|M.Sc Student||Matani Shihab|
|Subject||The Influence of Density and Moisture Content on the|
Resilient Modulus of Clayey Subgrdes
|Department||Department of Civil and Environmental Engineering||Supervisors||Dr. Arieh Sidess|
|Professor Mark Lloyd Talesnick|
|Full Thesis text|
strength of subgrades is one of the important parameters in design and
performance prediction of pavements in terms of fatigue, rutting, roughness, and so on. Traditionally, the Californian Bearing Ratio (CBR) test was used to characterize the strength of subgrades, neglecting the cyclic loading conditions. The development of rational performance-based design methods requires the subgrade materials to be
characterized by stress-strain parameters like the Resilient Modulus (𝑀𝑟).
Mechanistically the resilient modulus is a nonlinear elastic property that is used to
characterize roadbed soils for pavement design.
Nowadays the most advanced specifications for predicting pavement performance such as AASHTO Guide, employ the resilient modulus for computing stresses, strains and displacements in pavements layers. There are 3 reliability levels for evaluating the resilient modulus. Level 1 is the most accurate. The modulus is obtained from direct laboratory or field testing, which is necessary under heavy traffic. Level 2 is concerned with moderate accuracy, in which the resilient modulus is determined from correlations with basic soil properties. Level 3 is with the highest uncertainty, where the modulus is determined from typical values depending on soil classification.
Determining the resilient modulus from laboratory tests (Level 1) is technically complex, time-consuming, and costly. Hence evaluating the Level 2 resilient modulus is expedient for reliable correlations. The level 2 evaluation may also be implemented in Quality Control based on Performance Related Specifications (PRS), replacing the current empirical specifications.
The objective of this study is to characterize the resilient modulus of local clayey subgrades, and to develop correlations for estimating the 𝑀𝑟 from soil properties. For this, two sets of testing systems were conducted: (a) compaction and testing of samples representing different initial post-compaction conditions, 3 different density levels in the range of 87-93% of Modified AASHTO and 3 moisture contents between 20-26%. (b) Samples compacted, wetted and tested, representing post-compaction moisture variations.
All samples were subjected to Haversine-shaped repeated loadings according to AASHTO T-307, under combinations of 3 confinement pressures, and 5 deviatoric stresses.
Analysis of the resilient modulus results yielded the following conclusions: (1) The confinement and deviatoric stresses highly affect the resilient modulus, mostly the latter. (2) The resilient modulus decreases with reducing density or increasing moisture content, and is more sensitive to moisture. (3) The testing results were calibrated to the Universal model. (4) Prediction correlations of the models' ki-coefficients from density and moisture content were developed. Combining these two properties into the Activity Ratio (𝐴𝑟) improved the correlations. (5) An alternative two parameters model was developed based on the Triaxiality Ratio. (6) The wetting process decreases the resilient modulus drastically, 2-4 times compared to the same initial moisture content. (7) Prediction curves of resilient modulus deterioration in terms of ki-coefficients and moisture changes were developed.
The contribution of this research is in predicting the resilient modulus of subgrades from post-compaction properties. The model enables advanced prediction of pavements performance (AASHTO Guide), and advanced quality control (PRS).