Evaluation of Bonded Concrete Overlays on Asphalt Pavements (2022)

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.

89   Several BCOA design methods have been developed over the past two decades, including BCOA-ME, the Colorado DOT method, the Illinois DOT method, and PaveME. One of the first design methods included the ACPA BCOA Thickness Designer program; however, this program is no longer supported and has been excluded from discussion and further analysis for this study. Structural Design Methods The following provides a brief discussion of each BCOA design method used in this study. BCOA-ME BCOA-ME, developed after Vandenbossche et al. (2011), is based on a comprehensive field survey and literature review of BCOA pavement projects in 11 U.S. states (Figure 70). Vandenbossche et al. (2011) used performance data from 52 projects to calibrate and validate the performance models (Vandenbossche et al. 2011; Li, Dufalla, and Vandenbossche 2013; Li et al. 2016; Vandenbossche and Sachs 2013b). BCOA-ME accounts for the dynamic asphalt modulus as a function of site-specific temperatures. Table 49 provides a summary of BCOA-ME inputs, which include project location and climate information, ESALs, failure criteria, existing structure information, concrete material properties, and joint spacing. BCOA-ME can evaluate slab sizes ranging from 2- × 2-ft to 15- × 12-ft with predicted BCOA thickness of 2 to 8 in. Colorado DOT The Colorado DOT design method was developed in 1998 (Tarr, Sheehan, and Okamoto 1998; Sheehan, Tarr, and Tayabji 2004). (See Figure 71 for an example design.) The design method recommends a minimum asphalt layer thickness of 3 in. and includes both concrete and asphalt fatigue cracking failure criteria (Colorado DOT 2020). The design method uses critical loading along the longitudinal edge to predict transverse cracking on the basis of field observa- tions and measured strains for partially bonded slabs. Failure criteria are based on a concrete stress ratio [total concrete stress divided by design concrete modulus of rupture (MOR)] near 0.50. Maintaining a stress ratio of about 0.50 implies an unlimited number of load repetitions (Tarr, Sheehan, and Okamoto 1998). This method does not include models for longitudinal cracking or corner breaks, nor does it consider the use of synthetic macrofibers. The design method eval uates 6- × 6-ft to 12- × 12-ft slab sizes with a BCOA layer thickness range of 5 to 8 in. Table 50 lists required inputs for the Colorado DOT design method. C H A P T E R 5 Performance Prediction

Figure 70. Screenshot of BCOA-ME. (Source: Adapted from Vandenbossche 2018.)

Input Criteria General information Latitude, longitude, and elevation Design-lane ESALs Maximum allowable percentage of slabs cracked and reliability (25% cracked slabs and 85% reliability recommended) Climate Annual mean daily average temperature region ID Map of sunshine zone Existing structure Post-milling asphalt pavement layer thickness Asphalt layer fatigue (adequate or marginal) k-value Presence (“yes” or “no”) of transverse cracking Concrete layer properties Flexural strength (average 28 day) Estimated elastic modulus CTE Fiber type and content Joint spacing 2- x 2-ft, 3- x 3-ft, 4- x 4-ft, 6- x 6-ft, 7- x 7-ft, 10- x 12-ft, 12- x 12-ft, and 15- x 12-ft BCOA thickness 2 to 8 in. Recommended: 3.0 to 5.5 in. for 4- x 4-ft and 6- x 6-ft slabs 5.5 to 6.5 in. for 12- x 12-ft slabs Table 49. BCOA-ME design method inputs. Whitetopping Parameters (input trial design variables) Highway Category (Primary or Secondary)* Primary Joint Spacing, in. 150 Trial Concrete Thickness, in. 7.5 Concrete Elastic Modulus, psi 3,400,000 Concrete Poisson's Ratio 0.15 Concrete Flexural Strength, psi 650 Asphalt Thickness, in. 9 Asphalt Elastic Modulus, psi 300,000 Asphalt Poisson's Ratio 0.35 Asphalt Fatigue Life Previously Consumed, % 50 Subgrade Modulus, pci 150 Temperature Gradient, °F/in. 3 Design ESALs 3,100,000 Converted Concrete Thickness, in. ESAL Conversion Factor Neutral Axis le = L/le = 7.78 0.9938 4.66 39.13 3.83 Critical Concrete Stresses and Asphalt Strains Load Induced Support AdjustmentBond Adjustment Stress, psi µstrain Stress, psi µstrain Stress, psi µstrain 1 2 3 4 5 6 162 187 267 159 303 159 ESAL Fatigue Analysis No. of Concrete Fatigue Analysis Asphalt Fatigue Analysis 18-kip Stress Allowable Fatigue, Asphalt Allowable Fatigue, ESALs Ratio ESALs % µstrain ESALs % 7 8 9 10 11 12 13 3.1E+06 0.466 7.2E+06 43.0 159 2.6E+06 118.1 Concrete Fatigue, % = 43.0 Asphalt Fatigue, % = 168.1 Required Whitetopping Thickness = 7.5 in. Figure 71. Screenshot of Colorado DOT whitetopping design.

92 Evaluation of Bonded Concrete Overlays on Asphalt Pavements Illinois DOT The Illinois DOT design method was developed in 2008 in cooperation with the University of Illinois Urbana–Champaign (Roesler et al. 2008). The Illinois DOT design method is an extension of the Portland Cement Association design method, with an updated concrete fatigue algorithm, consideration of synthetic macrofibers in terms of residual strength, elimination of asphalt layer fatigue as a failure mode (as it did not appear to affect performance), and the effective temperature gradient for Illinois (Roesler et al. 2008; Illinois DOT 2020). The contribution of synthetic macrofibers was determined by calculating an effective flexural strength of the concrete, which is the actual concrete strength plus the residual strength of the FRC material based on ASTM C1609 results. Evaluated slab joint spacings range from 4 to 6 ft, with recommended BCOA thicknesses ranging from 3 to 6 in. and a minimum asphalt thickness of 2.5 in. (Figure 72). The BCOA design spreadsheet tool requires the use of FRC for thicknesses ≤4 in. (Roesler et al. 2008). Corner loading is considered a critical load location, considering that much of the observed BCOA distress in Illinois was corner breaks. Existing structure Asphalt thickness Asphalt elastic modulus Asphalt Poisson’s ratio Asphalt fatigue life previously consumed (%) Subgrade modulus k-value Concrete layer properties Elastic modulus Poisson’s ratio Flexural strength Joint spacing 72 to 144 in. BCOA thickness 5 to 8 in. Input Criteria General information Highway category (primary or secondary) Trial concrete thickness (5 to 8 in.) Design ESALs Climate Temperature gradient Table 50. Colorado DOT design method inputs. Figure 72. Screenshot of Illinois DOT inlay/overlay design spreadsheet. (Source: Illinois DOT 2018.)

