Using Existing Pavement in Place and Achieving Long Life (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.

41 A p p e n d i x A The R23 team conducted a thorough literature search for information on highway renewal using existing pavements. The resources utilized for this task include the following: • The Transportation Research Information Service (TRIS) database, • The International Transportation Research Documenta- tion (ITRD) database, • The Transportation Libraries Catalog (TLCat), • The National Technical Information Service (NTIS) database, • The Transportation Research in Progress (RiP) database, • The Online Library Catalog of the University of Illinois at Urbana-Champaign, • ProQuest’s ABI/INFORM Complete database of periodi- cals, professional journals, and trade publications, • The Federal Highway Administration’s (FHWA’s) National Highway Specifications website, • The Bureau of Transportation Statistics’ National Transpor- tation Library (which searches some of the above databases and others, as well as the websites of the departments of transportation of all 50 states and the District of Columbia), • The Virtual Library of publications of the World Road Asso- ciation (PIARC), • The United Kingdom’s Transport Research Laboratory (TRL) publication database, • The Netherlands’ Foundation Center for Research and Contract Standardization in Civil and Traffic Engineering (CROW) publication database, • The publication databases of the American Concrete Pavement Association (ACPA), Asphalt Institute (AI), and National Asphalt Paving Association (NAPA), and • The publication databases of the roadway authorities of Aus- tria, Germany, France, Belgium, the Netherlands, the United Kingdom, Canada, Australia, South Africa, and other coun- tries facing demands of heavy traffic on high-volume roads and the need for in-place renewal of existing pavement structures. In addition, questionnaires were sent to each of the state highway agencies. These surveys were followed by a series of phone calls to learn more about state-sponsored reports. The state survey portion of the project is described in more detail in Chapter 2. A number of definitions for long-life pavements exist, depending on the location, pavement type, and road- way facility. For purposes of this study, long-life pavement is defined as pavement sections designed and built to last 50 years or longer without requiring major structural rehabilitation or reconstructions and needing only periodic surface renewal in response to distresses confined to the top of the pavement. Table A.1 shows typical ranges of service life for reconstruction and various major rehabilitation techniques for each pavement type (Thompson 1989). These ranges are general estimates only and represent the “conventional wisdom” about the ser- vice lives that may reasonably be expected of the different reha- bilitation techniques. Based on these estimates, the most promising renewal strategies for long life using existing pave- ments are the following: • Thick AC overlay over existing AC pavement, • PCC overlay over existing AC pavements, • Thick AC overlay over fractured PCC pavement, • Bonded PCC overlay over existing PCC pavement, • Unbonded PCC overlay over existing PCC pavement, • Thick AC or PCC overlay over fractured PCC of AC/PCC composite pavement, and • Unbonded PCC overlay over existing AC/PCC composite pavement. The following sections provide details on each rapid renewal strategy along with considerations for long-life pave- ment based on relevant literature and agency information. Literature Review

42 Asphalt Concrete (AC) Renewal Approaches AC over AC Methods The team sought information on the following potential flex- ible pavement renewal methods: • AC over existing AC, • AC over crushed and shaped AC, and • AC over reclaimed AC. The three overlay methods listed above have been used by many states and other countries, for conventional rehabilita- tion purposes—that is, for rehabilitation design lives typically not exceeding 15 years. Nonetheless, it seems entirely feasible from a conceptual standpoint that a new AC surface of suffi- cient thickness and durable mix design could be placed on an existing AC-surfaced pavement as a long-life renewal approach. However, to determine the structural and material requirements needed to achieve a true long-life renewal of the existing pavement, it may be necessary to think of the new AC surface as new construction on a high-quality base rather than as a conventional overlay. AC over Existing AC Pavement The hot-mix asphalt (HMA)-over-existing-HMA strategy ranges from “milling and filling” for the lower levels of traffic to “milling and strengthening” for the higher levels of traffic. Figure A.1 shows the example design cross sections for long-life performance of HMA pavements developed by Von Quintus for the Michigan Asphalt Pavement Association (Asphalt Pave- ment Alliance 2002). It includes suggested types of HMA mix- tures to be placed within the pavement structure. Von Quintus recommends that the asphalt mixture for the HMA base layer be designed to have 3% air voids to mitigate bottom-up fatigue cracking. The surface course mixture is a dense graded Super- pave in the case of 3 and 10 million equivalent single axle load (ESAL) levels, and an SMA in the case of 20 and 30 million ESAL levels (20-year life). The strategies presented in the figure are for planning purposes only. AC over Crushed and Shaped AC Pavement The technique involving HMA over crushed and shaped HMA consists of crushing the existing HMA layer and shap- ing it into a base layer before overlaying it with a new HMA layer (Figure A.2). This strategy is suitable for severely cracked HMA pavements. Marginal base material can be upgraded with admixtures to provide high-quality support. To avoid reflection cracking, crack-relieving separator layers or mem- branes can be used, including (1) geotextile or fabrics and (2) stress-relieving or stress-absorbing membrane interlayers. Crushing is usually more economical than hot mix recycling, unless the asphalt surface is quite thick. When the existing mat is quite thick (greater than 6 in.), a common procedure is to mill off part of the HMA then crush the remainder. AC over Reclaimed AC Pavement An alternative to crushing and shaping is to recycle the HMA layer using hot mix or cold mix in-place recycling techniques. The product is a renewed HMA base layer that is overlaid with new HMA. Recycling can involve cold mix for the lowest Table A.1. Typical Ranges of Service Lives for Rehabilitation Treatments Treatment Typical Range of Service Life (years) Reconstruction Reconstruction in asphalt 15–20 Reconstruction in concrete 20–30 Asphalt pavement rehabilitation Structural asphalt overlay of asphalt pavement 18–15 Structural concrete overlay of asphalt pavement 20–30 Surface recycling without overlay 4–8 Nonstructural asphalt overlay of asphalt pavement 4–8 Nonstructural (ultrathin) concrete overlay of asphalt pavement 5–15 Asphalt patching without overlay 4–8 Concrete pavement rehabilitation Structural asphalt overlay of concrete pavement 8–15 Asphalt or concrete overlay of fractured concrete slab 15–25 Unbonded concrete overlay of concrete pavement 20–30 Nonstructural asphalt overlay of concrete pavement 4–8 Bonded concrete overlay of concrete pavement 15–25 Restoration without overlay 5–15 Asphalt-overlaid concrete pavement rehabilitation Structural asphalt overlay of asphalt concrete (AC)/portland cement concrete (PCC) pavement 8–15 Asphalt or concrete overlay of fractured concrete slab 15–25 Unbonded concrete overlay of AC/PCC pavement 20–30 Surface recycling without overlay 4–8 Nonstructural asphalt overlay of AC/PCC pavement 4–8 Nonstructural (ultrathin) concrete overlay of AC/ PCC pavement 5–15 Source: Hall et al. 2001.

43 layers or hot mix for the upper base layer. The categories of pavement recycling options are shown in Figure A.3 (National Highway Institute 2003). Only the categories applicable to using existing pavement in place are discussed next. Cold In-PlaCe ReCyClIng (CIPR) CIPR involves the reuse of an asphalt concrete pavement that is processed in place with the addition of asphalt emulsions, cutbacks, portland cement, lime, and/or other materials as required to achieve desired mix quality, followed by place- ment and compaction. CIPR is accomplished by a special machine that scarifies the existing surface to a given depth, crushes it in a pug mill, adds asphalt cement, and lays the resultant mix back down, almost in its original location. The Asphalt Recycling and Reclaiming Association (ARRA) dif- ferentiates two different CIPR procedures as full depth and partial depth. Partial-depth CIPR involves the recycling of the asphalt-bound layers to a depth of 3 to 4 in. Full-depth CIPR, also termed full-depth reclamation, involves the recycling of the asphalt-bound layers and the unbound granular layers in the flexible pavement. The finished product is considered a base only, and a hot mix surface course is necessary. CIPR has been performed on all types of roadways, with the concentra- tion being on lower-volume roadways. However, full-depth reclamation has successfully been conducted on high-volume Interstate pavements. This process can directly address struc- tural problems through the production of an improved stabi- lized layer when full-depth reclamation is used. Partial-depth reclamation is limited to correcting only those distresses that are surface problems in the asphalt layer (Hall et al. 2001). Records on performance are highly variable because a com- mon definition has not been applied to judge the comparative performance levels. Causes commonly noted for poor perfor- mance using CIPR include (1) use of an excessive amount of recycling agent; (2) application of a surface seal prematurely; (3) recycling only to the depth of an asphalt layer, resulting in delamination from the underlying layer; and/or (4) allowing a project to remain open for too long into the winter season (Hall et al. 2001). Source: Asphalt Pavement Alliance 2002. Figure A.1. Michigan design catalog for long-life HMA pavements. Figure A.2. Schematic of HMA over crushed and shaped HMA pavement. Source: National Highway Institute 2003. Figure A.3. Categories of pavement recycling options.

