Inclusions have a detrimental effect upon casting properties but it is known empirically that a slow liquid metal flow has a beneficial effect by reducing the number of inclusions entering the casting. Positron Emission Particle Tracking (PEPT) is a method that can be used to track the path taken by radioactive particles, and can be used to follow the behaviour of inclusions as they make their way from a furnace, along the launder and into the casting. A new PEPT camera geometry was developed and used to track radioactively labelled γ-alumina particles, in the region of 600 μm size. The camera detectors were arranged radially around a launder, into which was poured 150 kg of liquid Al alloy, the radioactive particles being released at intervals during the pour. The positron camera was 0.7 m in length and 0.1 m2 in cross-sectional area, and this matched the dimensions of an industrial launder. A model of inclusion behaviour in a flowing liquid Al alloy was also developed, and the Positron Imaging system described was used to validate this model.

The behavior of inclusions in castings was studied using radioactive labeling of oxide particles located by Positron Emission Particle Tracking (PEPT). This uses the isotope18F, which has a half-life of 110 minutes, and allows particle detection within an accuracy of a few mm. Alumina and glass particles with a size range of 110 to 600 μm were made radioactive by an ion-exchange/surface adsorption process involving irradiated water. Individual radioactive particles were placed in resin-bonded sand moulds at known initial positions, and the moulds were filled with Al alloy, causing the particle to be entrained into the metal stream during the casting process. A modular γ-ray positron camera was used to track the paths of the particles within the mould, demonstrating the applicability of the technique to the study of inclusion behaviour in castings.

No abstract

Inclusions in castings are particularly damaging as crack initiators, causing failure in service. It is therefore desirable to attempt to remove them prior to their entry into the casting. In this work the behaviour of inclusions was studied using a computer simulation to predict the movement of particles in flowing liquid metal. The computer simulation was created by a combination of Computational Fluid Dynamics (CFD) and Discrete Element Modelling (DEM), to try to predict the agglomeration characteristics of inclusions in liquid Al alloy in a launder system. The model was tailored to simulate dispersed particles in a fluid, where particle collisions with each other and with side walls are significant. The launder model had a wall or baffle placed along its length to investigate its effect on trapping inclusions, results from the model showing particle distribution.