If a turbine flowmeter is used to measure media of different viscosities, the accuracy of the turbine flowmeter is affected by the changes in the viscosity of the measured medium and its motion. It is not possible to obtain consistent and predictable calibration results from variable viscosity experiments using volumetric flow rate and instrument coefficients. Using dimensional analysis to derive Reynolds number and Strouhal number as dimensionless parameters for describing the performance of turbine flow meters.
By changing the volume concentration of the propylene glycol water solution, five media with different kinematic viscosities were obtained and used to calibrate a turbine flowmeter. The comparison results show that the calibration curves at different viscosities exhibit separation in the region where the Reynolds number is less than 7400, and the maximum difference in calibration data is 0 9%. As the Reynolds number increases, the influence of bearing blockage in the instrument coefficient relatively decreases, and the calibration curve cluster tends to converge from dispersion, with a difference of calibration data less than 0 1%.
The sudden change in resistance during the turbulent transition of the flow boundary layer on the blade surface leads to a hump in the calibration curve, and the lower the kinematic viscosity, the smoother the hump tends to be. The static resistance in bearing blockage is the main reason for the difference in instrument coefficients at the same Reynolds number, which increases as the Reynolds number decreases,
Therefore, when there is a significant difference in the kinematic viscosity between the calibration medium and the working medium, the turbine flowmeter should avoid operating in the low Reynolds number region.