44 Hot In-Place RecyclIng (HIPR) ARRA defines three types of HIPR operations: heater scarifi- cation, repaving, and remixing. Each is described below (Hall et al. 2001). Heater scarification involves the following steps: (1) heating the existing pavement surface to about 110°C to 150°C, using one or more propane-fired radiant heaters; (2) scarifying the softened surface to a depth of about one-half to three-quarters of an inch; (3) applying a liquid rejuvenating agent (if needed); (4) mixing and leveling the loose mixture with an auger and/ or lay-down machine; and (5) compacting with rollers. Repaving is heater scarification combined with placement of a new asphalt concrete overlay. The process involves the follow- ing steps: (1) heating the existing pavement surface to about 190°C, using infrared heaters; (2) scarifying the softened sur- face to a depth about one-half to three-quarters of an inch; (3) applying a liquid rejuvenating agent (if needed); (4) mixing the loose mixture with an auger; (5) spreading and screeding the recycled mixture; (6) placing a new asphalt concrete layer over the recycled mixture; and (7) compacting with rollers. Remixing is similar to repaving but involves mixing mineral aggregate or new asphalt concrete hot mix into the scarified, rejuvenating material rather than placing a layer of new asphalt concrete on top. Remixing not only increases the structural capacity of the pavement, as does repaving, but also permits improvement of the gradation or binder properties of the existing asphalt concrete layer. Remixing involves heating and reworking material to a greater depth than in heater scarifica- tion and repaving. The steps in the remixing process are the following: (1) heating the existing pavement surface to about 85°C to 105°C, using one or more propane-fired radiant heat- ers; (2) milling the softened surface to a depth of about 1 to 2 in.; (3) mixing the hot milled material, rejuvenating agent, and new asphalt concrete material in a pug mill; (4) placing the mixture; and (5) compacting with rollers. HIPR without an accompanying overlay or addition of new asphalt concrete material is estimated to have a service life of about 4 to 8 years. How much HIPR in conjunction with an overlay or additional asphalt concrete thickness benefits over- lay performance has not yet been quantified (Hall et al. 2001), although remixing with a thick HMA overlay provides the best potential of achieving long life. AC-over-PCC Methods When in-place renewal of an existing PCC pavement is con- sidered, the structural design considerations that must be taken into account to ensure good long-term performance are the adequacy of the subgrade, protection of the subgrade from excessive deformation, limiting strain in the existing PCC, limiting stress and strain in the new AC or PCC surface, and minimizing reflection cracking in the new surface. Although AC overlay is undoubtedly the most common major rehabilitation method for jointed PCC pavements, the service life of this technique is limited by the rate at which reflection cracks develop and deteriorate to unacceptably rough levels. Thus, an AC overlay of a jointed PCC pavement is typically considered a conventional rather than a long-life rehabilitation approach, with an expected service life of about 10 to 15 years. However, exceptions exist: Iowa, for example, has experi- ence with jointed PCC pavements built in the 1930s and 1940s, widened with PCC or AC from 18, 20, or 22 ft to 24 ft in the 1970s, and then overlaid over time with a total of 5 or more inches of AC. Now, some 30 years later, these old AC/ PCC pavements are being widened again, to 28 or 32 ft, and are being overlaid with PCC (J. K. Cable, personal communi- cation, 2008). The most promising long-life rigid pavement methods, however, appear to be the following: • AC over continuously reinforced concrete pavement (CRCP), • AC over cracked and seated jointed plain concrete pavement (JPCP), and • AC over rubblized PCC. AC over Existing CRCP AC overlays of CRCPs can reasonably be expected to perform much longer than AC overlays of jointed concrete pavements, especially when (a) working cracks and punchouts in the existing CRCP are repaired with continuously reinforced full- depth concrete and (b) the existing CRCP does not have D-cracking. Permanent patching of punchouts and working cracks will delay for many years the occurrence and deterio- ration of reflection cracks in asphalt overlays of continuously reinforced concrete pavements. Reflection crack control treatments are not necessary for asphalt overlays of continu- ously reinforced concrete pavements, as long as continuously reinforced concrete repairs are used for deteriorated areas and cracks (Barnett, Darter, and Laybourne 1981; Darter, Barnett, and Morrill 1982; Hall and Darter 1989). It has often been suggested that an adequate thickness of AC over a sound CRCP may be the perfect application for long-life design, which would require nothing more than periodic renewal of the AC surface. However, such rehabilita- tion projects are not currently typically designed for lives in excess of about 20 years. The most commonly used approach to structural design of asphalt overlays of concrete pavements and asphalt-overlaid concrete pavements is the structural deficiency approach, exemplified by the 1993 AASHTO Guide procedure. The required AC overlay thickness is determined by multiplying the structural deficiency (Df, the required concrete thickness

45 for future traffic, minus Deff, the effective thickness of the existing concrete slab) by an adjustment factor, A, that con- verts the thickness deficiency from inches of concrete to inches of asphalt. A value of 2.5 has traditionally been used for the adjust- ment factor A. This value was based on the results of acceler- ated traffic tests conducted by the Corps of Engineers in the 1950s. The value 2.5 does not represent the best fit of the rela- tionship of concrete thickness deficiency to asphalt overlay thickness in those field tests, but rather a conservative value suggested by the Corps for use in design. However, an A value of 2.5 can lead to excessive overlay thickness for larger con- crete thickness deficiencies. A formula for the A factor as a function of the magnitude of the concrete thickness defi- ciency was developed by Hall (1991) using elastic layer analy- sis and is recommended in the 1993 AASHTO Guide in place of a constant A factor. The NCHRP 1-37A Mechanistic-Empirical Pavement Design Guide (MEPDG) procedure’s software for design of AC over CRCP allows the user to select some or all of the following performance criteria by which the adequacy of a trial overlay design is judged: • Longitudinal cracking of the AC overlay, • Thermal cracking of the AC overlay, • Rutting of the AC overlay, and • Punchout damage in the existing CRCP. New pavement models for rutting in AC layers, longitudi- nal (top-down) cracking in AC, thermal cracking in AC, and punchouts in CRCP are adapted for use in the prediction of AC overlays of CRCP in the MEPDG methodology. The smooth- ness parameter used for AC overlays of PCC pavements in the MEPDG methodology is the international roughness index (IRI), predicted from an empirical model as a function of the existing pavement’s IRI at the time of overlay placement, the time elapsed since placement of the overlay, the average rut depth, and the average spacing of medium- and high-severity transverse cracks. The viability of the AC-over-CRCP method as an in-place renewal option, and the AC overlay thickness and CRCP con- dition requirements necessary to make it viable, need to be explored in this study in coordination with the work done on composite pavements in SHRP 2 Renewal Project R21. AC over Cracked and Seated Pavement Cracking and seating a plain jointed concrete pavement before overlaying it with AC has been done in the United States as far back as the 1940s. The technique attracted renewed interest beginning in the 1980s as an approach to reflection crack con- trol (J. K. Cable, personal communication, 2008; Barnett et al. 1981; Darter et al. 1982). A great number of crack and seat projects have been built on highways in the United States, including test sections in the Long-Term Pavement Perfor- mance (LTPP) specific pavement study (SPS)-6 (Rigid Pave- ment Rehabilitation) experiment. This technique is suitable for JPCPs. It involves breaking the existing concrete into pieces about 12 to 48 in. (305 to 1,220 mm), as shown in Figure A.4. In principle, the smaller the cracked piece, the larger the potential for reduction in reflection cracking, and the larger the reduction in the struc- tural strength of the concrete pavement. Cracking and seating is done in four major steps: (1) cracking the concrete slab, (2) seating the cracked slab, (3) special treatments, and (4) HMA overlay. Cracking of the pavements can be accomplished with drop hammers, guillotine hammers (Figure A.5), modified pile drivers, or whip hammers, with the most commonly used equipment being the drop hammer. These are self-propelled units that raise a heavy mass several feet above the pavement and then release the weight, which then falls and strikes the surface of the slabs. Some agencies require cracking in both transverse and longitudinal directions. The resulting pieces should be large enough to retain interlock between aggre- gates, but also small enough to minimize the joint movement of the unreinforced PCC pavement. Excessive cracking can be detrimental to the PCC pavement. After cracking, the slab is seated using 66- to 110-kip- capacity (30- to 50-ton-capacity) rubber-tired rollers (Fig- ure A.6). Seating of the concrete is done to (1) ensure reestab- lishment of the support between the subbase and the slab by reducing the existing voids, (2) create a relatively uniform grade for supporting paving operations, and (3) locate soft zones in the underlying layers that may need to be removed and/or replaced with more stable material. Excessive rolling may be harmful to the slab. The main concern with break or crack and seat is the reduction in the structural capacity of the pavement. To com- pensate for the reduction in structural capacity, thickness of the overlay should be increased. Pavement rehabilitated with Source: National Highway Institute 2003. HMA Overlay Subgrade Subbase Firm Foundation Cracks Short slab length (~2’) Good granular interlock Figure A.4. Schematic of HMA over cracked and seated pavement.

46 the crack and seat technique can perform well when the sub- grade support is uniform and the subgrade modulus is more than 15,000 psi after cracking. Nondestructive testing (NDT) should be used to analyze and design the cracked and seated pavements. Some studies have suggested that cracking and seating only succeeds in delaying the onset of reflection cracking by a few years (e.g., five or fewer), and that once reflection cracking appears, it tends to progress at much the same rate as it does in an AC overlay of an intact PCC pavement (e.g., about a year per inch of overlay thickness in reaching the surface) (Carpenter and Darter 1989). Improvements in slab cracking techniques and the use of greater overlay thicknesses have resulted in better performance from crack and seat on later projects. The term “breaking and seating,” rather than cracking and seating, is applied to the technique of fracturing a jointed reinforced concrete pavement prior to placement of an AC overlay. In general, breaking and seating has been found to be less effective at reflection crack control than cracking and seating because of the difficulty of ensuring that the reinforc- ing steel in the concrete is completely ruptured in the process of breaking the slab. An example of a successful application of this technique as a long-life HMA pavement is the California Interstate 710 (Figure A.7) in Los Angeles County, known as the Long Beach Freeway, with a design lane traffic of 100 million to 200 million ESALs for a 40-year period. AC over Rubblized Concrete Pavement Rubblization originally developed as an improvement in reflection crack control over cracking and seating. The LTPP SPS-6 experiment includes several rubblized sections built as supplements to the main experimental test sections. At AC overlay thicknesses comparable to those built on crack and seat projects, rubblizing projects typically are expected to provide about 5 to 10 years of additional service life. How- ever, in recent years the two U.S. manufacturers of concrete pavement rubblizing equipment (Antigo and PB4) have both been involved in rubblizing projects in which a much more substantial thickness of AC, e.g., 15 in. or more, has been placed. Such structures are in essence full-depth AC pave- ments on high-strength granular bases, and thus it appears reasonable to expect that they are viable candidates for long- life in-place renewal projects. (a) (b) Source: National Highway Institute 2003. Figure A.5. Crack of pavement with guillotine hammer. (a) Guillotine hammer. (b) Fractured slab with guillotine hammer. Source: National Highway Institute 2003. Figure A.6. Heavy roller used to seat the cracked pavement.

