Particle Size Classification Through Aerodynamic Jet Vectoring

An experimental and numerical demonstration of a new, non-contact particle sorting technique called Aerodynamic Vectoring Particle Sorting (AVPS) is presented. AVPS uses secondary blowing and suction control flows to sharply turn a 2D, particle-laden jet. As the jet is turned, particles present in the flow experience a resultant force, dependent upon their size and due to the combined effects of pressure, inertia, and drag. Since the balance of these forces determines the particle’s trajectory, turning the flow leads to a separation of particles downstream. This simple, low-pressure-drop sorting technique classifies particles with less risk of damage or contamination than currently available sorting devices. AVPS is experimentally demonstrated using a rectangular air jet. Particle size are measured using the Shadowgraphy method. Numerical simulations are performed using the commercial CFD solver FLUENT to calculate the 2D turbulent vectored jet flow field using a RANS approach. Examination of the mean and the standard deviation of measured and computed particle trajectories is used to determine the range of particle sizes that can be effectively sorted using AVPS. Our results indicate that while vectoring can be achieved with smaller control flow rates when blowing and suction are used together, fluctuations in the velocity field are much smaller when suction only is used. Furthermore, we have demonstrated that the jet flow can be vectored 90 with pure suction and 180 using a new geometry that allows for modification of the blowing angle on the fly. Using pure suction, particles from 10–40 micron and 2.5 times the density of water have been sorted to an accuracy of 1.5 micrometers. Sorting of heavy particles such as these is accomplished at very low speeds, reducing the tendency of damage to the particles. Lighter particles are sorted at higher speeds. Also using pure suction, particles from 5–40 μm and 0.6 times the density of water were sorted to an accuracy of 6.6 μm.