This report presents causes of cracking in asphalt concrete pavement in North Carolina through field investigation and laboratory experiments with field extracted material. Specific objectives of this research were to: (1) investigate whether systematic bias exists in the NCDOT's volumetric mix design that pertains to such factors as dry asphalt concrete mixtures, aggregate structures, etc.; (2) investigate the strength and uniformity of the pavement substructure; and (3) identify dominant causes of premature crack propagation patterns, such as top-down cracking, bottom-up cracking, and bidirectional cracking. A total of 34 pavement sites were selected for this study that constitute 6 old and well performing pavements and 24 young and poor performing pavements. The research team visited these 34 pavement sites and conducted the following investigations: (1) visual condition survey, (2) falling weight deflectometer testing, (3) full depth coring and visual observation of cracking patterns, and (4) dynamic cone penetrometer testing. In order to assess condition of existing pavement, alligator cracking index (ACI) was developed by combining the amounts of alligator cracking and longitudinal cracking on the wheel path. Deduct values for low, moderate, and high severity cracking were subtracted from 100 to determine the ACI value. Therefore, higher ACI values indicate a better condition than lower ACI values. In addition to the in-situ testing, laboratory tests were performed on the cores, including: (1) dynamic modulus and fatigue testing, (2) ignition oven testing, and (3) frequency sweep and linear amplitude sweep testing of extracted and recovered binder. Finally, pavement performance simulations were done using the Pavement ME program and the Layered ViscoElastic pavement analysis for Critical Distresses (LVECD) program.
It was found that the asphalt content in the top layer that exhibits top-down cracking or bottom-up cracking has a proportional relationship to ACI values. The air void content in a bottom layer that exhibits top-down cracking or bottom-up cracking shows an inverse proportional relationship to ACI values. These observations reflect reasonable results. A comparison between ACI and asphalt film thickness values does not produce noteworthy findings, but somewhat reasonable results are evident once the range of comparison is narrowed down. Thicker film thicknesses show higher ACI values. From field core visual observations, road widening is identified as a major cause of longitudinal cracking. Regions with observed layer interface separation (debonding) tend to have low ACI values. Through tensile strain simulation based on actual field conditions, it is observed that sites with observed bottom-up cracking have higher tensile strain levels at the bottom of the asphalt layer than sites with observed top-down cracking. Extracted binder fatigue test results indicate that the performance difference between good and poor sections of a given site is not the result of differences in the binder properties. Hence, other mixture design factors are at work in controlling the site variability in terms of fatigue resistance.
