Kafie Martinez, Jackeline (2017-12). On Fatigue Life Estimations of Riveted Details under Variable Amplitude Loading. Doctoral Dissertation. Thesis uri icon

abstract

  • Old riveted railroad bridges currently in service in North America are being subjected to much higher live loads that they were originally designed for. Current fatigue provisions have been proven to provide robust estimations for riveted details. The purpose of this study is to determine how an increase of axle loads ultimately affect the fatigue life of riveted bridges. A linear elastic fracture mechanics approach along with a cycle-by-cycle integration method, is used to estimate fatigue life of riveted connections in railroad bridges under variable amplitude loading. Only the stress ranges with stress intensity factors larger than the fatigue threshold are assumed to produce crack growth. The analysis is developed using the variable amplitude stresses generated due to a traversing heavy axle train on a simply supported riveted girder. Stress intensity factors for quarter elliptical corner crack emanating from a through- thickness (rivet) hole under remote tension were used. Effect of the initial crack size assumption on the fatigue life estimation was studied. Moreover, the effect of overloads on the fatigue behavior of the component was also studied. Special care has been taken to analyze the effect on fatigue life of thermal residual stresses due to original rivet installation. Thermal residual stresses, generated during the installation process, were quantified through a finite element simulation. Superposition principle was utilized to account for localized residual stresses present on the base material. Fatigue life estimations calculated using S-N curves (current AREMA provisions) and linear elastic fracture mechanics with and without considering clamping stress are compared. Plate thickness plays a role in the fatigue life of the components; thin plates count with a larger local compressive stress that hinders crack propagation in this direction. Crack growth proved to be different under localized residual stresses when compared to the base material. Finally, the ratio of crack depth to plate thickness also was found to play a role in the fatigue life.
  • Old riveted railroad bridges currently in service in North America are being subjected to much higher live loads that they were originally designed for. Current fatigue provisions have been proven to provide robust estimations for riveted details. The purpose of this study is to determine how an increase of axle loads ultimately affect the fatigue life of riveted bridges.
    A linear elastic fracture mechanics approach along with a cycle-by-cycle integration method, is used to estimate fatigue life of riveted connections in railroad bridges under variable amplitude loading. Only the stress ranges with stress intensity factors larger than the fatigue threshold are assumed to produce crack growth.
    The analysis is developed using the variable amplitude stresses generated due to a traversing heavy axle train on a simply supported riveted girder. Stress intensity factors for quarter elliptical corner crack emanating from a through- thickness (rivet) hole under remote tension were used. Effect of the initial crack size assumption on the fatigue life estimation was studied. Moreover, the effect of overloads on the fatigue behavior of the component was also studied.
    Special care has been taken to analyze the effect on fatigue life of thermal residual stresses due to original rivet installation. Thermal residual stresses, generated during the installation process, were quantified through a finite element simulation. Superposition principle was utilized to account for localized residual stresses present on the base material.
    Fatigue life estimations calculated using S-N curves (current AREMA provisions) and linear elastic fracture mechanics with and without considering clamping stress are compared. Plate thickness plays a role in the fatigue life of the components; thin plates count with a larger local compressive stress that hinders crack propagation in this direction. Crack growth proved to be different under localized residual stresses when compared to the base material. Finally, the ratio of crack depth to plate thickness also was found to play a role in the fatigue life.

publication date

  • December 2017