| Air classifiers are used in a variety of industries (e.g., electronics, food, chemical, petroleum) that need to mass-produce fine particles in a limited size range for controlling powder properties such as homogeneity, flowability, taste, texture, and surface area. The BLMT (Boundary Layer Momentum Transfer) classifier is a new type of centrifugal air classifier in which particle-carrying air is forced to flow radially inward between closely spaced, rotating, coaxial, annular disks. The disks accelerate the air tangentially, causing coarser particles to be removed by centrifugal force. Finer particles are carried with the air through the disks by drag force. The BLMT classifier differs from other forced vortex centrifugal classifiers in that it does not supply rotational momentum by collision of the particles with rotating vanes. Thus, it should be more effective in separating fragile particles without breaking them. A pilot-plant-scale prototype gave very good classification for a wide range of cut sizes with both glass spheres and non-ideal, industrial particles. We originally expected cut size to decrease with increasing disk rpm because of higher centrifugal force on the particles. However, over a certain range of rpm, cut size increased with increasing rpm. To explain this behavior, a simple, two-dimensional model for the flow between the disks was developed using computational fluid dynamics. The model shows that disk rotation causes flow near the disk surface to reverse. To compensate for the reversed flow, the velocity radially inward at the center between the disks increases, increasing the drag force, which may explain the observed increase in cut size with increasing rpm. As rpm increases further, the flow becomes turbulent according to the Renormalization Group k-ϵ model. Turbulence makes the velocity profile more uniform, reducing the reversed flow and thus the center radial velocity inward. More uniform velocity profiles are also in agreement with sharper cuts observed at higher rpm. |
