Fatigue Crack Growth (FCG) Analysis  
The FCG is a complicated process that is sensitive to numerous factors, including material properties, microstructure, environment, load spectrum, crack size (small crack behavior) and crack orientation. The current state of the art for predicting FCG considers the effects of various factors by using empirically determined material properties and parameters for a given situation. Therefore, to accurately predict FCG for a given situation, it is important to use the material properties and parameters determined for the given condition. A review of the literature at the time of the work showed that a very limited amount of material property data useful in FCG analysis exists for tank wall materials (e.g., A51670, TC128B, and A515 for older tank cars). There was no FCG data available in the literature for TC128B. Several sources of fatigue data for A51670 were found in the literature. It is also noted that no data exists to consider load sequence effects under spectrum loading for these steels. To predict FCG under variable Rratio tank car load spectrum by using the available constant Rratio Paris equation, a “primary Rratio” concept is proposed in order to have a closer match of the actual Rratios with appropriate Paris equation parameters. The primary Rratio at a given potential fatigue critical location (PFCL) is the Rratio range of the load steps that contributes most to the fatigue damage in the spectrum. The primary Rratio range will be either (a) low Rratio range (LR): 0<R<0.2, (b) medium Rratio range (MR): 0.2<R<0.5, or (c) high Rratio range (HR): R>0.5. FCG analyses of the seventeen identified PFCLs identified that the PFCL designated as N1 is the most critical. Thus, location N1 controls the safety of the tank shell, provided that equal size and shape of initial cracks exist at each PFCL. The fatigue critical location (FCL), N1, is located on the external surface of the head at the weld of the front sill pad, as shown in the lowerleft figure, along with the crack parameters. The lowerright figure compares inspection intervals predicted for A515 steel tank car of different initial crack cases with and without welding residual stresses. It can be seen that the tensile residual stresses can significantly reduce the inspection intervals. 

Full Paper 