47 Rubblizing involves breaking the existing concrete pave- ment into pieces, and thereby destroying any slab action, and overlaying with HMA. The sizes of the broken pieces usually range from 2 to 6 in. (51 to 152 mm) (Asphalt Pavement Alli- ance 2002). The technique is suitable for both JPCPs and jointed reinforced concrete pavements (JRCPs). It has also been used on severely deteriorated CRCPs, although the heavy reinforcement in the CRCP presents some challenges and requires extra care in quality control/quality assurance (QC/QA) procedures. A rubblized PCC pavement behaves like a high-quality gran- ular base layer. This loss of structure must be accounted for in the HMA overlay design thickness. A study by NAPA indicated that strength of the rubblized layer is one and a half to three times greater than a high-quality dense graded crushed stone base (National Asphalt Pavement Association 1994). Rubblization is considered a viable, rapid, and cost-effective rehabilitation option for deteriorated PCC pavements. Good performance of rubblized pavements requires a high-quality process of rubblization, effective rubblizing equipment, and maintaining a strong base and/or subgrade soil. Also, poor performance can occur when the underlying soils are satu- rated. Installation of edge drains prior to rubblization has proven to be successful for this type of condition. If the exist- ing concrete pavement is deteriorated due to poor subgrade support, then rubblization may not be a viable option. Two types of equipment are used in the rubblization process: (1) a resonant breaker and (2) a multiple-head breaker. The resonant rubblizer (Figure A.8) is composed of a sonic shoe (hammer) located at the end of a pedestal, which is attached to a beam whose dimensions vary from one machine to another, and a counterweight situated on top of the beam. The principle on which the resonant breaker operates is that a low-amplitude (about 0.5-in.) high-frequency resonant energy is delivered to the concrete slab, which causes high tension at the top. This causes the slab to fracture on a shear plane inclined at about 35° from the pavement surface. Sev- eral equipment variables affect the quality of the rubblization process, including shoe size, beam width, operating fre- quency, loading pressure, velocity of the rubblizer, and the degree of overlapping of the various passes. The rate of pro- duction depends on the type of base or subbase material and is approximately 1.0 to 1.5 lane miles per day. During its operation, a resonant rubblizer encounters dif- ficulty in the vicinity of pavement discontinuities such as joints or cracks. At a discontinuity, the microprocessor con- troller increases the rubblizer speed, causing a decrease in the energy delivered to the concrete, or it causes a shutdown. Bituminous patches or unmilled overlays can also be prob- lematic, because the shoe penetrates the asphalt, causing a large loss in the energy delivered to the concrete. Last, the type of base or subbase material, the roadbed or subgrade soil, and the condition of the concrete pavement being rub- blized all affect the quality of the rubblized product. For example, if the base or subbase materials are softer than the roadbed soil, shear failure may result. The multihead breaker operation includes multiple drop hammers arranged in two rows on a self-propelled unit and a vibratory grid roller (Figure A.9). The hammers strike the pavement approximately every 4.5 in. The bottom of the hammer is shaped as to strike the pavement on 1.5-in.-wide and 8-in.-long loading strips. The hammers in the first row Source: TRB 2001. Figure A.7. Cross-sectional design for the I-710 Freeway.

48 strike the pavement at an angle of 30° from the transverse direction. The hammers in the second row strike the pave- ment parallel to the transverse direction. The sequence of hammer drops is irregular because each cylinder is set on its own timer or frequency system. By disabling some cylinders, the width of the rubblized area can be varied from 2.5 to 12.67 ft. Typically, a 10-ton vibratory grid roller follows the multihead breaker to reduce the size of the broken concrete. The rate of production of the multihead breaker depends on the type of base or subbase material and is about 0.75 to 1 lane mile per day. Several variables affect the rubblization process, including speed, height, weight, and frequency of the drop hammers. The multihead breaker encounters difficul- ties on weak or saturated subbase and/or roadbed soil, which fail in shear, causing large concrete pieces to rotate and/or penetrate the underlying material. Such failure would result in poor pavement performance. Examples of successful application of the rubblization technique as a long-life HMA pavement include (1) I-440, Raleigh Beltway, North Carolina (average daily traffic (ADT) > 100,000); (2) I-65, Alabama; and (3) I-496 near Lansing, Michigan. Figure A.10 shows example design cross sections for long-life performance of HMA over rubblized concrete pavements developed by Von Quintus for the Michigan APA (Asphalt Pavement Alliance 2002). Thompson has demonstrated that a mechanistic-empirical approach to evaluation of the structural capacity of in-service asphalt pavement can be used to determine the required overlay (a) Sources: (a) Karim Chatti; (b) National Highway Institute 2003. (b) Figure A.8. Resonant frequency pavement breaker. (a) Resonant breaker machine. (b) Close-up of the sonic shoe. (a) (b) Figure A.9. (a) Multihead breaker. (b) Grid roller.

49 thickness for rubblized concrete pavements (Thompson 1999). An algorithm to predict the tensile strain at the bottom of the asphalt overlay as a function of a deflection basin parameter, called the area under the pavement profile (AUPP), has been validated with measurements from instrumented full-depth and conventional flexible pavements. Falling weight deflec- tometer (FWD) data from rubblized concrete pavements with asphalt concrete overlays were used to develop a relationship between AUPP and an overlay stiffness parameter (Eh3, where E is the asphalt concrete modulus and h is the asphalt overlay thickness). The estimated strain is an input to an asphalt con- crete fatigue model. The asphalt overlay thickness is selected to limit the asphalt concrete tensile strain to an acceptable level. The NCHRP 1-37A MEPDG procedure’s software for design of AC overlay of rubblized PCC allows the user to select some or all of the following performance criteria by which the adequacy of a trial overlay design is judged: • Rutting, • Alligator cracking, • Longitudinal cracking, • Transverse cracking, and • Smoothness. In the MEPDG software, the elastic modulus of the rub- blized PCC is assigned a modulus of 150 ksi for Level 3 design (the simplest approach, requiring the fewest and simplest user inputs). For Level 1 design (the most sophisticated approach, requiring the most numerous and precise user inputs), how- ever, the rubblized PCC modulus may be assigned a value from 300 to 600 ksi, depending on the expected level of con- trol on the breaking process, and the anticipated coefficient of variation of the fractured slab modulus. Criteria for Long-Life AC Renewal Approaches For asphalt concrete pavements, achieving long life requires the combination of a rut- or wear-resistant top layer with a rut-resistant intermediate layer and a fatigue-resistant base Source: Asphalt Pavement Alliance 2002. Figure A.10. Michigan design catalog for long-life HMA pavements over rubblized concrete.

50 layer, as illustrated in Figure A.11 (Newcomb, Buncher, and Huddleston 2001). This requires a high-quality HMA wearing surface or an open graded friction course, a thick, stiff dense graded inter- mediate layer, and a flexible (asphalt-rich) bottom layer. In addition, the pavement foundation must be strong enough to satisfy the limiting strain criteria. Suggested values for the horizontal tensile strain at the bottom of the AC layer and ver- tical subgrade strain are 65 and 200 microstrains, respectively. The value for the endurance limit of the tensile strain at the bottom of the AC layer is still debated. Original work by Moni- smith and others suggests a value of 65 microstrains (Fig- ure A.12). Others believe that this value is too conservative, and that a higher value (100 to 120 microstrains) should be used to ensure that the AC renewal solution is economical. When applied to existing pavements, a fourth condition is added: the inhibition of reflective cracking. This is true regardless of the existing pavement type (i.e., distressed HMA Source: Newcomb et al. 2001. Figure A.11. Long-life HMA pavement design concept. Source: Thompson and Carpenter 2006. Figure A.12. Endurance fatigue limit for long-life AC pavements.

51 or PCC), although experience shows that reflective cracking can be more predominant when the existing pavement is a PCC pavement. Reflection cracking can occur in an HMA overlay over any joint or crack in the PCC pavement. The current state of the art does not provide accurate methods to predict the occurrence and growth of the reflection crack. Figure A.13 schematically illustrates reflection crack distress in an HMA overlay placed over a joint or crack of an existing PCC slab. Figure A.14 illustrates the mechanism through which the crack develops and propagates in the HMA layer (National Asphalt Pavement Association 1994). PCC slabs expand and contract with seasonal changes in temperature. This movement causes the development of forces at the bottom of the HMA layer as shown in Fig- ure A.14, part A. The combination of forces at the bottom of the HMA overlay will eventually cause the development of a microcrack at the bottom of the HMA overlay, as shown in B. With time, this microcrack will grow and eventually reflect upward to the surface of the HMA overlay, as shown in C and D. As temperature and loading cycles continue, multiple cracks will form and eventually result in significant deteriora- tion of the HMA surface, as shown in E and F. Figure A.15 illustrates a distressed reflection crack area in an HMA over- lay over an existing PCC pavement. Existing CRCP is an excellent foundation for a new long- life HMA pavement since reflection cracking is not a problem as long as cracks are of low severity and failed areas (punch- outs and deteriorated cracks) are repaired prior to overlaying. Pavements with D-cracking are not good candidates for HMA overlays without slab fracturing. Studies have shown that the placement of HMA overlay can accelerate D-cracking, and field data showed poor performance of HMA overlays of con- crete pavement with D-cracking (Liu et al. 2003). Source: National Asphalt Pavement Association 1994. Figure A.13. Schematic representation of a reflection crack. Source: National Asphalt Pavement Association 1994. Figure A.14. Growth of a reflection crack.

