Laboratory remolded subgrade soil samples have been widely used to study subgrade resilient modulus (M sub r). But physical conditions, such as moisture content and density, of such specimens may not represent in situ conditions very well. Therefore, AASHTO and the Long-Term Pavement Performance program have recommended that undisturbed, thin-walled tube samples be used to study subgrade resilient behavior. The Minnesota Department of Transportation (MnDOT) is developing mechanistic-empirical pavement design approaches through the Minnesota Road Research project and has realized the importance of M sub r in the design approaches. Currently, MnDOT is making an effort to study the M sub r of unbound pavement materials through laboratory experiments. Under a research project at MnDOT, several thin-walled tube samples of subgrade soil were obtained from six different pavement sections at the Minnesota Road Research project. Repeated loading triaxial tests were conducted on the soil specimens to determine the M sub r at the MnDOT laboratory. Also, some soil properties, such as resistance, R-value, and plasticity index, were obtained. R-value is an indicative value of performance when soil is placed in the subgrade of a road subjected to traffic. Two constitutive models (the Uzan-Witczak universal model and the deviator stress model) were applied to describe the M sub r. The objectives of the research were to compare these two well-known constitutive models in describing subgrade soil resilient behavior and to study the effects of material properties on the M sub r. From the specimens tested, the experimental results showed that the universal model described the subgrade M sub r slightly better than the deviator stress model, and the coefficients in these two constitutive models were found to have correlation to material properties. Also, no well-defined relationships between the R-value and the coefficients in the constitutive models were observed from the results of the tested specimens.
Abstract