Vortex Shedder Fluid Flow Sensor

Abstract

This paper was motivated by the possibility of extracting from a vortex-shedding strut, in addition to flow velocity V, information on fluid density rho or temperature T, and combining them to obtain mass flowrate. Shedder shapes were diamond and bluff polygon. These shapes are compared as vortex shedders in flowing air or water. V is obtained from the shedding frequency f. In water, V ranged from 0.5 to 4 m/s and, in air, from 0.3 to 15 m/s. Clamp-on ultrasonic transducers generated and, on the diagonally opposite side of the pipe, received the beam that obliquely traversed the wake of the shedder. A continuous-wave transmission across the fluid was modulated by vortices passing through the beam. The modulation frequency yielded f. In air, the bluff polygon yielded f over a 50 : 1 flow range, which was better than the diamond's flow range of 20 : 1. Whether the shedder was a diamond or a bluff polygon, and the fluid air or water, f correlated approximately linearly with the flow velocity V. Using one path of an ultrasonic tag clamp-on flowmeter system, the measured vortex-shedding frequencies were found to be in reasonable agreement with computational-fluid-dynamic predictions for diamond and for bluff-polygon struts. Collectively, the pipe Reynolds number (Re) range was 1000-200 000. With both shedders, operation was demonstrated in laminar- and turbulent-flow regimes. In water flow tests, rotating the diamond (aspect ratio=3) through 90deg about its axis, from broadside to airfoil, diminished the Strouhal number by 17%. When the diamond shedder was tested as a torsional density sensor in flowing air or water, no torsional transit time effect of V was observed, confirming for the first time a 1989 prediction. The negative result in flowing water implies that there were no attached bubbles or microbubbles