Performance of Nozzle, Venturi, and Orifice Meters Relative to Extrapolation Criteria

Abstract

Flow meter performance is described by the dimensionless numbers of discharge coefficient and Reynolds number. To achieve the best flow measurement uncertainty, meters are tested (calibrated) to determine the discharge coefficient behavior versus Reynolds number (magnitude and slope). Various meter designs have differing Reynolds number dependence. In many cases calibration laboratories can not achieve the Reynolds number at which the flow meter will operate. This deficiency is usually due to fluid properties (density and viscosity) at operating conditions being considerably different than those in a water-based calibration laboratory. Testing using fluids such as natural gas may increase the achievable Reynolds number but it is difficult to achieve the low uncertainty of the discharge coefficient possible in a water calibration due to the additional uncertainty of the expansion factor required with compressible fluids and the problems associated with gravimetric measurements of compressible fluids. In some power industry applications, operating Reynolds numbers may be an order of magnitude higher than can be achieved during calibration. Therefore, calibration data must be used to infer the discharge coefficient at operating conditions (Reynolds number), defining extrapolation. In Code tests, minimum flow measurement uncertainty is the objective and the uncertainty must be estimated. The largest uncertainty component in a flow measurement application usually is the discharge coefficient, which is dependent on the care of fabrication, the calibration data, and the extrapolation process. Measured discharge coefficients of Throat Tap Nozzles, Venturi meters Wall Tap Nozzles, and Orifice Meters are compared to predictive equations.