Performance Prediction 93   The design method failure criteria, not modifiable by the user, include 20% slab cracking at 85% reliability. This method considers BCOA designs applicable for traffic levels below 5 million ESALs. To simplify the design process, three asphalt stiffness options are included and linked to the distress condition (poor, fair, and good) of the asphalt layer. Design method inputs are listed in Table 51. PaveME In 2016, PaveME was updated to accommodate the analysis of short-jointed plain concrete pavement (SJPCP), specifically for joint spacing of 5 to 8 ft and thinner slabs of 4 to 8 in. (AASHTO 2016). PaveME uses the BCOA-ME longitudinal cracking model and the existing bonded concrete over asphalt transverse cracking model. The SJPCP models excluded the ACPA corner break model; therefore, PaveME is restricted to slab sizes 5 ft and greater. In addition, PaveME uses axle load spectra for characterizing traffic, rather than the ESALs used in BCOA-ME. Currently, PaveME predicts only the percentage of slabs with longitudinal cracking (the inter- nally set performance criterion is 15% of slabs at 50% reliability). PaveME inputs are listed in Table 52. A comparison of overlay thickness, slab size, and failure criteria for BCOA design procedures is summarized in Table 53. Input Criteria General information Design traffic factor (ESALs/1 x 106) Climate Internal value (Illinois conditions only) Existing structure Remaining thickness of asphalt Elastic modulus of asphalt k-value Concrete layer properties Modulus of rupture FRC residual strength ratio CTE Elastic modulus Joint spacing 48 to 72 in. BCOA thickness 3 to 6 in. Table 51. Illinois DOT design method inputs. Input Criteria General information Design life Construction and opening to traffic month and year Axle load spectra Climate Site specific Existing structure Layer thickness (asphalt limited to 3 to 10 in., base, subbase) Asphalt layer air voids and effective binder content Asphalt layer Poisson’s ratio Asphalt binder grade and type Subgrade layer type Subgrade layer Poisson’s ratio Subgrade layer resilient modulus Subgrade layer gradation, liquid limit, plastic limit Concrete layer properties Thickness Poisson’s ratio CTE Aggregate type Cement type, cementitious material content, water-cement ratio Fibers Strength Elastic modulus Joint spacing 5 to 8 ft BCOA thickness 4 to 8 in. Table 52. Selected PaveME SJPCP design method inputs.

94 Evaluation of Bonded Concrete Overlays on Asphalt Pavements Current BCOA design methods present several major limitations: • None of the design methods explicitly models the concrete–asphalt bond. Two introduce a partial bond and increase the concrete slab stress by a constant factor (Colorado and Illinois DOT). Two model the bond indirectly by using calibration factors (BCOA-ME and PaveME). It is believed that deterioration of the concrete–asphalt bond is one major factor leading to concrete fatigue damage and a key factor in BCOA performance. • None of the performance prediction models considers concrete drying shrinkage and mois- ture warping. • All of the design methods use a fixed LTE in the structural model, whereas a reduction in LTE can occur in the field over time. For example, BCOA-ME uses a constant LTE value of 90% and PaveME calibrated its longitudinal cracking model assuming 80% LTE. • None of the design methods considers faulting (Han 2005; Mateos et al. 2015). • Curling stress calculation is oversimplified through the use of either a single design gradient or an adjustment to slab responses caused by thermal gradients. • None of the design methods uses a rational approach for the contribution of structural fibers, except through adjustment of flexural strength. Sensitivity Analysis A sensitivity analysis was conducted to determine input variability on design method results. The sensitivity analysis evaluates each variable independently, rather than how the interaction of variables may influence design recommendations. The sensitivity analysis included the evalu- ation of asphalt layer thickness, truck traffic, subgrade support, and concrete flexural strength on recommended BCOA layer thickness. BCOA-ME Input values used for the BCOA-ME sensitivity analysis are shown in Table 54. In general, asphalt layer thickness ranging from 3 to 7 in. results in a reduction in BCOA layer thickness of approximately 0.5 to 1.5 in. (Figure 73). The inclusion of structural fibers also reduces the recommended BCOA layer thickness. However, the BCOA layer thickness for a 12- × 12-ft slab (no fibers) is not influenced by asphalt layer thickness. The 12- × 12-ft slab without synthetic macrofibers results in an unexpected increase in BCOA layer thickness from an asphalt layer thickness of 7 to 8 in. For 4- × 4-ft slabs without synthetic macrofibers, the minimum BCOA layer thickness (3 in.) occurs at 1,000 ESALs, compared with 100,000 ESALs when synthetic macrofibers are included Design Procedure Overlay Thickness (in.) Slab Size Failure Criteria BCOA-ME 2 to 8 2- x 2-ft to 15- x 12-ft % cracked slabs Reliability User-specified input Colorado DOT 3 to 8 6- x 6-ft to 12 -x 12-ft Concrete stress ratio near 0.50 Not changeable by user Illinois DOT 3 to 6 4- x 4-ft or 6- x 6-ft 20% cracked slabs Not changeable by user PaveME 4 to 8 5- x 5-ft to 8- x 8-ft 15% cracked slabs at 50% reliability Not changeable by user Table 53. Comparison of BCOA design methods.

Performance Prediction 95   (Figure 74). The minimum BCOA layer thickness for 6- × 6-ft slabs (3.0 in.) with or without synthetic macrofibers occurs at 1,000 ESALs. In addition, regardless of ESAL level, a 3-in. BCOA layer is recommended when synthetic macrofibers are used. The recommended BCOA layer thickness for 12- × 12-ft slabs with synthetic macrofibers is not influenced by ESAL level. Compared with asphalt layer thickness and ESALs, BCOA thickness is not sensitive to k-value, which is typically true with new concrete pavement designs (Figure 75). For all slab sizes and conditions evaluated, k-value changes from 100 to 500 pci resulted in a BCOA thickness change of approximately 0.6 in. The 12- × 12-ft slabs with no fibers had a 1.4-in. reduction in BCOA layer thickness with a k-value change from 50 to 100 pci. The addition of synthetic macrofibers is addressed through an artificial increase in the con- crete effective flexural strength, and BCOA thickness is sensitive to changes in flexural strength (Figure 76). The 6- × 6-ft slabs without fibers were the most sensitive, with a 1.1-in. thickness change when the flexural strength decreased from 700 to 600 psi. 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 BC O A Th ic kn es s ( in ch ) Asphalt Layer Thickness (inch) 4x4 - No Fibers 6x6 - No Fibers 12x12 - No Fibers 4x4 - With Fibers 6x6 - With Fibers 12x12 - With Fibers Figure 73. BCOA-ME sensitivity analysis of asphalt layer thickness. Design Input Value Annual mean daily average temperature region ID 5 Map of sunshine zone 2 Elevation (ft) 874 ESALs 1,000,000 (1,000 to 10,000,000)a Maximum allowable % slabs cracked 25 Desired reliability (%) 85 Post-milling hot-mix asphalt thickness (in.) 6 (3 to 10)a Existing hot-mix asphalt fatigue cracking Adequate (8% to 20%) k-value (pci) 150 (50 to 500)a Existing hot-mix asphalt transverse cracks (yes or no) Yes Average 28-day flexural strength (psi) 650 (450 to 850)a Concrete elastic modulus (psi) 4,000,000 CTE (x10-6 in./in./°F) 5.5 Fiber type None or synthetic macrofibers (80 lb/yd3) a Values in parentheses represent the range of values used in the sensitivity analysis. Table 54. BCOA-ME design inputs and values for sensitivity analysis.

96 Evaluation of Bonded Concrete Overlays on Asphalt Pavements 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 0.001 0.01 0.1 1 10 100 BC O A Th ic kn es s ( in ch ) ESALs (millions) 4x4 - No Fibers 6x6 - No Fibers 12x12 - No Fibers 4x4 - With Fibers 6x6 - With Fibers 12x12 - With Fibers Figure 74. BCOA-ME sensitivity analysis of ESALs. 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 0 100 200 300 400 500 BC O A Th ic kn es s ( in ch ) k-value (pci) 4x4 - No Fibers 6x6 - No Fibers 12x12 - No Fibers 4x4 - With Fibers 6x6 - With Fibers 12x12 - With Fibers Figure 75. BCOA-ME sensitivity analysis of k-value. In summary, the BCOA-ME design method is sensitive to thinner existing asphalt layers (3 to 7 in.), ESALs, and concrete flexural strength and insensitive to changes in k-value from 100 to 500 pci. Colorado DOT The sensitivity analysis inputs for the Colorado DOT design method are shown in Table 55. In this analysis, asphalt layer fatigue was held constant and only concrete fatigue resistance was considered. The result of the asphalt layer thickness sensitivity analysis is presented in Figure 77.

