Papers by Author: Marzio Grasso

Abstract: Fatigue failures of rails often occur at the rail foot, since the geometry of this zone gives rise to stress concentrations under service loads or defects during rail manufacture and installation. In this paper, the fatigue behavior of cracks at the web/foot region of a rail is analyzed numerically. Analytical models in the literature for a semi-elliptical surface crack in a finite plate assume that the geometry of the front remains semi-elliptical during the whole propagation phase and the ellipse axes do not undergo translations or rotations. Fatigue tests show that this is not the case for such cracks in rails. A predictive model for crack growth has been developed by assuming an initial small crack at one probable initiation point between the web and foot of the rail in reference to a service condition loading. SIF values have been estimated by means of the finite element method and the plastic radius correction. The results attained were compared with crack growth experimental data.

Abstract: In order to carry out the railway axle design according to the “Damage Tolerance” philosophy, reliable crack-growth models for these kind of components are necessary. Indeed, damage tolerance principles have received more and more attention from railway technical community, thanks to its ambitious task concerning the inspection intervals prevision of railway components subjected to non-stationary loading conditions. In this paper, a simple routine is exposed that is able to calculate the characteristic parameters of the Linear Elastic Fracture Mechanics (LEFM) for a generic cracked mechanical component. Such parameters are then used in a two parameters propagation law to estimate the necessary time for a crack to become critical.

Abstract: A practice for analyzing some raw data from fatigue crack growth tests, carried out on specimens obtained from a railway component (wheel), is presented. Despite the data sample is small, it allows to identify the material parameters of the “threshold” propagation model that make this latter amazingly realistic for the examined material. The proposed analysis method is based on the Evolutionary Algorithms theory and uses raw data obtained from C(T) specimens instrumented with crack gauges.