Buried pipelines subjected to non-continuous ground movement such as frost heave, thaw settlement, slope instability and seismic movement experience high compressive strains that can cause local buckling (or wrinkling), in which the pipe wall buckles like a thin cylindrical shell in axial compression. In a strain-based design and assessment framework, excessive local buckling deformation that may cause loss of serviceability, or even pressure containment in some cases, is managed by limiting the strain demand below the strain limit. The determination of compressive strain limit is typically ...
Buried pipelines subjected to non-continuous ground movement such as frost heave, thaw settlement, slope instability and seismic movement experience high compressive strains that can cause local buckling (or wrinkling), in which the pipe wall buckles like a thin cylindrical shell in axial compression. In a strain-based design and assessment framework, excessive local buckling deformation that may cause loss of serviceability, or even pressure containment in some cases, is managed by limiting the strain demand below the strain limit. The determination of compressive strain limit is typically performed by full-scale structural testing or nonlinear finite element analysis that takes into account material and geometric non-linearity associated with the inelastic buckling of cylindrical shells. Before performing testing and numerical analysis (or when such options do not exist), empirical equations are used to estimate the strain limit. In this report a number of representative equations were evaluated by comparing strain limit predictions to full-scale test results. Work prior to this study has identified the importance of key variables that have the greatest impact on the local buckling behaviour. Examples of these variables include the diameter-to-thickness (D/t) ratio, internal pressure and shape of the stress strain curve. The evaluation focused on how existing equations address these key variables, and the performance with respect to key variables and in different ranges.