6533b7ddfe1ef96bd12747ce

RESEARCH PRODUCT

Fatigue crack initiation and subsequent crack growth in fillet welded steel joints

Tom LassenZbigniew Mikulski

subject

Materials scienceCarbon steelMechanical EngineeringMonte Carlo methodFatigue testingFatigue damageFracture mechanics02 engineering and technologyWeldingengineering.material021001 nanoscience & nanotechnologyIndustrial and Manufacturing Engineeringlaw.invention020303 mechanical engineering & transports0203 mechanical engineeringMechanics of MaterialslawModeling and SimulationengineeringGeneral Materials ScienceComposite material0210 nano-technologyFillet (mechanics)Stress intensity factor

description

Abstract The fatigue damage evolution in fillet welded steel joints where cracks are emanating from the weld toe is investigated. Based on existing experimental data for as-welded joints including crack depth measurements of the early crack growth it is proposed to make a distinction between the crack initiation phase and the subsequent crack growth phase. The welded detail in question is an F class detail with plate thickness 25 mm made of medium strength carbon steel. It is found that the crack initiation phase defined at a crack depth of 0.1 mm is close to 25% of the fatigue life even at a relatively high constant stress range of 150 MPa. At lower stress ranges it is concluded that the initiation phase is the dominating part of the fatigue life. The present work is focusing on the crack propagation phase that is defined from a crack depth of 0.1 mm to final failure of the detail. It is demonstrated that the recommendation given in rules and recommendations (DNVGL and BS 7910) for applying Linear Elastic Fracture Mechanics (LEFM) for the crack propagation phase is valid for the propagation of such small surface breaking cracks. A model based on the rule-based formulas for the Stress Intensity Factor Range (SIFR) and the growth parameters C and m in Paris law agree well the with the measured crack growth curves. For these small semi-elliptical cracks at the weld toe notch important topics like the existence of a threshold limit for the SIFR and the influence of the stress ratio R are discussed. Due to the inherent scatter in the variables characterizing the fatigue damage evolution stochastic approaches are applied for the analyses. Observations and measurements are presented by descriptive statistics and simulations are carried out using Monte Carlo techniques.

https://doi.org/10.1016/j.ijfatigue.2018.11.014