52 Because the pavement foundation is critical to the con- struction and performance of a long-life HMA pavement, the question of whether an existing pavement can be used in place largely depends on the quality of the existing founda- tion. A careful consideration of the existing condition of the pavement foundation must therefore be made. This is in light of the fact that there will be cases where the condition of the existing subgrade does not warrant using the existing pave- ment in place (e.g., drainage problems or soft layer underneath existing pavement structure). Several end-result specifications for the foundation layers have been used in Europe (United Kingdom, France, and Germany), requiring a minimum modulus under FWD loading or imposing a maximum toler- able surface deflection (Newcomb et al. 2001). The state of Illinois requires a minimum CBR- or DCP-based cone index value below which the subgrade soil must be modified (using lime treatment), as shown in Figure A.16. Overlay Design Approaches for AC Surfaced Pavements The two most commonly used approaches to structural design of asphalt overlays of asphalt pavements are (1) the structural deficiency approach, exemplified by the 1993 AASHTO proce- dure (AASHTO 1993) and (2) the deflection-based approach, exemplified by the Asphalt Institute procedure (Asphalt Insti- tute 1999). Much less common is the mechanistic approach, in which fatigue and rutting performance are predicted using mechanistic-empirical models (Hall et al. 2001). In a mechanistic-empirical approach to design of asphalt overlays of asphalt pavements, performance of the overlay is Source: Martin 1973. Figure A.15. Reflection cracking in HMA overlay over PCC pavement. Source: Illinois Department of Transportation 1982. Figure A.16. Illinois granular thickness requirement for foundations.

53 predicted using mechanistic-empirical distress models. The distresses considered should include at least fatigue cracking, and ideally rutting and thermal cracking as well. The existing pavement layers and foundation are characterized using non- destructive deflection testing and backcalculation of their elastic moduli. Material properties for the overlay are assumed. The overlay thickness that will yield acceptable performance in terms of the distresses considered is determined by itera- tion. A conceptual overview of the mechanistic-empirical approach to design of asphalt overlays of asphalt pavements is given by Monismith (1992). The individual tools used in mechanistic-empirical design of asphalt pavements (e.g., fatigue models, rutting models, seasonal adjustment) can be adapted to some extent to design of asphalt overlays. However, there are additional aspects of the problem that need to be considered to develop a full design procedure for asphalt overlays of asphalt pavements. Among these are consideration of the extent, type, and qual- ity of preoverlay repairs, prediction of reflection crack propa- gation and deterioration (a problem for asphalt overlays of both asphalt and concrete pavements), and calibration of asphalt overlay performance prediction models to the observed performance of asphalt overlays. Several examples of mechanistic-empirical procedures for design of asphalt pavements exist, such as the Shell procedure (Shell International Petroleum Company 1978), the Asphalt Institute procedure (1981; Shook et al. 1982), the NCHRP 1-26 procedure (Thompson 1989), and the MEPDG procedure developed under NCHRP 1-37A (Applied Research Associ- ates 2004). Fewer examples exist, however, of mechanistic- empirical procedures for design of asphalt overlays of asphalt pavements. The NCHRP 1-37A MEPDG procedure’s software for design of AC overlay of AC allows the user to select some or all of the following performance criteria by which the adequacy of a trial overlay design is judged: • Rutting, • Alligator cracking, • Longitudinal cracking, • Transverse cracking, and • Smoothness. According to the MEPDG, “the models used for the predic- tion of structural distresses (i.e., excluding smoothness pre- diction) in the overlaid pavement are basically the same as those described in Part 3, Chapter 3 [for design of new AC pavements] with some modifications to the rates of distress accumulation in the existing layers.” The smoothness parameter used for AC overlays of AC pavements in the MEPDG methodology is the IRI, predicted from an empirical model as a function of the existing pave- ment’s IRI at the time of overlay placement, the time elapsed since placement of the overlay, the percent of the wheelpath area with fatigue cracking, the average spacing of medium- and high-severity transverse cracks, the length of medium- and high-severity sealed longitudinal cracks in the wheelpath, the percent of the total lane area with medium- and high-severity patches, and the percent of the total lane area with potholes. Among the few state departments of transportation (DOTs) that have developed a mechanistic-empirical design proce- dure for asphalt overlays of asphalt pavements are Washington (Mahoney et al. 1989), Idaho, and Nevada (Nevada Depart- ment of Transportation 1996; Sebaaly et al. 1996). The Wash- ington State DOT procedure uses a model to predict fatigue as a function of horizontal tensile stress at the bottom of the asphalt overlay and at the bottom of the original asphalt layer, as well as a model to predict rutting as a function of vertical compressive stress at the top of the subgrade. The critical stress locations considered are illustrated in Fig- ure A.17. A flowchart of the Washington State procedure is illustrated in Figure A.18. The overlay thickness required to keep fatigue and rutting below critical levels is determined through a process of iteration. Figure A.19 compares traditional mechanistic-empirical (M-E) design to long-life pavement design. The basic concept in designing long-life AC pavements is to use limiting strain criteria (see Figure A.19b). Structural Design of AC Overlay over Fractured Slab The approach taken in the 1993 AASHTO Guide to design of asphalt overlays of fractured slabs (both crack and seat and rubblizing) is a structural deficiency approach. The overlay must satisfy the deficiency between the structural number (SNf) required to support traffic over some future design period, and the effective structural number (SNeff) of the existing pavement (after fracturing). Perhaps the most contentious aspect of overlay design for fractured slabs by the structural deficiency approach is what structural coefficient should be assigned to the fractured slab. The 1993 AASHTO Guide recommends the following ranges for structural coefficients for different types of slab fracturing: • Rubblized: 0.14–0.30, • Crack and seat: 0.20–0.35, and • Break and seat: 0.20–0.35. Other recommendations for overlay design for fractured slabs, including recommended ranges of structural coeffi- cients and overlay thickness design tables, have been devel- oped by NAPA. A study done for the ACPA recommended a range of 0.15 to 0.25 for the structural coefficient of all three types of fractured slabs (Hall 1999).

54 A mechanistic procedure for design of AC overlays of cracked and seated concrete pavements was developed by Thompson at the University of Illinois as part of the FHWA/ Illinois DOT study Mechanistic Evaluation of Illinois Flexible Pavement Design Procedures. For a given overlay thickness, the required inputs are the design AC elastic modulus, the subgrade resilient modulus, and the “equivalent modulus” of the cracked and seated concrete. In the development of the design procedure, the finite ele- ment program ILLI-PAVE was used to estimate the asphalt concrete bending strain for a range of overlay thicknesses. Transfer functions for the number of repetitions to failure for a given bending strain were developed for typical Illinois DOT Class I asphalt concrete mixtures (Schutzbach 1988, 1989). Additional guidance on the use, design, and construction of AC overlays of cracked and seated PCC pavements is given by Thompson in NCHRP Synthesis No. 144 (Thompson 1989). Ahlrich has documented the use of FWD testing on intact PCC slabs and testing after cracking and seating and overlay- ing with AC to determine the “effective modulus” of the cracked and seated PCC layer (Ahlrich, 1989). In field studies conducted by the U.S. Army Corps of Engineers Waterways Experiment Station, at the Rock Island Arsenal in Illinois and Fort Wainwright in Alaska, concrete slabs with an elastic modulus of about 6 million psi were reduced by cracking and seating to a fractured concrete layer with an effective elastic modulus of about 1 to 1.5 million psi. Similar results from analysis of FWD deflections measured on test sections at the LTPP SPS-6 test site on I-57 in Illinois have been reported by Hall (1991). The NCHRP 1-37A MEPDG procedure’s software for design of AC overlay of cracked and seated PCC allows the user to select some or all of the following performance criteria by which the adequacy of a trial overlay design is judged: • Rutting, • Alligator cracking, • Longitudinal cracking, • Transverse cracking, and • Smoothness. Source: Washington State Department of Transportation 2005. Figure A.17. Critical stress locations considered in Washington State DOT overlay design procedure.

55 Source: Washington State Department of Transportation 2005. Figure A.18. Washington State Department of Transportation overlay design procedure flowchart. Source: Timm 2005. (a) (b) Figure A.19. Traditional versus long-life AC pavement design. In the MEPDG software, the elastic modulus of the cracked and seated PCC is assigned as a function of the crack spacing (i.e., 200 ksi for 12-in. spacing, 250 ksi for 24-in. spacing, and 300 ksi for 36-in. spacing) for Level 3 design (the simplest approach, requiring the fewest and simplest user inputs). For Level 1 design (the most sophisticated approach, requiring the most numerous and precise user inputs), however, the rubblized PCC modulus may be assigned a value from 300 to 600 ksi, depending on the expected level of control on the breaking process and the anticipated coefficient of variation of the fractured slab modulus. Renewal of Rigid Pavements When in-place renewal of an existing PCC pavement is con- sidered, the structural design considerations that must be taken into account to ensure good long-term performance are the adequacy of the subgrade, protection of the subgrade from excessive deformation, limiting strain in the existing PCC, limiting stress and strain in the new AC or PCC surface, and minimizing reflection cracking in the new surface. While AC overlay is undoubtedly the most common major rehabilitation method for jointed PCC pavements, the service life of this technique is limited by the rate at which reflection cracks develop and deteriorate to unacceptably rough levels. Thus, an AC overlay of a jointed PCC pavement is typically considered a conventional rather than a long-life rehabilitation approach, with an expected service life of about 10 to 15 years. However, exceptions exist: Iowa, for example, has experi- ence with jointed PCC pavements built in the 1930s and 1940s, widened with PCC or AC from 18, 20, or 22 ft to 24 ft the 1970s, and then overlaid over time with a total of five or more inches of AC. Now, some 30 years later, these old AC/ PCC pavements are being widened again, to 28 or 32 ft, and

