overlays were placed at the Spirit of St. Louis Airport in 1995 (8,53). Although the project was not a highway application, valuable information was collected, which has resulted in improved design and construction spec- ifications. The ACPA continues to monitor the whitetopping overlays at the airport to determine bond strength and stresses in the concrete and in the HMA layer below the overlays. Colorado Santa Fe Drive, Denver, Colorado In May 1996, two 152-m (500-ft) test sections were con- structed at this location (17,20). These sections included a 100-mm (4-in.) TWT on an unprepared 125-mm (5-in.) HMA and a 125-mm (5-in.) TWT on a milled 100-mm (4-in.) HMA. On some sections, a new HMA mat was constructed 2 to 3 days before the overlay. The slabs on both sections were 1.5 m2 (5 ft2). A total of three test slabs were instrumented for strain, deflection, and temperature. Load testing on these slabs was conducted in August 1996 with a 90-kN (20-kip) axle load at several times during the day. Cylinders, beams, and cores were taken for thickness verification, PCC–HMA bond shear strength, strength (compressive and flexural), PCC modulus of elas- ticity, HMA-resilient modulus, and unit weight. On the basis of observations made on this project, it was recommended that the whitetopping overlay not be placed on top of a newly laid HMA layer. Subsequent work has sup- ported this conclusion, but emphasizes that the properties of the HMA—existing or new—should be considered in the overlay design (2). State Road 119, Longmont, Colorado On this project, a number of design and construction vari- ables were used, including varying TWT thickness, joint spac- ing, use of dowels, and HMA surface preparation (17,20). A total of five slabs were instrumented on this project for strain, deflection, and temperature. Each slab had a unique combi- nation of the aforementioned design features. Construction was done in August 1996, with load testing being conducted the following month. An 80-kN (18-kip) axle load was used for the load testing at this location. Cylinders and cores were also collected and measured for strength (compressive), PCC modulus of elasticity, thickness verification, PCC–HMA bond shear strength, HMA-resilient modulus, and unit weight. One of the more important findings from this project was the rec- ommendation of tie bars along the longitudinal construction joint. It was found that without the tie bars, the adjacent lanes would separate and move longitudinally with respect to each other. Montana Great Falls, Montana To repair an intersection subjected to heavy rutting and shov- ing, the Montana DOT constructed a TWT overlay at the intersection of N.W. Bypass and 3rd Street N.W. in Great Falls (54,55). Constructed in the fall of 1999, the project has

9been considered a success in meeting the needs of the Mon- tana DOT. Approximately 2 years after construction, one section was found to have developed a localized failure. Slab shattering was observed on an area approximately 3 m (10 ft) by 4.5 m (15 ft). In March 2001, the damaged area was repaired by removing the existing panels along with the underlying HMA and replacing it with full-depth concrete. On removal and inspection of the pavement materials, it was apparent that an underlying storm drain had been a source of moisture that led to a weakened subgrade and stripping within the HMA layer. Before placement of the full-depth patch, the subgrade was partially removed and replaced with a select unbound aggre- gate. Since the placement of the original patch, some addi- tional cracking (shattering) has been observed on additional panels in the vicinity. Kalispell, Montana In September 2000, the Montana DOT participated in the resurfacing of a segment of East Idaho Street in the city of Kalispell (54). The existing HMA pavement was reported to be heavily distressed, with rutting, shoving, and transverse cracking. A TWT was identified as the viable alternative, constructed by removing 130 mm (5 in.) of the HMA surface and replacing it with an equal thickness of PCC. The project was built using concrete paving practices typ- ical of other projects in the state. Milling of the existing HMA was done, followed by vacuuming and sweeping to remove the remaining debris that may contribute to debonding. Con- struction was staged to accommodate both traffic patterns and the selection of paving equipment. During the course of the paving, wash water (slurry) from the sawcutting opera- tions was found to cover the milled surface. As a result, addi- tional vacuuming and sweeping were performed before PCC placement to ensure a good bond. During placement, some locations were found to have insufficient HMA thickness to contribute to composite action. As a result, the HMA was completely removed at these locations, and the PCC was placed on top of the existing base. Thicker sections were also used on areas where additional load transfer was desired. The mix included 1.8 kg/m3 (3 lb/yd3) of polypropylene fibers, which were introduced into the trucks at the batch plant. A Bid-Well deck paver was used to place most of the concrete. Vibration during placement was found to be criti- cal. A single handheld “stinger” type vibrator was used, but the recommendation was subsequently made that two units would have been more helpful. After placement was complete, one small section of con- crete was found to be of poor quality as a result of improper vibration during construction. The panels were subsequently removed and replaced, dowelling the replacement panels into the existing overlay. Overall, however, the Montana DOT research reported that this TWT project is functioning as intended. Of 4,200 panels, only 12 have exhibited minor cracking, and no debonding is evident. People living adjacent to this project have also reported how quiet their neighborhood has become thanks to the lack of heavy vehicles “bouncing” through the intersections.