Experimental evaluation of through-thickness stresses in a welded pipe using the consistent-splitting model

Experimental data for evaluating through-thickness residual stresses in a girth-welded pipe have been reanalyzed using an improvement in the analysis method. Data were obtained using a destructive experimental technique which involves three steps: parting-out, splitting, and layer removal. The improved analysis, called the consistent-splitting model, has been applied to the data in the splitting step. In addition to satisfying necessary equilibrium and compatibility conditions for the splitting step, the consistent-splitting model is easy to apply and requires no more data than are usually gathered for existing analysis models. Results from the consistent-splitting model show substantial corrections when compared to the originally computed residual-stress distributions. Comparisons of the results based on the consistent-splitting model with values from a finite-element computational model show generally good agreement. Questions about the assumptions embodied in analysis methods applied to the parting-out and layer-removal steps of the experimental procedure suggest future areas of potential improvement.