Base Course Resilient Modulus for the Mechanistic-Empirical Pavement Design Guide

Abstract

The Mechanistic-Empirical Pavement Design Guidelines (MEPDG) recommend use of modulus in lieu of structural number for base layer thickness design. Modulus is nonlinear with respect to effective confinement stress, loading strain, and moisture. For design purposes, a single effective modulus of a base layer is desirable, and this modulus should be able to approximately account for nonlinearities. However, the MEPDG does not describe a procedure for determining this single modulus value. This research focused on laboratory characterization of base modulus nonlinearity, developing a nonlinear response model using laboratory data for nonlinear pavement analysis, and a methodology to determine a single effective modulus for a base layer via the nonlinear response model. Resonant column tests were conducted on two base materials used in Florida to characterize shear modulus (G) nonlinearity under different confinements and moisture contents. The suction effect increases G in the strain range of 10 to the negative 5th power% to 10 to the negative 1st power%, with very significant increases at strain levels below 10 to the negative 3rd power%. Using laboratory nonlinear modulus data, a nonlinear response model was developed via the Plaxis finite element methodology. The model is an effective means for assessing the effects of unbound material nonlinearity on the response of pavements. A representative modulus can be determined by a backcalculation procedure in which surface deflections from a nonlinear analysis are matched via an equivalent linear analysis. The single effective modulus varies over a range of conditions, including the moisture content of the base, pavement layer thicknesses, and the modulus of the subgrade. There is a significant effect of moisture on the effective modulus of limerock base materials used in Florida and the modulus/moisture relationship employed in the MEPDG underpredicts this increase. An equivalent linear analysis using effective moduli for both an unbound base and the subgrade can predict the structural response of an asphalt surface layer in a flexible pavement. It should be possible to utilize these structural response predictions in the assessment of cracking performance of the surface layer. However, caution is warranted in predicting the structural response of the unbound base and subgrade layers using an equivalent linear analysis. Use of an effective modulus for a nonlinear base layer appears reasonable for very thick pavement structures, but appears to underpredict vertical strain at the top of subgrade as the nonlinearity increases. Use of effective moduli for both a nonlinear base and subgrade appears to underpredict top of subgrade vertical strain even for very thick pavements.