Improved Fatigue Life Estimation Of Tubular Joints

ABSTRACT This paper presents previously unpublished results of over 50 elastic tests on large scale steel tubular joints of realistic size and construction. The tests encompass a wide range of joint configurations and geometries including the extremes of practical range where few data are available. These results, together with a screened database of SCF and fatigue results, are used to conduct a two-level exhaustive reliability analysis for fatigue life estimation of tubular joints. Comparisons with existing parametric SCF formulae are made, and the reliability of these formulae when coupled with the family of S-N curves currently in use is examined. The fatigue life estimation of tubular joints is an important design tool for offshore structures, particularly for deeper water structures and harsher environments. The new test data and the associated reliability assessments presented in this paper will enable the estimation of fatigue lives to known levels of confidence. The recommended design approach provides a mechanism for incorporation of varying degrees of safety in line with the adopted safety and reliability philosophy for the structure. This can be used to establish a priority basis for developing a selective inspection programme to be followed during the life of the installation. INTRODUCTION Steel offshore jacket structures consist primarily of tubular joints which are formed by the intersection of brace and chord members, with the outside diameter of the brace less than or equal to the outside diameter of the chord. The complex intersection gives rise to severe stress concentrations; wave loading on a tubular structure cause fluctuations in these stress levels at the joints, leading to fatigue crack growth and eventual failure fatigue failure is defined as the number of stress cycles (and hence the time) taken to reach a pre-defined failure criterion. Fatigue analysis is by no means a rigorous science and the idealisations and approximations inherent in it prevent the calculation of abs9lute fatigue life (time to failure) for even the simplest joint. Any fatigue analysis approach encounters at least four areas of considerable unpredictability:The operational environment of a structure and the relationship between the environment and the actual forces in a structure.The internal stresses at a critical point in the structure (around the intersection of a tubular joint, for example) induced by external forces acting on the structure.The time to failure due to the accumulated stress history at the critical point.The definition of 'failure' used in design. The potential for error in the above predictions clearly precludes the exact calculations of absolute time to failure. Nevertheless, fatigue analysis is an important design tool for the prediction of relative magnitudes of fatigue lives at potentially critical points. The calculated lives enable critical parts to be ranked in terms of fatigue sensitivity and may be used