Experimental analysis and neural network modelling of the rheological behaviour of powder injection moulding feedstocks formed with bimodal powder mixtures

Abstract

Feedstock behaviour during powder injection moulding (PIM), has a critical influence on the physical and mechanical properties of the final components. In order to quantify this behaviour, a rheological study has been performed using binary blends of stainless steel powders that exhibit various particle sizes, morphologies, and size distributions. The feedstocks were obtained by mixing the blended powders with a standard binder system, and their rheological properties were investigated using torque and capillary rheometry methods. The resulting data were employed to develop a neural network for advising on the selection of desirable solids loadings for the PIM feedstocks. The system asks the user to input the particle characteristics, blend composition, shear rate, and binder viscosity. By relating these input parameters to the recommended feedstock viscosity, the neural network enables the operator to identify the value of solids loading to be employed for production of optimal quality PIM components.