Nondestructive field assessment of flexible pavement and foundation layers
Pavana K. Vennapusa
Falling weight deflectometer (FWD) and ground penetrating radar (GPR) are nondestructive test devices widely used by transportation agencies to assess pavement conditions. The two papers in this thesis evaluated the uncertainties associated with interpreting data from these devices and assessed potential applications.
In the first paper, FWD tests were conducted on asphalt pavements with varying supporting conditions, and individual layer modulus values were estimated using forward-and back-calculation methods. Dynamic cone penetrometer (DCP) test device was used to independently measure individual layer penetration resistance (PR) values to compare with the estimate moduli values. Results indicated that the predicted subgrade moduli values from forward- and back-calculations are strongly correlated but produce slightly different values. The predicted asphalt and base layer moduli values from forward- and back-calculations, however, showed significant scatter. Comparison between DCP-PR and the predicted base and subgrade layer modulus yielded non-linear relationships. The relationships produced lower standard errors when only data from subgrade layer is considered. The relationships developed in this study fell within the upper and lower bounds of relationships documented in the literature.
In the second paper, the efficacy of using a ground-coupled GPR system and a hand-held dielectric property measurement device to determine the asphalt and pavement foundation layer thicknesses is assessed. The actual pavement thicknesses were measured from pavement cores and foundation layer thicknesses were obtained using dynamic cone penetrometer (DCP) tests. Further, the viability of using GPR to detect moisture variations in the base layers is assessed. Tests were conducted on various asphalt pavement test sections built at a test site in Iowa with different foundation support and drainage conditions, and layer thicknesses. A comparative analysis of core measurements and asphalt thickness estimated from GPR showed a 10% average error. Base layer thicknesses could not be evaluated using GPR data due to variations in moisture contents. Based on the dielectric properties calculated from GPR scans, the estimated moisture contents in the base layer varied from about 5 to 15%. The variations in moisture contents between the test sections are attributed to variations in gradation and permeability properties of the base layer.