Computer model to simulate the random behaviour of particles in a thermal-spray jet

Thermal-spray processes involve many sources of uncertainty: the feedstock powder has a wide particle size distribution, the powder particles leave the powder injection nozzle in a variety of angles and velocities and enter a thermal jet that has enormous temperature and velocity gradients. Particles of various sizes therefore travel through the gas jet in many different trajectories and impact upon the substrate in a wide range of velocities, temperatures and angles. The process inevitably has a large non-deterministic and random element. Since coating performance is crucially dependent on the nature of the particles on impact, this degree of randomness must have a major influence on the effectiveness of industrial thermal spraying. Simulations based on the behaviour of a single particle cannot therefore provide accurate predictions of the deposition process. In this paper, a probability approach is adopted using the Monte-Carlo method and applied to simulate the behaviour of particles in a plasma jet. The distributions of particle temperature, velocity and trajectory are directly linked to the particle size distribution and powder injection parameters. This approach enables the prediction of the mean value and standard deviation of the particle temperature and velocity together with the deposition impact position. A unique attribute of the model is its ability to calculate the standard deviations of the parameters, which provides information on the variability of coating performance. The research shows that there are a limited number of process parameters and particle properties that are expected to govern the performance of the as-deposited coating. Charts showing the effect of process combinations on particle properties are used to simplify the data and are proposed as a basis for industrial process control.