56 are being overlaid with PCC (J. K. Cable, personal communi- cation, 2008). The most promising long-life rigid pavement methods, however, appear to be the following: • AC over CRCP, • AC over cracked and seated JPCP, • AC over rubblized PCC, • Unbonded PCC over PCC, and • Bonded PCC over PCC. Definition of Long-Life Concrete Pavements Long-life concrete pavements (LLCPs) have been quite attain- able for a long time in the United States, as evidenced by the number of very old pavements that remain in service; how- ever, recent advances in design, construction, and concrete materials technology give engineers the knowledge and tech- nology needed to consistently achieve what they know to be attainable. A working definition of long-life concrete pave- ment in the United States is summarized as follows (Tayabji and Lim 2007): • Original concrete service life is 40+ years. • Pavement will not exhibit premature construction and materials-related distress. • Pavement will have reduced potential for cracking, fault- ing, and spalling. • Pavement will maintain desirable ride and surface texture characteristics with minimal intervention activities, if war- ranted, for ride and texture, joint resealing, and minor repairs. The quest for long-life concrete pavements necessitates a much better understanding of design and construction factors that affect both short-term and long-term concrete pavement performance. Essentially, this requires a better understanding of how concrete pavements deteriorate or fail. Concrete pave- ments deteriorate over a period of time as a result of distresses that develop due to a combination of traffic and environmen- tal loading. Typical distresses that can develop include the following: 1. Cracking: Typically transverse cracking occurs, but longi- tudinal, random, and corner cracking may also develop due to poor design and construction practices. Cracking is typically referred to as a stress-based distress. 2. Joint faulting: Joint faulting may develop with or without outward signs of pumping. Faulting is typically referred to as a deflection-based response. Joint faulting is signifi- cantly affected by the type of load transfer provided at transverse joints. 3. Spalling: Spalling may develop along joints or cracks and may be caused by poor joint-sawing practices, incom- pressible materials in joints or cracks, winter snow removal operations, or poor-quality concrete. 4. Materials-related distress: The more significant materials- related distresses may include alkali-silica reactivity and D-cracking in freezing environments. 5. Roughness: The lack of pavement smoothness, or rough- ness, is affected by the development of various distresses in the concrete pavement, as listed in items 1 through 4 above. The effect of each distress type is additive and results in pavement roughness over a period of time. Some pave- ment roughness is also built in during construction. Initial pavement smoothness is needed so that the pavement does not become prematurely rough. Construction specifica- tions typically utilize incentives and disincentives to con- trol new pavement smoothness. 6. Texture loss: Although not conventionally considered a distress, texture loss is a significant distress for pavements in high-volume, high-speed applications. It is realized that it would be impossible or impractical to design and construct concrete pavements that exhibit very little or no distress. Distress development over the pavement’s service life is expected. However, the rate of distress develop- ment is managed by incorporating sound designs, durable paving materials, and quality construction practices. Gener- ally recognized threshold values in the United States for dis- tresses at the end of the pavement’s service life are listed in Table A.2 for JPCPs and CRCPs. Unbonded PCC over PCC An unbonded PCC overlay (sometimes called a separated overlay) contains an interlayer between the existing PCC pavement and the new PCC overlay (Figure A.20). Unbonded overlays of all types (jointed plain, jointed reinforced, and continuously reinforced) can be placed on all types of concrete Table A.2. Threshold Values for Long-Life Concrete Pavement Distresses Distress Threshold Value Cracked slabs, % of total slabs (JPCP) 10–15 Faulting, mm (in.) (JPCP) 6–7 (0.25) Smoothness (IRI), m/km (in./mi) (JPCP and CRCP) 2.5–3.0 (150–180) Spalling (length and severity) (JPCP and CRCP) Minimal Materials-related distress (JPCP and CRCP) None Punchouts, no./km (no./mi) (CRCP) 10–12 (12–16)

57 pavements, including those with existing asphalt overlays. Unbonded concrete pavements are appropriate for pavements with little or no remaining structural life and/or extensive and severe durability distress. Unbonded concrete overlays require little or no preoverlay repair and are thus well suited to badly deteriorated concrete and asphalt-overlaid concrete pavements. An unbonded concrete overlay is an attractive alternative to reconstruction when construction duration is a pressing issue (e.g., for high traffic volumes and/or very poor subgrade conditions). Jointed unbonded PCC overlays of PCC highway pave- ments have been built in the United States since the 1920s. The first unbonded continuously reinforced concrete (CRC) overlay of an existing jointed PCC highway pavement in the United States was constructed in Texas in 1959 (Martin 1973). In subsequent years CRC overlays were placed on hundreds of miles of both asphalt and jointed concrete pavements. Illinois built its first experimental test sections of CRC overlay on jointed reinforced PCC pavement in 1967 (Dhamrait and Schwartz 1978). Georgia built its first CRC overlay of a jointed plain PCC pavement in 1973 (Tyner, Gulden, and Brown 1981). The first unbonded CRC overlay of an existing CRC highway pavement in the United States was constructed on I-59 in Mississippi in 1982 (Crawley 1982). There is little doubt that unbonded concrete overlays, be they jointed or CRC, are substantial pavement structures with expected performance characteristics as good as or better than new concrete pavement construction. They are essentially new concrete pavements on high-quality founda- tions, and the consensus from past field studies is that, as long as an adequate separation layer is used, their performance is fairly insensitive to the condition of the overlaid pavement. Thus, they are certainly viable candidates for long-life in- place renewal projects. To date, unbonded overlays have typi- cally been designed for service lives in the range of 20 to 30 years. The PCC overlay thickness design approaches, slab thicknesses, and other design details required to achieve ser- vice lives of 40 or 50 years needs to be studied (Hall, Darter, and Seiler 1993). Traditionally, unbonded concrete overlays have been designed using some form of the familiar “square root” equa- tion shown below: 2 2h h hol f eff= − where hol = unbonded overlay thickness, hf = required slab thickness for future traffic, and heff = effective thickness of the existing slab. The square root equation dates back to the Bates Road Test in the 1920s, and its use in unbonded overlay design proce- dures started in the 1940s (Older 1924). Full-scale field tests of concrete overlays conducted by the Corps of Engineers in the 1940s and 1950s indicated that the square root equation yielded conservative results (Mellinger 1963). Although many engineers have the impression that the square root equation (also called the Corps of Engineers equation) for unbonded overlay design is completely empiri- cal, it has a theoretical basis. Several researchers have demon- strated that an overlay slab and a base slab can be represented by an equivalent single slab in a variety of ways—for example, equivalent surface deflection, equivalent tensile stress in the overlay slab, and equivalent tensile stress in the base slab. There are, however, some important limitations to the char- acterization of an unbonded overlay and base slab as an equiv- alent single slab. The Corps of Engineers square root equation is a simplified form of the equations for stress in either the base slab or the overlay slab equivalent to stress in the equivalent single slab. This simplified equation is only valid when the two slabs are equal in thickness and equal in elastic modulus. Another important limitation to characterizing an unbonded overlay slab and base slab in terms of an equivalent single slab is that it assumes full contact between the overlay and base slabs. They may bend independently, but they must have the same radius of curvature. To whatever extent the overlay slab curls and/or warps to a different shape than the underlying slab, it will experience different, and in some cases much greater, stresses under combined load and curling than the equivalent thickness concept implies. The third major limitation of the Corps of Engineers equa- tion is the structural deficiency concept itself, namely, the assumption that an overlay satisfies a structural deficiency between a required single slab thickness and an existing slab’s effective (i.e., damage-adjusted) thickness. As can be seen by examining the square root equation, the structural deficiency concept implies that for a given required slab thickness for future traffic, a thicker existing pavement will require a thinner unbonded overlay than a thinner existing pavement in the same condition. Conversely, it implies that a given thickness of unbonded overlay will perform better on a thicker existing pave- ment than on a thinner existing pavement in the same condition. Source: McGhee 1994. Figure A.20. Typical cross section of an unbonded PCC overlay.