Performance Prediction 97   3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 400 500 600 700 800 900 BC O A Th ic kn es s ( in ch ) Flexural Strength (psi) 4x4 - No Fibers 6x6 - No Fibers 12x12 - No Fibers 4x4 - With Fibers 6x6 - With Fibers 12x12 - With Fibers Figure 76. BCOA-ME sensitivity analysis of concrete flexural strength. 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 BC O A Th ic kn es s ( in ch ) Asphalt Layer Thickness (inch) 4x4, 6x6 and 12x12 Figure 77. Colorado DOT sensitivity analysis of asphalt layer thickness. Design Input Value Highway category (primary or secondary) Primary Concrete modulus (psi) 4,000,000 Poisson’s ratio 0.15 Concrete flexural strength (psi) 650 (450 to 850)a Asphalt thickness (in.) 6 (3 to 10)a Asphalt elastic modulus (psi) 350,000 Asphalt Poisson’s ratio 0.35 Asphalt fatigue life consumed (%) 50 k-value (pci) 150 (50 to 500)a Temperature gradient (°F/in.) 3 Design ESALs 1,000,000 (10,000 to 10,000,000)a a Values in parentheses represent the range of values in the sensitivity analysis. Table 55. Colorado DOT design inputs and values for sensitivity analysis of asphalt layer thickness.

98 Evaluation of Bonded Concrete Overlays on Asphalt Pavements The BCOA thickness is not influenced by asphalt layer thicknesses less than 7 in. An increase in asphalt layer thickness from 7 to 10 in. results in a 3-in. decrease in the BCOA thickness. The design method does not differentiate recommended BCOA layer thickness on the basis of slab size. For the inputs evaluated, a minimum BCOA layer thickness (6 in.) is obtained at 10,000 ESALs and a maximum thickness of approximately 8 in. is obtained at 10,000,000 ESALs (Figure 78). The Colorado DOT design method is sensitive to ESALs increasing from 10,000 to 100,000, as noted by a nearly 2-in. increase in BCOA layer thickness. As shown in Figure 79, the BCOA layer thickness is sensitive to changes in k-values. Changes in k-value from 150 to 400 pci result in a 3-in. reduction in the recommended BCOA layer thick- ness. For k-values less than 150 pci and above 400 pci, the maximum and minimum BCOA layer thicknesses were reached, respectively. 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 0 100 200 300 400 500 BC O A Th ic kn es s ( in ch ) k-value (pci) 4x4, 6x6 and 12x12 Figure 79. Colorado DOT sensitivity analysis of k-value. 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 0.01 0.1 1 10 BC O A Th ic kn es s ( in ch ) ESAL (million) 4x4, 6x6 and 12x12 Figure 78. Colorado DOT sensitivity analysis of ESALs.

Performance Prediction 99   Flexural strength values ranging from 450 to 700 psi resulted in the maximum BCOA layer thickness (Figure 80). As flexural strength increased from 700 to 850 psi, the required BCOA thickness decreased from 8 to 6 in. In summary, the Colorado DOT design method is sensitive to asphalt layer thicknesses greater than 7 in., ESALs, k-value ranging from 150 to 400 pci, and concrete flexural strength greater than 700 psi. The Colorado DOT procedure does not consider the inclusion of synthetic macrofibers, nor does it differentiate recommended thicknesses on the basis of slab size. Illinois DOT The input values to evaluate the sensitivity of the Illinois DOT design method are shown in Table 56. In all cases evaluated, BCOA layer thickness decreases with increases in asphalt layer thick- ness (Figure 81). For 4- × 4-ft and 6- × 6-ft slabs without synthetic macrofibers, BCOA layer thickness decreases as asphalt layer thickness increases from 5 to 7 in., and 7 to 8 in., respectively. Similarly, for 4- × 4-ft and 6- × 6-ft slabs with synthetic macrofibers, BCOA layer thickness decreases as asphalt layer thickness increases from 3 to 5 in., and 4 to 7 in., respectively. Recommended overlay thickness for the 4- × 4-ft slabs was less than recommended overlay thickness for the 6- × 6-ft slabs (Figure 82). With respect to synthetic macrofibers, for 6- × 6-ft 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 400 500 600 700 800 900 BC O A Th ic kn es s ( in ch ) Concrete Flexural Strength (psi) 4x4, 6x6 and 12x12 Figure 80. Colorado DOT sensitivity analysis of concrete flexural strength. Design Input Value Traffic factor (ESALs/1 x 106) 1 (0.1 to 10)a MOR (psi) 650 (450 to 850)a Asphalt layer thickness (in.) 6 (3 to 8)a Concrete elastic modulus (psi) 4,000,000 CTE (x 10-6 in./in./°F) 5.5 Asphalt layer modulus (psi) 350,000 k-value (pci) 150 (50 to 500)a Macrofiber residual strength ratio None (0%) or synthetic macrofibers (20%) a Values in parentheses represent the range of values in the sensitivity analysis. Table 56. Illinois DOT design inputs and values for sensitivity analysis of asphalt layer thickness.

100 Evaluation of Bonded Concrete Overlays on Asphalt Pavements slabs, the maximum overlay thickness is reached at 10,000 ESALs and the minimum BCOA layer thickness is unchanged regardless of ESALs for 4- × 4-ft slabs (for range evaluated). Including synthetic macrofibers results in a reduction of required BCOA layer thickness (approximately 1 to 3 in. depending on slab size and traffic factor). The recommended overlay thickness is not sensitive to changes in k-value (Figure 83). A k-value change from 50 to 500 pci resulted in a maximum reduction of only 0.6 in. As expected, the recommended BCOA thickness is sensitive to changes in concrete flexural strength (Figure 84). For the 4- × 4-ft slabs, a concrete flexural strength increase of 100 psi resulted in a decrease in the recommended BCOA thickness of 2.5 in. Similar patterns were noted for 6- × 6-ft slabs. The addition of synthetic macrofibers reduced the required thickness, as well as increased the concrete flexural strength. 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 0.01 0.1 1 10 BC O A Th ic kn es s ( in ch ) Traffic Factor (ESALs/1 × 106) 4x4 - No Fibers 4x4 - With Fibers 6x6 - No Fibers 6x6 - With Fibers Figure 82. Illinois DOT sensitivity analysis of ESALs. 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 3.0 4.0 5.0 6.0 7.0 8.0 BC O A Th ic kn es s ( in ch ) Asphalt Layer Thickness (inch) 4x4 - No Fibers 4x4 - With Fibers 6x6 - No Fibers 6x6 - With Fibers Figure 81. Illinois DOT sensitivity analysis of asphalt layer thickness.

Performance Prediction 101   In summary, the Illinois DOT design method is sensitive to changes in asphalt layer thick- ness, traffic factor, and MOR. The Illinois DOT design method is fairly insensitive to changes in k-value. PaveME Table 57 provides the variables and range of values used in the sensitivity analysis to evaluate the impact of design inputs on predicted longitudinal cracking (default values were used for all other inputs). Asphalt layer thickness significantly influences longitudinal cracking on 4-in. SJPCP thickness— a decrease of slabs with longitudinal cracking from 100% to 3% when asphalt thickness increases 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 400 500 600 700 800 900 BC O A Th ic kn es s ( in ch ) Concrete Flexural Strength (psi) 4x4 - No Fibers 4x4 - With Fibers 6x6 - No Fibers 6x6 - With Fibers Figure 84. Illinois DOT sensitivity analysis results of MOR. 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 0 100 200 300 400 500 BC O A Th ic kn es s ( in ch ) k-value (pci) 4x4 - No Fibers 4x4 - With Fibers 6x6 - No Fibers 6x6 - With Fibers Figure 83. Illinois DOT sensitivity analysis of k-value.

