No abstract

This paper presents the results of the first phase of a research project dealing with the constitutive description of the behaviour of well-graded granular materials when used for base or subbase layers in flexible pavement structures (so-called ''unbound granular materials'', UGMs). Monotonic and cyclic loading is under consideration. The present paper concentrates on test results and the constitutive description of monotonic loading. Hypoplasticity in the version proposed by von Wolffersdorff is used as the constitutive model. Sets of material constants for typical UGM materials do not exist in the literature. The experimental determination of a set of constants according to the procedure proposed by Herle is described in this paper. In the monotonic triaxial tests specimens with a square cross-section were used. The paper presents a preliminary test series comparing triaxial results obtained with cylindrical and with prismatic specimens. Re-calculations of the element tests are also presented. The simulations show a good congruence with the experiments.

Cyclic stresses due to passing wheels impose an accumulation of permanent strains in layers of unbound granular materials (UGMs) of flexible pavements. The hollow cylinder triaxial test would be the most appropriate test to simulate the in situ stress conditions but it is difficult to perform on UGMs due to their large maximum grain size. The simpler axi-symmetric cyclic triaxial test does not consider the shear stress components. It can be performed with a constant (CCP) or a variable confining pressure (VCP). CCP and VCP tests are commonly assumed to deliver similar residual and resilient strains as long as the average stress is the same. Thus, the simpler CCP test is mostly used in pavement engineering. However, this assumption is based on limited test data in the literature and may not be on the safe side. The present paper documents a comparative study of CCP and VCP tests on UGM. The study is mainly dedicated to the permanent deformations. The results show that only for some special stress paths both types of test deliver similar permanent axial or volumetric strains. For some other stress paths the CCP test may underestimate the permanent axial strain in comparison to the corresponding VCP test.

No abstract

No abstract