58 Field observations do not support the implication that un- bonded overlay performance is as sensitive to existing pave- ment thickness as the structural deficiency concept suggests. One alternative to the Corps of Engineers equation for design of an unbonded overlay is to design the overlay as if it were a new pavement, with the existing pavement structure characterized as a foundation for the new slab. The elastic modulus, modulus of rupture, and load transfer coefficient inputs to the design model are typically the anticipated values for the overlay slab. Two key differences exist between this approach and the Corps of Engineers approach. The first dif- ference is that the existing pavement is not considered to con- tribute any structural capacity to the total structural capacity of the overlaid pavement. The existing pavement is instead considered a foundation for the new slab. This leads to the second major difference between the two methods. The k value of the foundation beneath the existing pavement is used to determine the required future slab thickness in the Corps of Engineers method, whereas the new pavement design method requires a k value beneath the overlay. The major dif- ficulty in application of the new design approach thus lies in selection of an appropriate design k value (Barenberg 1981). Conventional practice in concrete pavement design for many years has been to assign a k value to a granular or stabi- lized base that is considerably higher than the k value of the subgrade and which was a function of the thickness and stiff- ness of the base layer. This convention is still employed for new concrete pavements in the 1993 AASHTO Guide and Portland Cement Association design procedures. Following this logic, an existing concrete pavement with an asphalt con- crete surfacing for a separation layer would be assigned a very high k value, such as 500 psi/in. or more for unbonded over- lay design. However, backcalculation results indicate that when an unbonded overlay is designed as a new pavement with the existing pavement as its foundation, it is neither nec- essary nor appropriate to use an extremely high k value such as 500 psi/in. or more. A design static k value in the range of 200 to 400 psi/in. is probably appropriate in most cases. Whenever possible, deflections should be measured on the existing pavement prior to overlay to backcalculate a dynamic k value for the existing foundation and to estimate from this a reasonable static k value for design. Another issue that should be considered is the effect of curling on performance. If a jointed overlay slab is designed as a new pavement with the existing pavement serving as its foundation, it will experience much higher curling stresses than a conventional concrete pavement on a weaker founda- tion (Voigt, Darter, and Carpenter 1989). These higher curl- ing stresses may be computed using finite element analysis or available equations. However, if the performance model used to determine the required slab thickness was developed for concrete pavements on weak foundations, the detrimental effect of high curling stress will not be adequately reflected in the predicted performance of the overlay. This would be the case if, for example, the 1993 AASHTO design procedure was used to determine the required slab thickness rather than a fatigue analysis that directly considered the combined effects of load and curling. Either increased slab thickness or reduced joint spacing may be necessary to achieve the performance from the unbonded overlay that is predicted by the model. Other alternatives to unbonded overlay design involve mod- eling the overlay and existing slab as either two elastic layers or two plates on a foundation. This is arguably the most realistic of the three design approaches described here, but also the most difficult. The basic approach is the same as for design of the overlay as a new pavement, except that the existing pave- ment structure is characterized more realistically, not as a uni- form foundation but as a multilayered system. Among the difficulties associated with this approach are the following: • Characterization of the existing slab, including deciding how (if at all) to account for existing deterioration; • Identifying the important structural responses (e.g., over- lay stress, overlay deflection, original slab stress); and • Identifying the important performance criteria (e.g., crack- ing in the original slab and/or cracking in the overlay slab). In jointed unbonded concrete overlays, the joints should be spaced more closely than they would be in a new pavement on a granular base, and the overlay’s transverse joints and the old pavement’s transverse joints should be mismatched to improve load transfer across the overlay joints. Mismatching the joints by at least 1 ft is advisable; several agencies specify a mismatch of 3 ft. According to the ACPA, dowels are not considered necessary for jointed unbonded overlays less than 8 in. thick. For overlays 8 to 9 in. thick, 1.25-in.-diameter dowels are recommended, and for overlays greater than 9 in. thick, 1.5-in.-diameter dow- els are recommended. The ACPA also provides guidelines for constructing transitions between unbonded concrete overlays and existing or reconstructed pavement sections. Additional information on the design and performance of unbonded concrete overlays is provided in NCHRP Synthesis 99, Resurfacing with Portland Cement Concrete (Hutchinson 1982); NCHRP Synthesis 204, Portland Cement Concrete Resur- facing (McGhee 1994); the ACPA’s Guidelines for Unbonded Concrete Overlays (1990); the Portland Cement Association’s Guide to Concrete Resurfacing Designs and Selection Criteria (1981); and NCHRP Report 415, Evaluation of Unbonded Port- land Cement Concrete Overlays (ERES Consultants 1999). The performance of unbonded PCC overlays of existing PCC pavements depends significantly upon obtaining effective separation between the two layers. Since the unbonded PCC overlays are placed on PCC pavements in a more advanced state of deterioration, distresses from the underlying pavement

59 can potentially reflect through the new overlay and compro- mise its performance. Typically, a fine-graded asphalt surface mixture is used for the separator layer. The thickness of the separator layer is a function of (1) the condition of the existing pavement and (2) the type of preoverlay repairs. Based on the review of the literature a minimum thickness of 1 in. is recom- mended for HMA separator layers. Thinner layers erode easily near joints and do not provide adequate isolation of the overlay from underlying PCC pavement. The separator layer is not intended to provide structural enhancement; therefore, the placement of an excessively thick layer should be avoided. Some state DOTs have modified the asphalt mixture because their surface mixes were not stable and were prone to scouring, particularly under heavy truck traffic. In an effort to reduce the scour pore pressure and increase stability, the sand content was reduced and the volume of 3/8-in. (9.5-mm) chip aggregate was increased (National Concrete Pavement Technology Cen- ter 2007). This modified mixture has a reduced unit weight and lower asphalt content. Other bituminous surface treatments such as slurry seals, cutbacks, and emulsions have been used for low-volume roads. In Germany, lean concrete is used as an interlayer. This is done in conjunction with breaking or fracturing the exist- ing pavement before overlaying the lean concrete interlayer. In addition to this process the interlayer is jointed to match the joints of the overlay. Belgium is the only country outside the United States iden- tified in this review as having reported appreciable experience with unbonded concrete overlays (Hall et al. 2007). Belgium constructed its first concrete overlay in 1960, over a concrete pavement originally constructed in 1934. The jointed con- crete overlay was constructed of 7-in.-thick reinforced con- crete slabs. Figure A.21 shows the overlay still in service nearly 45 years later. Source: Hall et al. 2007. Figure A.21. Belgium’s first concrete overlay after 45 years in service. Source: Hall et al. 2007. Figure A.22. CRC overlay construction on E40/A10 in Belgium. Source: Hall et al. 2007. Figure A.23. CRC overlay paving on E40/A10 in Belgium. Construction of a concrete overlay on the E40/A10 road from Brussels to Ostende in Belgium is shown in Figure A.22. Two mobile concrete plants were used to produce the 2,600 cubic yards of concrete a day required for this project. The average paving rate was 3,900 ft per day, 24 ft wide. A closer view of the paver is shown in Figure A.23. Due to the very tight schedule for this project, concrete was placed with- out interruption, 24 hours a day, 7 days a week. As a result, the CRC overlay has no construction joints. A slipform paver was also used to construct the safety barriers on this job, as shown in Figure A.24.

60 Bonded PCC over PCC Bonded PCC overlays of PCC are generally not considered very long-life pavement rehabilitation techniques because of their sensitivity to the condition of the underlying pavement and the difficulty of achieving the long-lasting bond neces- sary for composite bending action. Bonded concrete overlays are not often used, because they perform best on pavements in good to fair condition, that is, pavements that are not in urgent need of rehabilitation. The bonded concrete surface is bonded to the existing con- crete pavement to form a monolithic section. This renewal strategy has the potential to increase the structural capacity of an existing concrete pavement or to improve the overall ride quality. The bonded concrete surface is typically 2 to 5 in. thick. The bonded concrete surface works best when the existing pavement is free of structural distress and in rela- tively good condition. This rapid renewal strategy is typically attractive when vertical clearances must be met, or in mill and inlay sections, or in conjunction with widening projects. The achievement of an effective bond between the existing pave- ment and the new surface is critical in ensuring satisfactory performance of the bonded concrete surface. The use of “bonding agents” and “direct placement” are two methods that are practiced for this type of rehabilitation. Figure A.25 shows a cross section of a typical bonded PCC overlay. The service life of a bonded PCC overlay of a PCC highway pavement is typically estimated at about 15 to 25 years at best. However, in the course of the work done for Task 1 of this study, a bonded PCC overlay recently constructed in Okla- homa was identified as one that is expected by some to be very capable of providing 40 years of service or more. The design and construction details of this project warrant study to gain insight into whether, and under what conditions, bonded PCC overlays might be viable candidates for long- life in-place renewal projects. Bonded PCC overlays have also been constructed in many different states, including California, Illinois, Iowa, Louisiana, New York, Pennsylvania, South Dakota, Texas, and Virginia, as well as in the countries of Belgium, Canada, Japan, and Sweden. By far the most common bonded PCC overlay type is JPCP, and these over- lays have been placed on existing JPCP, JRCP, and CRCP designs (Sebaaly et al. 1996). Some bonded JRCP overlays have been used on existing JPCP and JRCP, although pres- ently they are rarely used. Texas and Virginia have both con- structed several bonded overlays on existing CRCP. For bonded PCC overlays of existing PCC pavements, achieving the bond between the two layers is critical mono- lithic slab behavior. To help achieve this, many state highway agencies place either a cement grout or an epoxy resin on the existing PCC pavement just ahead of the paver. Cement grouts are generally produced in a mobile mixer from a mix- ture of portland cement and water; the grout should have a maximum w/c of 0.62 (American Concrete Pavement Asso- ciation 1990). Epoxy bonding agents should be applied in accordance with the manufacturer’s instructions. Prior to the placement of either type of bonding agent, the pavement sur- face should have already been prepared and should be dry (American Concrete Pavement Association 1990). Renewal by Lane Replacement (Inlay) or Lane Addition When a lane replacement or lane addition is contemplated as an approach to in-place renewal of an existing AC or PCC Source: Hall et al. 2007. Figure A.24. Slipform paving of the safety barriers on the E40/A10 CRC overlay project. Source: McGhee 1994. Figure A.25. Typical cross section of a bonded PCC overlay.