102 Evaluation of Bonded Concrete Overlays on Asphalt Pavements from 4 to 8 in. (Figure 85). Asphalt layer thickness has less of an impact on longitudinal crack- ing for the 6-in. SJPCP thickness (a decrease in cracked slabs from approximately 40% to 2% when asphalt layer thickness increases from 4 to 6 in.). Asphalt layer has no effect on the 8-in. SJPCP thickness. Figure 86 illustrates the predicted impact of average annual daily truck traffic (AADTT) on longitudinal cracking. The 4-in. BCOA thickness evaluation resulted in a significant increase in longitudinal cracking with increasing AADTT (nearly 100% slabs cracked at 5,000, or approxi- mately 22 million ESALs over the 20-year design period). The thicker SJPCP overlays (both 6 and 8 in.) showed no to minimal increase in predicted longitudinal cracking with increasing AADTT. Subgrade strength had minimal impact on reducing longitudinal cracking in the 4-in. SJPCP (approximately 100% of slabs cracked over the subgrade modulus range evaluated) and no impact on 8-in. SJPCP (no slabs cracked). It, however, had significant impact on 6-in. SJPCP (Figure 87). Increasing concrete flexural strength affected all SJPCP thicknesses and slab sizes evaluated (Figure 88). A significant reduction was noted for both the 4- and 6-in. SJPCP (from 100% to less than 20% cracked slabs as flexural strength increased from 450 to 850 psi). For the 8-in. SJPCP, predicted longitudinal cracking reduced to zero for concrete flexural strength greater than 600 psi. 0 20 40 60 80 100 4 5 6 7 8 Lo ng itu di na l C ra ck in g (% sl ab s) Asphalt Layer Thickness (inch) 4-inch 5x5 4-inch 8x8 6-inch 5x5 6-inch 8x8 8-inch 5x5 8-inch 8x8 Figure 85. PaveME sensitivity analysis of asphalt layer thickness. Design Input BCOA (in.) Joint Spacing (ft) Value Average annual daily truck traffic 4, 6, and 8 5 and 8 50 to 5,000 Asphalt layer thickness (in.) 4, 6, and 8 5 and 8 4, 6, and 8 Subgrade modulus (psi) 4, 6, and 8 5 and 8 5,000, 10,000, and 15,000 Flexural strength 4, 6, and 8 5 and 8 450, 600, and 850 Table 57. Analysis of PaveME sensitivity to longitudinal cracking.

Performance Prediction 103   0 20 40 60 80 100 0 1,000 2,000 3,000 4,000 5,000 Lo ng itu di na l C ra ck in g (% sl ab s) Average Annual Daily Truck Traffic 4-inch 5x5 4-inch 8x8 6-inch 5x5 6-inch 8x8 8-inch 5x5 8-inch 8x8 Figure 86. PaveME sensitivity analysis of AADTT. 0 20 40 60 80 100 0 5,000 10,000 15,000 20,000 Lo ng itu di na l C ra ck in g (% sl ab s) Subgrade Modulus (psi) 4-inch 5x5 4-inch 8x8 6-inch 5x5 6-inch 8x8 8-inch 5x5 8-inch 8x8 Figure 87. PaveME sensitivity analysis of subgrade modulus. The BCOA-ME, Colorado DOT, and Illinois DOT design methods produce different concrete thicknesses for a given set of inputs. This is expected considering that the design methods have different structural models, failure criteria, internal assumptions, and so on. The results of the sensitivity analysis are provided in Table 58 and further illustrated in Figure 89 and Figure 90. As shown in Figure 89, over the input ranges evaluated for the 4- × 4-ft slabs, all three design methods result in BCOA thickness reductions with increasing asphalt layer thickness. All three design methods result in a similar BCOA layer thickness increase with increasing ESALs;

104 Evaluation of Bonded Concrete Overlays on Asphalt Pavements Input % Change in BCOA Thickness over Evaluated Input Rangea BCOA-ME Colorado DOT Illinois DOT 4- x 4-ft 6- x 6-ft 12- x 12-ft All 4- x 4-ft 6- x 6-ft Nob Yes No Yes No Yes No No Yes No Yes Asphalt thickness −0.6 −1.7 −1.3 −0.3 −0.3 0.0 −1.1 −0.9 −1.8 −3.0 −3.1 ESALs 2.0 0.8 1.3 0.0 0.9 0.0 2.1 2.4 0.0 2.4 1.8 k-value −0.1 0.0 −1.0 0.0 −1.6 0.0 −3.1 −0.7 0.0 −0.4 0.0 Flexural strength −1.3 −1.6 −3.6 −1.1 −1.1 −0.5 −2.0 −3.1 −2.4 −3.1 −3.1 a Negative values represent a decrease in BCOA thickness over the evaluated input range. b No = excludes synthetic macrofibers; Yes = includes synthetic macrofibers. Table 58. Comparison of sensitivity analysis results of BCOA design methods. -4.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 Ch an ge in BC O A Th ic kn es s( in ch ) BCOA-ME w/o fibers BCOA-ME w/fibers CODOT w/o fibers ILDOT w/o fibers ILDOT w/fibers Asphalt Layer Thickness ESALs k-value Concrete Flexural Strength Figure 89. Change in BCOA layer thickness of 4- ë 4-ft slabs. 0 20 40 60 80 100 400 500 600 700 800 900 Lo ng itu di na lC ra ck in g (% sl ab s) Flexural Strength (psi) 4-inch 5x5 4-inch 8x8 6-inch 5x5 6-inch 8x8 8-inch 5x5 8-inch 8x8 Figure 88. PaveME sensitivity analysis of concrete exural strength.

Performance Prediction 105   however, BCOA-ME results in less than half the BCOA layer thickness when synthetic macro- bers are used. Increasing k-value for the Colorado DOT design method causes a signicant reduction in BCOA thickness (approximately 3 in.) compared with the other two methods. All design methods show a decrease in BCOA layer thickness with increasing exural strength; the BCOA-ME result in a thinner recommended BCOA layer with synthetic macrobers. For 6- × 6- slabs, the Illinois DOT design method signicantly reduces BCOA layer thick- ness with increasing asphalt layer thickness, compared with the other two methods (Figure 90). All methods, excluding BCOA-ME with synthetic macrobers, result in thicker BCOA layers with increasing ESALs. As with the 4- × 4- slabs, increasing k-value signicantly reduces BCOA layer thickness for the Colorado DOT design method, compared with the other two design methods. Increasing exural strength reduces BCOA layer thickness for all design methods; however, BCOA-ME without synthetic macrobers results in signicantly thinner BCOA layers than the other two methods, and the Illinois DOT design method results in approximately the same reduction in BCOA layer thickness, with or without synthetic macrobers. In general, asphalt layer thickness, ESALs, concrete strength, and slab size signicantly aect the recommended BCOA thickness. For design methods allowing inclusion of synthetic macro- bers, slab thickness is signicantly reduced with the addition of synthetic macrobers at a certain minimum residual strength for the same set of inputs. PaveME sensitivity analysis results are summarized in Table 59. inner SJPCP shows more signicant change in longitudinal cracking over the input ranges evaluated. For 8- × 8- slabs -4.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 Ch an ge in BC O A Th ic kn es s( in ch ) BCOA-ME w/o fibers BCOA-ME w/fibers CODOT w/o fibers ILDOT w/o fibers ILDOT w/fibers Asphalt Layer Thickness ESALs k-value Concrete Flexural Strength Figure 90. Change in BCOA layer thickness of 6- ë 6-ft slabs. Input % Change in Predicted Longitudinal Cracking over Evaluated Input Rangea 5- x 5-ft 8- x 8-ft 4 in. 6 in. 8 in. 4 in. 6 in. 8 in. Asphalt thickness −94 −97 −43 −38 0 0 AADTT 98 95 4 4 0 0 Subgrade modulus −2 −5 −63 −84 0 0 Flexural strength −83 −94 −100 −100 −41 −55 a Negative values represent a decrease in predicted longitudinal cracking (% slabs). Table 59. Sensitivity analysis results using PaveME.

