Failure analysis for the optimisation of internally pressurised additive manufactured components

Abstract

Additive manufacturing (AM) is increasingly recognized as a viable method for producing internally pressurized components with complex geometries that are infeasible to fabricate using conventional machining techniques. Among AM processes, Powder Bed Fusion using Laser for Polymers (PBF-LB/P), commonly referred to as Selective Laser Sintering (SLS), enables the fabrication of polyamide-12 (PA-12) components with optimized internal structures. However, ensuring the structural integrity of these components under internal pressure conditions remains a critical challenge. In this work, a response surface methodology (RSM) model was developed to predict burst pressure in elbowed components fabricated via PBF-LB/P, using data obtained from hydraulic fracture tests. The model achieved a coefficient of determination (R²) of 0.974 and a root mean square error (RMSE) of 9.56 bar, demonstrating high predictive accuracy. Additionally, a comparative analysis with classical burst pressure models revealed that most traditional models significantly overestimate burst pressures, except for Faupel’s model, which showed the closest agreement with experimental results. Furthermore, Scanning Electron Microscopy (SEM) analysis confirmed that failure occurs predominantly through brittle fracture mechanisms, with no significant difference in fracture morphology between thin-walled and thick-walled components. These findings highlight the importance of additive manufacturing process parameters in failure behaviour and validate the applicability of response surface modelling for predicting burst pressure in AM components. This study represents a novel contribution by applying response surface methodology (RSM) to the failure prediction of PBF-LB/P-fabricated pressure components. Unlike previous works focused on isotropic materials or straight-walled geometries, this work targets the burst pressure prediction of elbowed, anisotropic AM parts, an area with limited prior exploration. Additionally, this is the first time that classical failure models are systematically benchmarked against experimental results for PBF-LB/P elbow geometries.