Abstract

A real-time gas holdup sensor comprising a gas exclusion cell connected to a magnetic flowmeter has been described and tested with water in a previous communication. Gas holdup calculation based on this approach requires, in addition to the measurement of the exit flow velocity provided by the magnetic flowmeter, the sensor's dimensions and the actual discharge coefficient. This communication characterizes discharge coefficient and gas holdup measurement errors, for one of the original prototypes working with solutions of glycerine and sodium chloride and suspensions of particles having different size distribution and concentration. An underlying fairly unique relationship between discharge coefficient and Reynolds number was found for the fluid tested. This result significantly simplifies the application of the gas hold up sensor at the industrial scale as calibration can be carried out off-line using water instead of the actual slurry without sacrificing accuracy. Gas holdup measurement from the sensor prototype fully immersed in aerated suspensions of different size and concentration were found to be practically unbiased, very precise and insensitive to the fluid properties.