106 Evaluation of Bonded Concrete Overlays on Asphalt Pavements and thicker overlays (6 and 8 in.), no change in longitudinal cracking is seen, excluding an increase in concrete flexural strength, over the range of values evaluated. Design Results BCOA thicknesses were determined using BCOA-ME, Colorado DOT, Illinois DOT, and PaveME with data obtained, to the extent possible, from the project site evaluations. The most sensitive variable in BCOA-ME is flexural strength, and the use of project-specific strength values would greatly change the predicted concrete slab thickness. However, testing included only split tensile and compression strength, and a correlation to 28-day flexural strength from either compression or split tensile strength testing and the age of the field specimens would be needed. Therefore, as a workable compromise, the analysis included a fixed 28-day flexural strength of 650 psi, given that no other information is available on the original design strengths. For the Illinois DOT design method, because of the low sensitivity of k-value on BCOA layer thickness, all projects used a k-value of 150 pci. The Illinois DOT design method recommends using an average 14-day flexural strength of 750 psi because it correlates better to construction specifications. However, to remain consistent with the other design methods, and on the basis of the results of the sensitivity analysis, all projects included a flexural strength of 650 psi. Inputs for all design methods are provided in Appendix G. Figures 91 through 109 summarize the design thickness recommendations and the percentage of slabs PaveME predicts will develop longitudinal cracking. BCOA-ME and Illinois DOT BCOA layer thickness recommendations are based on a maximum allowable percentage of cracked slabs—25% and 20%, respectively. Colorado DOT results are based on a concrete stress ratio near 0.50, and PaveME is based on performance criteria of 15% slabs with longitudinal cracking at 50% reliability. In addition, the percentage of slabs with longitudinal cracking, the total per- centage of slabs cracked (corner breaks, longitudinal cracks, and transverse cracks), faultmeter results from the visual distress survey (unless noted otherwise), and IRI from the automated distress survey are plotted using a linear extrapolation from year of BCOA construction to in-service age. Initial IRI is based on agency results following BCOA construction. A project assessment is provided beneath each figure. A comparison of recommended and as-constructed BCOA layer thicknesses from the BCOA-ME, Colorado DOT, and Illinois DOT design methods, as well as PaveME-predicted (at in-service pavement age) and field-observed longitudinal cracking, are shown in Table 60. Segments were considered overdesigned when the design method recommendation is 1 in. or greater than the as-constructed thickness. Underdesigned segments were considered when the design method recommendation is 1 in. or less than the as-constructed thickness. For PaveME, underperforming segments were considered when the predicted longitudinal cracking was 5% or greater than the field-observed longitudinal cracking. Conversely, overperforming seg- ments were defined when predicted longitudinal cracking was 5% or less than the field-observed longitudinal cracking. A summary of findings includes the following results: • According to BCOA-ME, one segment (3%) was underdesigned, 11 segments (29%) were adequately designed, and 26 segments (68%) were overdesigned. • According to Colorado DOT, 20 segments (53%) were underdesigned, 10 segments (26%) were adequately designed, and eight segments (21%) were overdesigned. • According to Illinois DOT, six segments (16%) were underdesigned, 21 segments (58%) were adequately designed, and 11 segments (26%) were overdesigned. • According to PaveME, three segments (8%) performed worse than predicted, three segments (8%) performed significantly better than predicted, and 32 segments (84%) had less than 5% difference between predicted and field-observed longitudinal cracking.

Performance Prediction 107   (a) (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 6 in. BCOA (in.) 6.8 6.0 BCOA-ME (in.) 3.8 3.8 In-service age (years) 7 Asphalt (in.) 15.1 12.1 Colorado DOT (in.) 3.0 3.3 Est. ESALs to date 17,500,000 Slab size 6- x 6-ft Il l inois DOT (in.) >6.0 >6.0 Joint seal Yes PaveME (% slabs) 0 0 Fibers No Features on CO I-70 include a thick asphalt layer, in-place BCOA thicker than design method recommendations, and ESALs exceeding the Illinois DOT recommendation. PaveME predicts no longitudinal cracking over the design period. NOTE: Because of challenges with faultmeter testing, faulting is based on an automated distress survey. Figure 91. Design method results for (a) good segment and (b) poor segment of CO I-70.

108 Evaluation of Bonded Concrete Overlays on Asphalt Pavements (a) (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 5 in. BCOA (in.) 5.5 10.4 BCOA-ME (in.) 3.9 3.9 In-service age (years) 19 Asphalt (in.) 9.2 7.5 Colorado DOT (in.) 6.3 >8.0 Est. ESALs to date 35,400,000 Slab size 6- x 6-ft Il l inois DOT (in.) 5.0 5.2 Joint seal Yes PaveME (% slabs) 0 0 Fibers No Features on CO SH-83A include a thicker asphalt layer, in-place BCOA thicker than BCOA-ME and Illinois DOT recommendations, and ESALs exceeding the Illinois DOT recommendation. PaveME predicts no longitudinal cracking over the design period. Figure 92. Design method results for (a) good segment and (b) poor segment of CO SH-83A.

Performance Prediction 109   (a) (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 6 in. BCOA (in.) 5.9 6.0 BCOA-ME (in.) 3.7 3.7 In-service age (years) 14 Asphalt (in.) 12.7 12.9 Colorado DOT (in.) >8.0 >8.0 Est. ESALs to date 23,000,000 Slab size 6- x 6-ft Il l inois DOT (in.) >6.0 >6.0 Joint seal Yes PaveME (% slabs) 0 0 Fibers No Features on CO SH-83B include a thick asphalt layer, in-place BCOA thicker than the BCOA-ME recommendation, and ESALs exceeding the Illinois DOT recommendation. PaveME predicts no longitudinal cracking over the design period. Figure 93. Design method results for (a) good segment and (b) poor segment of CO SH-83B.

110 Evaluation of Bonded Concrete Overlays on Asphalt Pavements (a) (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 6 in. BCOA (in.) 6.0 5.7 BCOA-ME (in.) 3.0 3.0 In-service age (years) 19 Asphalt (in.) 13.1 13.5 Colorado DOT (in.) > 8.0 7.5 Est. ESALs to date 13,000,000 Slab size 6- x 6-ft Il l inois DOT (in.) >6.0 3.0 Joint seal Yes PaveME (% slabs) 0 0 Fibers No Features on CO SH-121A include a thick asphalt layer, in-place BCOA thicker than BCOA-ME and Illinois DOT and thinner than Colorado DOT recommendations, and ESALs exceeding the Illinois DOT recommendation. PaveME predicts no longitudinal cracking over the design period. Figure 94. Design method results for (a) good segment and (b) poor segment of CO SH-121A.

Performance Prediction 111   (a) (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 6 in. BCOA (in.) 7.5 6.7 BCOA-ME (in.) 4.7 3.7 In-service age (years) 8 Asphalt (in.) 3.7 9.0 Colorado DOT (in.) >8.0 6.3 Est. ESALs to date 6,800,000 Slab size 6- x 6-ft Il l inois DOT (in.) >6.0 >6.0 Joint seal Yes PaveME (% slabs) 6.0 0 Fibers No Features on CO SH-121B include a thicker asphalt layer on the poor segment, in-place BCOA thicker than BCOA-ME design method recommendations, and ESALs exceeding the Illinois DOT recommendation. PaveME predicts 6% of slabs on the good segment will show longitudinal cracking over the design period. Figure 95. Design method results for (a) good segment and (b) poor segment of CO SH-121B.