61 pavement, the design considerations that must be taken into account to ensure good long-term performance include the adequacy of the foundation, the required thickness and any constraints on it, the method of connection to the adjacent lane, the design of transitions, and, in the case of widening, geometric considerations such as the availability of horizon- tal and vertical space for relocating shoulders, slopes, ditches, and/or drainage systems, interchanges, and bridges. Lane Replacement The evident viability of this technique as a long-life in-place renewal method seems at odds with the relatively little use that it has seen to date in the United States. When a portion of the thickness of an AC lane is milled out and replaced with PCC, it can be considered, and designed and constructed as, a conventional white-topping overlay. One caution, however, is that in some such applications of concrete inlays with undoweled joints, premature joint faulting has occurred and has been attributed to the “bathtub effect” of water collecting under the PCC overlay slab. The ACPA recommends that either doweled jointed PCC or CRC be used when construct- ing an inlay to replace a portion of the thickness of an AC traffic lane subjected to heavy traffic in one direction and wet climatic conditions. An inlay is a renewal option that involves the replacement of all or part of an existing pavement travel lane (or lanes) without significantly raising the surface eleva- tion. Inlays are practical for deteriorated concrete pavements. Single-lane and multilane inlays are common for concrete reconstruction. When a lane replacement or lane addition is contemplated as an approach to in-place renewal of an exist- ing PCC pavement, the design considerations that must be taken into account to ensure good long-term performance include the adequacy of the foundation, the required thick- ness and any constraints on it, the method of connection to the adjacent lane, the design of transitions, and, in the case of widening, geometric considerations such as the availability of horizontal and vertical space for relocating shoulders, slopes, ditches, and/or drainage systems, interchanges, and bridges. More information on design and construction of concrete inlays in existing AC or PCC pavements is provided in the ACPA publication Reconstruction Optimization Through Concrete Inlays (American Concrete Pavement Association 1993). Belgium’s experience with concrete inlays dates back to 1933 (ERES Consultants 1999). Concrete inlays in Belgium are con- structed with either JPCP or CRCP. Figure A.26 is a photo of a CRC inlay being placed on the A10 freeway in Belgium (Caestecker and Lonneux 2004). The roadway had three lanes in each direction, and the existing pavement was AC over JPCP. Rutting, reflection cracking, and roughness over AC patches in the PCC layer, particularly in the outer lanes, were resulting in steadily increasing annual maintenance costs. Lane Addition Although adding new lanes to an existing pavement struc- ture is also clearly a viable option for in-place long-life pave- ment renewal, it is costly and thus usually is only done when it is essential to increase the capacity of an existing roadway. In the course of the Task 1 work done for this study, several examples of lane addition projects on major highways in the eastern north-south corridor of the United States, including some that are currently under construction, were identified. Both the structural design aspects and the construction logistics aspects of such projects need to be studied to iden- tify the requirements for achieving good performance over an extended service life. Caestecker described an example of this type of work: the replacement of an outer AC shoulder with a fourth traf- fic lane on a heavily trafficked section of a six-lane highway on the A3 motorway toward Brussels, Belgium (Caestecker 1993). An important reason that the lane addition option was chosen was that highway noise is a significant environ- mental concern in Belgium, and designers were confident that a concrete-surfaced lane addition could achieve the capacity increase desired while minimizing the traffic noise generated by the roadway. The new lane was placed with a GOMACO slip form paver, operating at its capacity of 300 to 500 m per day. Immediately after paving, the surface was sprayed with a retarding agent and covered with plastic sheeting, to be brushed later to achieve the kind of exposed aggregate surface that has become popular in some European countries for both noise control and friction. The bituminous surface material salvaged from the old pavement shoulder was used in the cement-bound base layer of the new shoulder con- structed alongside the new traffic lane. Source: Hall et al. 2007. Figure A.26. CRC inlay construction in Belgium.

62 Concrete Overlay Materials Needed for Long Life Much of the emphasis in defining the characteristics of in- place pavement renewal options with the potential for service lives in the range of 50 years is necessarily on the structural design of the new material. Decisions regarding PCC mix materials are affected by the type of mixture—conventional or fast track (accelerated)—desired for a specific project. For the purpose of this report, only conventional PCC mixtures are discussed. Conventional PCC Mixtures for Overlay Construction Conventional concrete paving mixtures are typically used in the construction of concrete overlays. As with conventional concrete pavements, an effective mixture design is essential to the performance of a concrete overlay. Each component of the concrete mixture should be carefully selected so that the resulting mixture is dense, relatively impermeable, and resistant to both environmental effects and material-related chemical reactions over its service life. As Shilstone points out, thickness is only one of two key components of long- life pavement materials; the other is durability (Shilstone 1993). For example, in portland cement concrete, Shilstone identifies the following characteristics as key to long-term durability: • Low permeability is achieved with low total water, well- graded aggregate, good mixture rheology, and high in-place relative density. • Freeze-thaw resistance is achieved with closely spaced small air voids, ultimate compressive strength of 40 MPa (6,000 psi) or higher, well-graded aggregate, low permeability, and good curing. • Low shrinkage is achieved with low total water, low cement factor, low water-cement ratio, and minimal use of sharp and elongated particles. • Low reactivity is achieved with proper selection of cement type and aggregates, low permeability to reduce the poten- tial for water penetration, low water-cement ratio, and use of a properly selected pozzolanic material in the mix. • Abrasion resistance is achieved with compressive strength of 40 MPa (6,000 psi) or higher, well-graded aggregate, low water content, hard and dense aggregate, and air content appropriate for the exposure conditions. Most agencies specify a minimum concrete strength requirement for their pavements. Typical values include a 28-day compressive strength of 4,000 psi or a 28-day, third- point flexural strength of 650 psi (these specifications vary among state highway agencies). Cementitious Materials In general, Type I and Type II cements are commonly used in concrete mixtures for concrete overlay construction. The standard specification for portland cements used in the United States is presented in AASHTO M85 (ASTM C150). There are many references available that provide detailed descriptions of the physical and chemical characteristics of cements [e.g., Design and Control of Concrete Mixtures (Kosmatka et al. 2002)], which are not discussed further in this section. Depending on the mix design and strength requirements, cement content is typically in the range of 500 to 700 lb/yd3 (226.8 to 317.5 kg/m3), although higher content is sometimes used. The American Concrete Institute and Portland Cement Association provide guidelines for the selection of the appropriate w/cm ratio. A maximum w/cm ratio value of 0.45 is common for pavements in a moist envi- ronment that will be subjected to freeze-thaw cycles. How- ever, lower w/cm ratio values are used for concrete resurfacing to minimize drying shrinkage. As with conventional paving, supplementary cementitious materials (SCMs) normally improve durability and can improve construction. Aggregates To ensure long life of the overlay, these aggregates should pos- sess adequate strength and physical and chemical stability within the concrete mixture. All aggregates used in the produc- tion of PCC mixtures should conform to ASTM C33. Extensive laboratory testing or demonstrated field performance is often required to ensure the selection of a durable aggregate. For concrete resurfacing of concrete pavements, the types of aggre- gates in both the original pavement and the overlay should be similar so that the thermal expansion is similar. The coefficient of thermal expansion of concrete significantly influences joint design. It is therefore recommended that the coefficient of thermal expansion of concrete be measured in accordance with AASHTO TP60. The maximum coarse aggregate size used in concrete mixtures is a function of the pavement thickness or the amount of reinforcing steel. It is recommended that the largest practical maximum coarse aggregate size be used to minimize paste requirements, reduce shrinkage, minimize costs, and improve mechanical interlock properties at joints and cracks. Typically, maximum coarse aggregate sizes of ¾ to 1 in. (1.9 to 2.5 cm) have been common in the past two decades; however, smaller maximum coarse aggregate sizes may be required for concrete (thin) resurfacing. The use of well-graded aggregates reduces shrinkage. Admixtures Typical admixtures and additives that are commonly used in concrete mixtures include air entrainment (6% to 7%) water

63 reducers, and supplementary cementitious materials (SCMs) such as fly ash and ground granulated blast furnace slag (GGBFS) may also be added to concrete mixtures. Summary In this appendix, various renewal approaches that are appli- cable to using existing pavements in place and achieving long life were described. These include (1) AC overlay over existing AC pavements, (2) AC over crushed and shaped AC, (3) AC over reclaimed AC, (4) AC over CRCP, (5) AC over cracked and seated JPCP, and (6) AC over rubblized PCC. An overview of the criteria required for achieving long life was also pre- sented, and various overlay design approaches for AC surfaced pavements were outlined. References Ahlrich, R. C. 1989. Performance and Structural Evaluation of Cracked and Seated Concrete. Transportation Research Record 1215, TRB, National Research Council, Washington, D.C. American Association of State Highway and Transportation Officials (AASHTO). 1993. Guide for Design of Pavement Structures. Ameri- can Association of State Highway and Transportation Officials, Washington, D.C. American Concrete Pavement Association. 1990. Guidelines for Unbonded Concrete Overlays. Technical Bulletin TB-005P. American Concrete Pavement Association, Skokie, Ill. American Concrete Pavement Association. 1993. Reconstruction Opti- mization Through Concrete Inlays. Technical Bulletin TB013P. American Concrete Pavement Association, Skokie, Ill. Applied Research Associates. 2004. Guide for Mechanistic–Empirical Design of New and Rehabilitated Pavement Structures. NCHRP Project 1-37A. TRB, National Research Council, Washington, D.C. Asphalt Institute. 1981. Thickness Design Manual (MS-1), 9th ed. College Park, Md. Asphalt Institute. 1999. Asphalt Overlays for Highway and Street Reha- bilitation. Manual Series MS-17. Lexington, Ky. Asphalt Pavement Alliance. 2002. Perpetual Pavements: A Synthesis. APA 101. Lanham, Md. Asphalt Recycling and Reclaiming Association. 2001. Basic Asphalt Recy- cling Manual. Publication NHI01-022. FHWA, U.S. Department of Transportation. Barenberg, E. J. 1981. Rehabilitation of Concrete Pavements by Using Portland Cement Concrete Overlays. Transportation Research Record 814, TRB, National Research Council, Washington, D.C. Barnett, T. L., M. I. Darter, and N. R. Laybourne. 1981. Evaluation of Maintenance/Rehabilitation Alternatives for CRCP. Report FHWA/ IL/UI-185. Illinois Cooperative Highway Research Program, Uni- versity of Illinois, Urbana. Caestecker, C. 1993. Widening Works on the Motorway E40 at Leuven and Bertem (Belgium). Proc., Fifth International Conference on Con- crete Pavement Design and Rehabilitation, Purdue University, West Lafayette, Ind. Caestecker, C., and T. Lonneux. 2004. Inlay in Continuous Reinforced Concrete on the A10 Brussels–Oostende at Ternat. Presented at 9th International Symposium on Concrete Roads, Istanbul, April 4–7, 2004. Carpenter, S. H., and M. I. Darter. 1989. Field Performance of Crack and Seat Projects. Transportation Research Record 1215, TRB, National Research Council, Washington, D.C. Chou, Y. T. 1984. Asphalt Overlay Design for Airfield Pavements. Pro- ceedings of the Association of Asphalt Paving Technologists, Vol. 53, Association of Asphalt Paving Technologists, Minneapolis, Minn. Crawford, C. 1985. Cracking and Seating of PCC Pavements Prior to Overlaying with Hot Mix Asphalt—State of the Art. Information Series 91. National Asphalt Pavement Association, Riverdale, Md. Crawley, A. B. 1982. CRC Overlay of Existing CRCP. Report MSHD-RD- 82-074-I. Mississippi State Highway Department, Jackson. Darter, M. I., T. L. Barnett, and D. J. Morrill. 1982. Repair and Preventa- tive Maintenance Procedures for Continuously Reinforced Concrete Pavement. Report FHWA/IL/UI-191. Illinois Cooperative Highway Research Program, University of Illinois, Urbana. Dhamrait, J. S., and D. R. Schwartz. 1978. Continuously Reinforced Con- crete Overlays on Existing Portland Cement Concrete Pavement. Report FHWA-IL-PR-80. Illinois Department of Transportation, Springfield. Eckrose, R. A., and W. E. Poston, Jr. 1982. Asphalt Overlays on Cracked and Seated Concrete Pavements. Information Series 83. National Asphalt Pavement Association, Riverdale, Md. ERES Consultants, Incorporated. 1999. NCHRP Report 415: Evaluation of Unbonded Portland Cement Concrete Overlays. TRB, National Research Council, Washington, D.C. Fitts, G. L. 2006. Transportation Research Circular E-C087: Rubblization Using Resonant Frequency Equipment. Transportation Research Board of the National Academies, Washington, D.C. Hall, K. T. 1991. Performance, Evaluation, and Rehabilitation of Asphalt- Overlaid Concrete Pavements. PhD thesis, University of Illinois at Urbana-Champaign. Hall, K. T. 1999. Structural Coefficients for Fractured Concrete Slabs. American Concrete Pavement Association, Skokie, Ill. Hall, K. T., and M. I. Darter. 1989. Rehabilitation Performance and Cost-Effectiveness: 10-Year Case Study. Transportation Research Record 1215, TRB, National Research Council, Washington, D.C. Hall, K. T., M. I. Darter, and W. J. Seiler. 1993. Improved Design of Unbonded Concrete Overlays. Proc., 5th International Conference on Concrete Pavement Design and Rehabilitation, Purdue University, West Lafayette, Ind. Hall, K. T., C. E. Correa, S. H. Carpenter, and R. P. Elliott. 2001. Guide for Selection of Pavement Rehabilitation Strategies. NCHRP Project 1-38. TRB, National Research Council, Washington, D.C. Hall, K. T., D. Dawood, S. Vanikar, R. Tally, Jr., T. Cackler, A. Correa, P. Deem, J. Duit, G. M. Geary, A. J. Gisi, A. N. Hanna, S. Kosmatka, R. O. Rasmussen, S. Tayabji, and G. Voigt. 2007. Long-Life Concrete Pavements in Europe and Canada. Report FHWA-PL-07-027. FHWA, U.S. Department of Transportation. Harrington, D. 2008. Guide to Concrete Overlays: Sustainable Solutions for Resurfacing and Rehabilitating Existing Pavements, 2nd ed. National Concrete Pavement Technology Center, Iowa State Univer- sity, Ames. Hutchinson, R. L. 1982. NCHRP Synthesis of Highway Practice 99: Resurfacing with Portland Cement Concrete. TRB, National Research Council, Washington, D.C. Illinois DOT. 1982. “Subgrade Stability Manual,” Policy Mat-10, Illinois Department of Transportation. Kosmatka, S. H., B. Kerkhoff, W. C. Panarese. 2002. Design and Control of Concrete Mixtures. Portland Cement Association, Skokie, Ill. Liu, J., D. G. Zollinger, S. D. Tayabji, and K. D. Smith. 2003. Repair and Rehabilitation of Concrete Pavements: Volume IV: Strategic Analysis

