(384a) Effect of Particle Size Distribution On Particle Entrainment Velocity in a Pneumatic Conveying System

The minimum pickup velocity of solid particles is defined as the minimum velocity which can pick up and entrain a particle formerly at rest on the bottom of the pipe. The effect of particle size distribution on the minimum pickup velocity has not been fully explored in literature although it holds paramount importance in the pneumatic conveying of solid particulates. The present work investigates the effects of binary size distribution on the minimum pickup velocity in a pneumatic conveying system at different finer particle fractions. The minimum pickup velocities of three binary mixtures of 50, 200, and 400 ìm glass beads are investigated. The experiments were conducted in a 2.05 m long horizontal acrylic pipe with an internal diameter of 0.0155m. The pipe was retrofitted with a removable pipe section where compressed air with flow rates ranging from 20 L/min to 62 L/min flowed through the upper hemispherical section to entrain the particles from a loaded cavity in the lower hemispherical section. It was observed that the minimum pickup velocity of the binary mixture increased either with increasing composition of smaller and cohesive particles or with increasing composition of larger and heavier particles, which is similar to the trends observed in the minimum fluidization velocity of a fluidized bed. The trends in the pick-up velocity as a function of the particle size distribution can be classified into different zones similar to the Geldart's particle classification. Empirical correlations to predict the minimum pick-up velocity of binary mixtures were developed from an existing correlation for a mono-modal size distribution using either arithmetic or sauter mean diameters as the effective diameter of the mixture. The correlations were expressed in terms of the dimensionless Archimedes and particle Reynolds numbers. The results indicates that the use of the sauter mean diameter in the correlation results in reasonable predictions for a majority of binary mixtures, even for those which exhibit high non-linearity in their behavior.