112 Evaluation of Bonded Concrete Overlays on Asphalt Pavements (a) (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 25 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 25 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 5.5 in. BCOA (in.) 5.8 5.8 BCOA-ME (in.) 4.9 4.1 In-service age (years) 21 Asphalt (in.) 3.4 3.4 Colorado DOT (in.) >8.0 >8.0 Est. ESALs to date 4,000,000 Slab size 12- x 12-ft Il l inois DOT (in.) >6.0 >6.0 Joint seal Yes PaveME (% slabs) 9 1 Fibers No Features on CO US-6 include an in-place BCOA thicker than BCOA-ME recommendations. PaveME predicts 9% and 1% of slabs on the good and poor segments, respectively, will show longitudinal cracking over the design period. NOTE: Because of challenges with faultmeter testing, faulting is based on an automated distress survey. Figure 96. Design method results for (a) good segment and (b) poor segment of CO US-6.

Performance Prediction 113   (a) (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 6 in. BCOA (in.) 6.5 6.5 BCOA-ME (in.) 4.1 4.1 In-service age (years) 7 Asphalt (in.) 9.3 7.6 Colorado DOT (in.) 6.5 >8.0 Est. ESALs to date 3,800,000 Slab size 6- x 6-ft Il l inois DOT (in.) >6.0 5.8 Joint seal No PaveME (% slabs) 0 0 Fibers No Features on IA US-71 include a thicker asphalt layer and in-place BCOA thicker than design method recommendations, excluding the Colorado DOT recommendation for the poor segment. PaveME predicts no longitudinal cracking over the design period. Figure 97. Design method results for (a) good segment and (b) poor segment of IA US-71.

114 Evaluation of Bonded Concrete Overlays on Asphalt Pavements (a) (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 5.25 in. BCOA (in.) 5.2 5.5 BCOA-ME (in.) 4.0 5.1 In-service age (years) 16 Asphalt (in.) 8.6 2.8 Colorado DOT (in.) 7.0 >8.0 Est. ESALs to date 500,000 Slab size 5.5- x 5.5-ft Il l inois DOT (in.) >6.0 5.5 Joint seal No PaveME (% slabs) 0 0 Fibers No Features of IL CH-27 include low ESALs and in-place BCOA within design recommendations for the BCOA-ME poor segment only. PaveME predicts no longitudinal cracking over the design period. Initial IRI is not available. Figure 98. Design method results for (a) good segment and (b) poor segment of IL CH-27.

Performance Prediction 115   (a) (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME 0 20 40 60 80 100 120 0 50 100 150 200 250 300 350 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 4 in. BCOA (in.) 4.0 3.5 BCOA-ME (in.) 3.3 3.3 In-service age (years) 7 Asphalt (in.) 5.6 12.7 Colorado DOT (in.) >8.0 3.0 Est. ESALs to date 9,200,000 Slab size 4- x 4-ft Il l inois DOT (in.) >6.0 3.0 Joint seal No PaveME (% slabs) 23 0 Fibers Yes Features on IL SR-53 include a thick asphalt layer on the poor segment, in-place BCOA slightly thicker than design method recommendations, and higher ESALs exceeding the Illinois DOT recommendation. PaveME predicts 23% of slabs on the good segment will show longitudinal cracking over the design period. Figure 99. Design method results for (a) good segment and (b) poor segment of IL SR-53.

116 Evaluation of Bonded Concrete Overlays on Asphalt Pavements (a) (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 6 in. BCOA (in.) 5.9 5.5 BCOA-ME (in.) 4.4 4.4 In-service age (years) 8 Asphalt (in.) 18.1 26.2 Colorado DOT (in.) 3.0 3.0 Est. ESALs to date 8,400,000 Slab size 6- x 6-ft Il l inois DOT (in.) >6.0 >6.0 Joint seal No PaveME (% slabs) 0 8 Fibers No Features on KS I-70 include a thick asphalt layer, in-place BCOA thicker than BCOA-ME and Colorado DOT recommendations, and higher ESALs exceeding the Illinois DOT recommendation. PaveME predicts 8% of slabs on the poor segment will show longitudinal cracking over the design period. NOTE: Because of challenges with faultmeter testing, faulting is based on an automated distress survey. Figure 100. Design method results for (a) good segment and (b) poor segment of KS I-70.

Performance Prediction 117   (a) (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 4 in. BCOA (in.) 14.0 4.6 In-service age (years) 21 Asphalt (in.) unknown 8.9 Est. ESALs to date 18,100,000 Slab size 4- x 4-ft Joint seal Yes Fibers No BCOA-ME (in.) >5.5 4.8 Colorado DOT (in.) >8.0 5.8 Il l inois DOT (in.) >6.0 >6.0 PaveME (% slabs) 0 0 Features on LA US-167 include an in-place BCOA for the good segment thicker than all design procedure recommendations but similar to BCOA-ME for the poor segment, and high ESALs exceeding the Illinois DOT recommendation. PaveME predicts no longitudinal cracking over the design period. NOTE: Because of challenges with faultmeter testing, faulting for the good segment is based on an automated distress survey. Figure 101. Design method results for (a) good segment and (b) poor segment of LA US-167.

118 Evaluation of Bonded Concrete Overlays on Asphalt Pavements Figure 102. Design method results for (a) good segment and (b) poor segment of LA US-425. (a) (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 4 in. BCOA (in.) 4.5 3.8 BCOA-ME (in.) 4.7 4.7 In-service age (years) 16 Asphalt (in.) 9.5 10.1 Colorado DOT (in.) 5.0 4.5 Est. ESALs to date 11,100,000 Slab size 4- x 4-ft Il l inois DOT (in.) >6.0 >6.0 Joint seal Yes PaveME (% slabs) 4 4 Fibers Yes Features on LA US-425 include high ESALs, a thicker asphalt layer, and an in-place BCOA thickness similar to BCOA-ME and Colorado DOT recommendations. The Illinois DOT recommendation exceeds the design method thickness limit and ESAL limit. PaveME predicts approximately 4% of slabs will show longitudinal cracking over the design period.

Performance Prediction 119   Figure 103. Design method results for (a) good segment and (b) poor segment of MN CSAH-7. (a) (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 5 in. BCOA (in.) 6.6 4.9 BCOA-ME (in.) 4.1 4.0 In-service age (years) 10 Asphalt (in.) 6.7 8.1 Colorado DOT (in.) >8.0 3.0 Est. ESALs to date 1,300,000 Slab size 6- x 6-ft Il l inois DOT (in.) >6.0 >6.0 Joint seal No PaveME (% slabs) 0 0 Fiber No Features on MN CSAH-7 include an in-place BCOA thicker than BCOA-ME and Colorado DOT (for the poor segment) recommendations. Colorado (good segment) and Illinois DOT recommendations exceed the design method thickness limit. PaveME predicts no longitudinal cracking over the design period.

120 Evaluation of Bonded Concrete Overlays on Asphalt Pavements Figure 104. Design method results for (a) good segment and (b) poor segment of MN CSAH-22. (a) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 6 in. BCOA (in.) 6.6 7.1 BCOA-ME (in.) 4.9 3.7 In-service age (years) 8 Asphalt (in.) 3.2 6.5 Colorado DOT (in.) >8.0 8.0 Est. ESALs to date 2,800,000 Slab size 6- x 6-ft Il l inois DOT (in.) >6.0 >6.0 Joint seal Yes PaveME (% slabs) 1 0 Fibers No Features on MN CSAH-22 include an in-place BCOA thicker than the BCOA-ME recommendation. PaveME predicts 1% of slabs on the good segment will show longitudinal cracking over the design period.

