Reflective Cracking Study: HVS Test Section Forensic Investigation

This report is one in a series of first-level analysis reports that describe the results of HVS testing on a full-scale experiment being performed at the Richmond Field Station (RFS) to validate Caltrans overlay strategies for the rehabilitation of cracked asphalt concrete. It describes the results of the forensic investigation on the HVS rutting sections (Sections 580RF through 581RF) and HVS reflective cracking testing sections (Sections 586RF through 591RF). The study forms part of Partnered Pavement Research Center Strategic Plan Element 4.10: “Development of Improved Rehabilitation Designs for Reflective Cracking.” Findings and observations based on the data collected during this forensic investigation include: • There was considerable variation in the thicknesses of the constructed layers of the test road. • In the rutting experiments, rutting occurred primarily in the underlying DGAC and not in the overlay. In the reflective cracking experiments, rutting occurred in both layers. Very little rutting occurred in the underlying layers. • Cracks were observed on some of the test pit profiles. In the underlying DGAC layer, cracks were generally clearly visible. However, in the overlays, heat generated from the saw cut operation appeared to seal any cracks and no conclusions could be drawn as to the depth that cracks had reflected into the overlays. Most cracks appeared to have initiated close to the bottom of the underlying DGAC. Some crack initiation was also observed at poorly bonded joints between lifts and overlays in the AR4000-D section. • Some post-construction cementation of the base material appeared to have occurred. This was substantiated with DCP tests, close inspection of the test pit profile, use of phenolphthalein to determine the pH of the base material, and examination of specimens under optical and scanning electron microscopes. This recementation appears to have contributed to the good performance of the sections. • Base material density was generally consistent over the section. Nuclear gauge determined wet densities averaged 2,176 kg/m3, which corresponds with the average of 2,200 kg/m3 recorded after construction. • Nuclear gauge-determined base moisture contents averaged 11.1 percent for the 18 test pits, with higher moisture contents in the top 50 mm compared to the remainder of the layer. This is higher than the predetermined optimum (8.9 percent) and the laboratory-determined gravimetric moisture contents (8.7 percent). • Subgrade densities were not measured. The average subgrade moisture content was 15 percent, considerably higher than the base moisture content. The presence of mottling in the subgrade material indicates that the moisture content probably fluctuated seasonally. • Air-void contents were lower in the wheelpath after HVS testing compared to before HVS testing, as expected. The findings of this investigation confirm the conclusions of the other first-level analysis reports on HVS testing. No recommendations as to the use of modified binders in overlay mixes are made at this time.