Search

International audience; Quantification of different effects (nonlinearity, heating, thixotropy, and fatigue) occurring during fatigue tests on bituminous mixtures is presented in this paper. A focus is given on the nonlinearity phenomenon.Continuous fatigue tests and a test with specific protocol (called fatigue tests to estimate biasing effects) were performed in tension/compression mode on cylindrical samples of the same material. The analysis of results reveals that reversible effects (nonlinearity, heating, and thixotropy) are important (more than 90% decrease at 100,000 cycles for a strain amplitude of 100 μm/m at 10 Hz) and cannot be ignored when interpreting classical fatigue tests. The nonlinearity effects respect the time‐temperature superposition principle, and they are more pronounced at “high” temperature (at the same frequency). Direction of nonlinearity curve in the Cole‐Cole axes is shown to be independent of temperature and frequency for the considered range.

An experimental campaign on the characterization of the tri-dimensional linear viscoelastic (3D LVE) behaviour of 2 bituminous mixtures was carried out. These 2 different mixtures were produced using the same aggregates and bitumen. As additives, two types of fillers, hydrated lime and limestone filler, were used. The goal was to evaluate the influence of hydrated lime on linear viscoelastic properties of bituminous mixtures. The linear viscoelastic behaviour of bituminous mixtures was studied using complex modulus tests (tension-compression test on cylindrical specimens) at different temperatures (from −25 to +52 °C) and frequencies (from 0.003 to 10 Hz). Sinusoidal cyclic loadings of 50 µm/m axial strain amplitude were applied. In addition to axial stress and axial strain, radial strains were also measured. The complex modulus E* and the complex Poisson’s ratios ν* were then obtained and the three-dimensional (3D) LVE behaviour (with isotropy hypothesis) was then completely described. The Time-Temperature Superposition Principle (TTSP) was verified with good approximation. The same values of shift factor were obtained for uni-dimensional (1D) and tri-dimensional (3D) conditions. The 2S2P1D model (3D) developed by ENTPE team was used for modelling linear viscoelastic properties of the two mixtures. Effect of hydrated lime is discussed.

An ongoing study on influence of hydrated lime on mechanical properties of bituminous mixtures is carried out. In this paper we focus on Linear ViscoElastic (LVE) properties and fatigue behavior. Nine mixtures were produced with the same 35/50 penetration grade bitumen, three different bitumen contents (4.4%, 5.1%, 5.8%) and three different hydrated lime contents (0%, 1.25%, 2.5%). Hydrated lime replaces a part of limestone filler (the mass of removed filler for the same mass of added hydrated lime). LVE properties of mixtures (complex Young’s modulus) were determined using a tension/compression test developed at University of Lyon/ENTPE. In the same time, two points bending tests were used to evaluate fatigue resistance. Surface response methodology and ANOVA (ANalysis Of VAriance) were carried out on experimental data obtained for different responses in order to evaluate the influence of bitumen content and hydrated lime content on linear viscoelastic properties and fatigue resistance. Time-Temperature Superposition Principle (TTSP) was verified with good approximation using very close shift factors for all tested materials. The use of hydrated lime filler (instead of limestone filler) induces an increase of complex modulus. Its effect increases with hydrated lime content. Hydrated lime does not change fatigue resistance of mixtures.