Performance Prediction 121   Figure 105. Design method results for (a) good segment and (b) poor segment of MN I-35. (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME (a) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 6 in. BCOA (in.) 6.0 6.5 BCOA-ME (in.) 4.3 4.2 In-service age (years) 10 Asphalt (in.) 14.0 14.0 Colorado DOT (in.) 3.0 >8.0 Est. ESALs to date 11,800,000 Slab size 6- x 6-ft Il l inois DOT (in.) >6.0 >6.0 Joint seal Yes PaveME (% slabs) 0 0 Fibers No Features on MN I-35 include in-place BCOA thicker than BCOA-ME and Colorado DOT recommendations and high ESALs exceeding the Illinois DOT recommendation. PaveME predicts no longitudinal cracking over the design period.

122 Evaluation of Bonded Concrete Overlays on Asphalt Pavements Figure 106. Design method results for (a) good segment and (b) poor segment of MN TH-30. (b) 0 25 50 75 100 125 150 0 50 100 150 200 250 300 0 5 10 15 20 25 30 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME (a) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 25 30 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 6 inches BCOA (in.): 6.0 6.5 BCOA-ME (in.): 4.5 4.5 In-service age (years) 26 Asphalt (in.): 5.7 5.5 Colorado DOT (in.): >8.0 >8.0 Est. ESALs to-date 220,000 Slab size (ft): 12x12 Il l inois DOT (in.): >6.0 6.0 Joint seal Yes PaveME (% slabs): 0 0 Fibers No Features on MN TH-30 include low ESALs and an in-place BCOA thicker than the BCOA-ME recommendation. PaveME predicts no longitudinal cracking over the design period.

Performance Prediction 123   Figure 107. Design method results for (a) good segment and (b) poor segment of MO US-60. (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME (a) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 4 in. BCOA (in.) 4.5 4.3 BCOA-ME (in.) 4.2 3.7 In-service age (years) 20 Asphalt (in.) 4.7 5.5 Colorado DOT (in.) >8.0 >8.0 Est. ESALs to date 28,000,000 Slab size 4- x 4-ft Il l inois DOT (in.) >6.0 3.8 Joint seal No PaveME (% slabs) 90 90 Fibers Yes Features on MO US-60 include in-place BCOA thicker than the BCOA-ME recommendation and high ESALs exceeding the Illinois DOT recommendation. PaveME predicts 90% of slabs will show longitudinal cracking over the design period.

124 Evaluation of Bonded Concrete Overlays on Asphalt Pavements Figure 108. Design method results for (a) good segment and (b) poor segment of MT SR-16. (b) (a) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 4 in. BCOA (in.) 4.5 4.3 BCOA-ME (in.) >5.5 >5.5 In-service age (years) 18 Asphalt (in.) 3.7 2.8 Colorado DOT (in.) >8.0 >8.0 Est. ESALs to date 4,500,000 Slab size 4- x 4-ft Il l inois DOT (in.) >6.0 >6.0 Joint seal No PaveME (% slabs) 0 22 Fibers No Features on MT SR-16 include in-place BCOA thinner than design method recommendations. PaveME predicts 22% of slabs on the poor segment will show longitudinal cracking over the design period. NOTE: Because of challenges with faultmeter testing, faulting is based on an automated distress survey.

Performance Prediction 125   Figure 109. Design method results for (a) good segment and (b) poor segment of PA SR-119. (a) (b) 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME 0 20 40 60 80 100 120 0 50 100 150 200 250 300 0 5 10 15 20 Cr ac ki ng (% sl ab s) o r F au lti ng (m il) IR I ( in /m i) Age (years) IRI Long. Crack Total Crack Fault PaveME In- Project Details As-Built Information BCOA Thickness Recommendations Good Poor Good Poor Design thickness 6 in. BCOA (in.) 6.7 6.7 BCOA-ME (in.) 5.3 5.3 service age (years) 9 Asphalt (in.) 9.1 9.1 Colorado DOT (in.) 6.5 7.5 Est. ESALs to date 3,700,000 Slab size 6- x 6-ft Il l inois DOT (in.) 5.2 >6.0 Joint seal Yes PaveME (% slabs) 0 0 Fibers No Features on PA SR-119 include a thicker asphalt layer and an in-place BCOA thicker than the BCOA-ME and Illinois DOT recommendations but within Colorado DOT recommendations. PaveME predicts no longitudinal cracking over the design period. NOTE: Because of challenges with faultmeter testing, faulting is based on an automated distress survey.

126 Evaluation of Bonded Concrete Overlays on Asphalt Pavements Table 60. Design procedure results. Project ID Seg-ment Trans. Joint Spacing (ft) In- Service Age (years) Design Thick (in.) As- Built Thick (in.) Recommended Design Thickness (in.) Longitudinal Cracking (% slabs)a BCOA -ME CO DOT IL DOT Field Pave- ME CO I-70 Good 6 7 6.0 6.8 3.8a 3.0b >6.0 0.0 0.0 Poor 6 7 6.0 6.0 3.8b 3.3b >6.0 0.0 0.0 CO SH-83A Good 6 19 5.0 5.5 3.9b 6.3 5.0 0.9 0.0 Poor 6 19 5.0 10.4 3.9b >8.0b 5.2b 0.3 0.0 CO SH-83B Good 6 14 6.0 5.9 3.7b >8.0c >6.0 0.4 0.0 Poor 6 14 6.0 6.0 3.7b >8.0c >6.0 2.0 0.0 CO SH-121A Good 6 19 6.0 6.0 3.0b >8.0c 3.0b 0.3 0.0 Poor 6 19 6.0 5.7 3.0b 7.5c 3.0b 2.4 0.0 CO SH-121B Good 6 8 6.0 7.5 4.7b >8.0 >6.0b 0.7 0.5 Poor 6 8 6.0 6.7 3.7b 6.3 5.6b 2.2 0.0 CO US-6 Good 12 21 5.5 5.8 4.9 >8.0c >6.0 13.4 3.0b Poor 12 21 5.5 5.8 4.9 >8.0c >6.0 30.7 0.1b IA US-71 Good 6 7 6.0 6.5 4.1b 6.5 5.8 0.3 0.0 Poor 6 7 6.0 6.5 4.1b >8.0c 5.2b 0.0 0.0 IL CH-27 Good 5.5 16 5.25 5.2 4.0b 7.0c 5.5 2.2 0.0 Poor 5.5 16 5.25 5.5 5.1 >8.0c >6.0 0.8 0.0 IL SR-53 Good 4 7 4.0 4.0 3.3 >8.0c 3.0 0.2 2.4 Poor 4 7 4.0 3.5 3.3 3.0 >6.0c 0.1 0.0 KS I-70 Good 6 8 6.0 5.9 4.4b 3.0b >6.0 0.1 0.0 Poor 6 8 6.0 5.5 4.4b 3.0b >6.0 3.8 0.7 LA US-167 Good 4 21 4.0 14.0 >5.5b >8.0b >6.0b 0.1 0.0 Poor 4 21 4.0 4.6 4.8 5.8c >6.0c 0.0 0.0 LA US-425 Good 4 16 4.0 4.5 4.7 5.0 >6.0c 0.6 2.4 Poor 4 16 4.0 3.8 4.7 4.5 >6.0c 1.1 2.3 MN CSAH-7 Good 6 10 5.0 6.6 4.1b >8.0c >6.0 0.2 0.0 Poor 6 10 5.0 4.9 4.0b 3.0b >6.0 0.0 0.0 MN CSAH-22 Good 6 8 6.0 6.6 4.9b >8.0c > 6.0 0.0 0.2 Poor 6 8 6.0 7.1 3.7b 8.0 >6.0b 3.7 0.0 MN I-35 Good 6 10 6.0 6.0 4.3b 3.0b >6.0 2.2 0.0 Poor 6 10 6.0 6.3 4.2b >8.0c >6.0 0.1 0.0 MN TH-30 Good 12 26 6.0 6.0 4.5b >8.0c >6.0 0.1 0.0 Poor 12 26 6.0 6.5 4.5b >8.0c >6.0 5.6 0.0b MO US-60 Good 4 20 4.0 4.5 4.2 >8.0c 3.8 12.3 90.3c Poor 4 20 4.0 4.3 3.7 >8.0c 3.0b 0.8 90.4c MT SR-16 Good 4 18 4.0 4.5 >5.5 >8.0c >6.0c 1.4 0.1 Poor 4 18 4.0 4.3 >5.5c >8.0c >6.0c 1.0 18.3b PA SR-119 Good 6 9 6.0 6.7 5.3b 6.5 5.2b 0.1 0.0 Poor 6 9 6.0 7.1 5.3b 7.5 >6.0b 1.1 0.0 Underdesigned (no. of segments) 1 20 6 3 Adequately designed (no. of segments) 11 10 21 32 Overdesigned (no. of segments) 26 8 11 3 Underdesigned (% segments) 3 53 16 8 Adequately designed (% segments) 29 26 58 84 Overdesigned (% segments) 68 21 26 8 a Percentage of slabs with longitudinal cracking at 2019 in-service age. b Overdesigned (as-constructed thickness is at least 1 in. greater than recommended BCOA design thickness). c Underdesigned (as-constructed thickness is at least 1 in. less than recommended BCOA design thickness). On the basis of a simple regression analysis, there does not appear to be a strong correlation between PaveME-predicted and field-observed longitudinal cracking (R2 = 0.56 for 4- × 4-ft slabs, analyzed as 5- × 5-ft slabs, and R2 = 0.23 for 6- × 6-ft slabs) (Figure 110). However, the pre- dicted results from the CO US-6 poor segment, the MO US-60 good and poor segments, and the MT SR-16 poor segment may be outliers (>70% difference between predicted and field-observed longitudinal cracking). If the outliers are removed, a stronger correlation exists (R2 = 0.79 for 4- × 4-ft slabs analyzed as 5- × 5-ft slabs and R2 = 0.80 for 6- × 6-ft slabs) (Figure 111). A paired sample, two-tailed t-test was also performed on the PaveME results to determine whether predicted and field-observed longitudinal cracking are statistically the same (Table 61). The t-test results indicate there is no statistical significance when comparing predicted and

