Abstract
This study presents the results of four years of data collection from a test site that is constructed as part of an actual roadway. The focus of the study was to evaluate the possibility of achieving similar or better performance characteristics from base course sections constructed with reclaimed asphalt pavement (RAP) blends compared to sections with 100 % virgin aggregate (VA). Two base course sections were constructed by blending VA with 20 % RAP and two with 30 % RAP (four RAP-VA sections). Blends were achieved using fine processed RAP with two different average binder contents (4.9 and 5.6 %). Additionally, two sections were constructed with 100 % VA for comparison. During construction, strain gauges were embedded into the asphalt and base course layers. Long-term performance evaluations were monitored based on data from the strain gauges and in situ testing. Falling weight deflectometer and light weight deflectometer tests were used to evaluate modulus values, and the results showed that all sections constructed with RAP-VA blends had equal or greater modulus values than the 100 % VA sections. International roughness index tests coupled with data from strain gauges were used to evaluate deformations. The results showed that three of the four RAP sections had deformations close to or less than 100 % VA sections (desired outcome). The only exception was in the section constructed with 30 % RAP with lower binder content (4.9 %). Findings from this study showed that field evaluations require multiple methods of testing, and modulus-based measurements alone should not be relied on for performance evaluations. This study demonstrated that it is possible to design base course sections with RAP that perform similar to or better than 100 % VA. A proper blend requires the gradation and binder content of RAP to be characterized and the percentage of RAP blended with VA to be determined based on these properties.