Consideration of Reduction in Stiffness due to Cracking and the Impact on Standard Stress Intensity Factor Solutions

Abstract

An API 579-1/ASME FFS-1 Failure Assessment Diagram based Fitness-for-Service assessment was carried out on an embedded crack-like flaw found in a nozzle to shell weld in a pressure vessel. Stress intensity factors were initially calculated by utilizing stress results from a Finite Element Analysis (FEA) of an uncracked configuration, with the standard embedded crack stress intensity factor solution given in API 579-1/ASME FFS-1. Due to the complex nozzle geometry and flaw size, a second analysis was carried out, incorporating a crack into the FEA model, to calculate the stress intensity factors and evaluate if the standard solution could be applied to this geometry. A large difference in the resulting stress intensity factors was observed, with those calculated by the FEA with the crack incorporated into the model to be twice as high as those calculated by the standard solutions, indicating the standard embedded crack stress intensity factor solution may be non-conservative in this case. An investigation was carried out involving a number of studies to determine the cause of the difference. Beginning with an elliptical shaped embedded crack in a plate, the stress intensity factor calculated with an idealized 3D crack mesh agreed with the API 579-1/ASME FFS-1 solution. Examining other crack locations, and crack shapes, such as a constant depth embedded crack, revealed how the solution began to differ. The greatest difference was found when considering a crack mesh with a small component height (i.e. the distance measured perpendicular from the crack face to the top of the mesh). A close agreement was then found between the stress intensity factors calculated in the nozzle model and an idealized crack mesh with component heights representative of the true geometry. This revealed that reduced structural stiffness is a key factor in the calculation of the stress intensity factors for this geometry, due to the close proximity of the embedded crack to the inner surface of the nozzle. It was found that this reduction is potentially significant even with relatively small crack sizes. This paper details the investigation, and aims to provide the reader with an awareness of situations when the standard stress intensity factor solutions may no longer be valid, and offers general recommendations to consider when calculating stress intensity factors in these situations.