Performance Prediction 127   Figure 110. PaveME-predicted versus field-observed longitudinal cracking for all projects. R² = 0.5578 R² = 0.2259 0 20 40 60 80 100 0 20 40 60 80 100 O bs er ve d Lo ng itu di na l C ra ck in g (% sl ab s) PaveME Longitudinal Cracking (% slabs) 4x4 6x6 12x12 Figure 111. PaveME-predicted versus field-observed longitudinal cracking with outliers removed. 0 5 10 15 20 0 2 4 6 8 10 12 14 16 O bs er ve d Lo ng itu di na l C ra ck in g (% sl ab s) PaveME Longitudinal Cracking (% slabs) 4x4 6x6 12x12 R2 = 0.80 (6x6) R2 = 0.68 (All) R2 = 0.79 (4x4 analyized as 5x5) Table 61. PaveME-predicted versus field-observed longitudinal cracking, t-test results. Slab Size All Projects Outliers Removed t-test % diff R2 t-test % diff R2 4- x 4-ft Not equal −94 −0.08 Not equal −94 −0.08 6- x 6-ft Equal 161 0.15 Equal −61 0.52 12- x 12-ft Equal −100 na Equal −100 na All Equal 131 0.18 Not equal −70 0.51 NOTE: “Equal” indicates the null hypothesis can be accepted, that is, the means of the two data sets are equal. “Not equal” indicates the null hypothesis can be rejected, that is, the means of the two data sets are not equal. Negative R2 values indicate a negative relationship. na = not applicable.

128 Evaluation of Bonded Concrete Overlays on Asphalt Pavements field-observed longitudinal cracking (in percentage of slabs). However, this comparison evalu- ates all projects and does not consider project-to-project evaluation. For this, the Pearson cor- relation or coefficient of determination was used to determine whether the model prediction is correct. For this case, the regression showed a poor correlation between predicted and field- observed longitudinal cracking. However, when outliers were removed, the coefficient of deter- mination improved (R2 = 0.51). This can be considered a moderate linear relationship between observed and predicted longitudinal cracking. Summary Several methods for designing BCOA layer thickness are available and include BCOA-ME, Colorado DOT, Illinois DOT, and PaveME. BCOA design methods vary by input require- ment, input sensitivity, and slab size–BCOA thickness combinations evaluated. BCOA-ME, Colorado DOT, and Illinois DOT methods use ESALs for quantifying traffic loadings, whereas PaveME uses axle load spectra. BCOA-ME and PaveME use site-specific climate conditions, but Colorado DOT and Illinois DOT methods use concrete slab temperature gradient. All methods include input requirements for asphalt layer thickness. BCOA-ME and Colorado DOT methods include asphalt layer cracking and Colorado DOT and Illinois DOT methods include asphalt layer elastic modulus. All methods use k-value for characterizing the under- lying layer, and PaveME characterizes the unbound materials using layer properties. BCOA-ME allows the user to establish the design criteria according to cracking and reliability, Colorado DOT and Illinois DOT criteria are internally set, and PaveME criteria include reliability and design life. Failure criteria for BCOA-ME and Illinois DOT include the percentage of cracked slabs, PaveME uses the percentage of slabs with longitudinal cracking, and Colorado DOT uses asphalt and concrete fatigue. All methods quantify the concrete mixture using the modulus of rupture and elastic modulus. In addition, BCOA-ME and Illinois DOT include macrofiber strength, Colorado DOT and PaveME include Poisson’s ratio, and PaveME includes other concrete mixture properties. Each method varies in slab size and BCOA layer thicknesses eval- uated and includes the following specifications: • BCOA-ME: joint spacing ≥ 2- × 2-ft and ≤ 15- × 12-ft and 2- to 8-in. BCOA layer thickness. This method recommends a 3.0- to 5.5-in. BCOA layer thickness for 4- × 4-ft and 6- × 6-ft slabs and 5.5- to 6.5-in. BCOA layer thickness for 12- × 12-ft slabs. • Colorado DOT: joint spacing ≥ 6- × 6-ft (recommended) and ≤ 12- × 12-ft and minimum BCOA thickness of 3 in., with recommended BCOA layer thickness of 5 to 8 in. • Illinois DOT: joint spacing 4- × 4-ft or 6- × 6-ft and BCOA thickness 3 to 6 in. • PaveME: joint spacing 5- × 5-ft to 8- × 8-ft and 4- to 8-in. BCOA layer thickness. Sensitivity analyses of the BCOA-ME, Colorado DOT, and Illinois DOT design methods resulted in different concrete thicknesses for a given set of inputs. Because the procedures have different structural models, failure criteria, internal assumptions, and so on, this result is not unexpected. Results indicated that all BCOA design methods evaluated are sensitive to changes in asphalt layer thickness except for BCOA-ME. All design methods are sensitive to changes in ESALs. BCOA-ME and Illinois DOT design methods are insensitive to changes in k-value, the Colorado DOT design method is sensitive to changes in k-value, and PaveME is sensitive to changes in subgrade modulus. All BCOA design methods are sensitive to changes in concrete strength. For PaveME, because the poor section of CO US-6, the good and poor sections of MO US-60, and the poor section of MT SR-16 exhibit extremely high longitudinal cracking, these are removed from the analysis. After removing these outliers, a stronger correlation exists between PaveME-predicted and field-observed longitudinal cracking.