64 of Pavement Evaluation and Repair (SAPER). Draft Final Report. FHWA, U.S. Department of Transportation. Mahoney, J., S. W. Lee, N. Jackson, and D. Newcomb. 1989. Mechanistic- Based Overlay Design Procedures for Washington State Flexible Pave- ments. Report WA-RD-170.1. Washington State Department of Transportation. Martin, R. 1973. Design Considerations for Resurfacing Pavement with Concrete. Highway Research Record 434, HRB, Washington, D.C. McGhee, K. H. 1994. NCHRP Synthesis of Highway Practice 204: Portland Cement Concrete Resurfacing. TRB, National Research Council, Washington, D.C. Mellinger, F. M. 1963. Structural Design of Concrete Overlays. Journal of the American Concrete Institute, Vol. 60, No. 2, pp. 225–238. Mellinger, F. M., and J. P. Sale. 1956. The Design of Non-Rigid Overlays for Concrete Airfield Pavements. Journal of the Air Transport Divi- sion, Vol. 82, No. AT 2, American Society of Civil Engineers. Monismith, C. L. 1992. Analytically Based Asphalt Pavement Design and Rehabilitation: Theory to Practice, 1962–1992. Transportation Research Record 1354, TRB, National Research Council, Washing- ton, D.C. National Asphalt Pavement Association. 1994. Guidelines for the Use of HMA Overlays to Rehabilitate PCC Pavements. Information Series 117, National Asphalt Pavement Association, Lanham, Md. National Concrete Pavement Technology Center. 2007. Guide to Con- crete Overlay Solutions. Iowa State University, Ames. National Highway Institute. 2003. Course 131083: Pavement Evalua- tions and Rehabilitation, Reference Manual. Publication FHWA- NHI 02-002. Nevada Department of Transportation. 1996. Pavement Structural Design and Policy Manual. Nevada Department of Transportation, Carson City. Newcomb, D. E., M. Buncher, and I. J. Huddleston. 2001. Transporta- tion Research Circular 503: Concepts of Perpetual Pavements. TRB, National Research Council, Washington, D.C. Older, C. 1924. Highway Research in Illinois. Transactions of the Ameri- can Society of Civil Engineers, Vol. 87. Portland Cement Association. 1981. Guide to Concrete Resurfacing Designs and Selection Criteria. Engineering Bulletin EB087.01P. Portland Cement Association, Skokie, Ill. Schutzbach, A. M. 1988. The Crack and Seat Method of Pavement Rehabili- tation. Physical Research Project 104. Bureau of Materials and Physical Research, Illinois Department of Transportation, Springfield. Schutzbach, A. M. 1989. Crack and Seat Method of Pavement Rehabili- tation. Transportation Research Record 1215, TRB, National Research Council, Washington, D.C. Scullion, T. 2006. Nondestructive Testing Results from the Rubblized Concrete Pavement on Interstate 10 in Louisiana. Transportation Research Circular E-C087: Rubblization of Portland Cement Concrete Pavements. Transportation Research Board of the National Acade- mies, Washington, D.C. Sebaaly, P. E., A. Hand, J. Epps, and C. Bosch. 1996. Nevada’s Approach to Pavement Management. Transportation Research Record 1524, TRB, National Research Council, Washington, D.C. Shell Pavement Design Manual. 1978. Shell International Petroleum Company, London. Shilstone, J. M. 1993. The Concrete Mixture: The Key to Pavement Durability. Proc., 5th International Conference on Concrete Pavement Design and Rehabilitation, Purdue University, West Lafayette, Ind. Shook, J. F., F. N. Finn, M. W. Witczak, and C. L. Monismith. 1982. Thickness Design of Asphalt Pavements: The Asphalt Institute Method. Proc., 5th International Conference on the Structural Design of Asphalt Pavements, University of Michigan and Delft University of Technology. Smith, K. D., H. T. Yu, and D. G. Peshkin. 2002. Portland Cement Con- crete Overlays: State of the Technology Synthesis. Report DTFH61- 00-P-00507. FHWA, U.S. Department of Transportation. Tayabji, S., and S. Lim. 2007. Long Life Concrete Pavements: Best Practices and Directions from the States. Concrete Pavement Technology Pro- gram, Technical Brief FHWA-HIF-07-030. FHWA, U.S. Department of Transportation. Thompson, M. R. 1989. NCHRP Synthesis of Highway Practice 144: Breaking/Cracking and Seating Concrete Pavements. TRB, National Research Council, Washington, D.C. Thompson, M. R. 1999. Hot-Mix Asphalt Overlay Design Concepts for Rubblized Portland Cement Concrete Pavements. Transportation Research Record 1684, TRB, National Research Council, Washing- ton, D.C. Thompson, M. R., and E. J. Barenberg. 1989. Calibrated Mechanistic Structural Analysis Procedures for Pavements. Phase I final report, NCHRP Project 1-26. TRB, National Research Council, Washing- ton, D.C. Thompson, M. R. and S. H. Carpenter. 2006. Considering Hot-Mix- Asphalt Fatigue Endurance Limit in Full-Depth Mechanistic- Empirical Pavement Design. Presented at 2006 International Conference on Perpetual Pavement, Columbus, Ohio, Septem- ber 2006. Timm, D. 2005. Perpetual Pavement Design. Presented to the Asphalt Pavement Alliance in Perpetual Pavement Open House, Ashton, Iowa, October 5, 2005. TRB Committee on General Issues in Asphalt Technology (A2D05). 2001. Transportation Research Circular 503: Perpetual Bituminous Pavement. TRB, National Research Council, Washington, D.C. Tyner, H. L., W. Gulden, and D. Brown. 1981. Resurfacing of Plain Jointed Concrete Pavement. Transportation Research Record 814, TRB, National Research Council, Washington, D.C. Voigt, G. F., M. I. Darter, and S. H. Carpenter. 1989. Field Performance Review of Unbonded Jointed Concrete Overlays. Transportation Research Record 1227, TRB, National Research Council, Washing- ton, D.C. Von Quintus, H. L., C. Rao, J. Mallela, B. Aho, Applied Research Associ- ates, Incorporated, and Wisconsin Department of Transportation. 2007. Guidance, Parameters, and Recommendations for Rubblized Pavements. Project WHRP 06-13, Project 16730. National Technical Information Service, Alexandria, Va. Washington State Department of Transportation. 2005. Pavement Design Manual. Washington State Department of Transportation, Olympia.