An empirical method to estimate the viscosity of mineral oil by means of ultrasonic attenuation.

This paper presents an empirical method for measuring the viscosity of mineral oil. In a built-in pipeline application, conventional ultrasonic methods using shear reflectance or rheological and acoustical phenomena may fail because of attenuated shear wave propagation and an unpredictable spreading loss caused by protective housings and comparable main flows. The empirical method utilizing longitudinal waves eliminates the unknown spreading loss from attenuation measurements on the object fluid by removing the normalized spreading loss per focal length with the measurement of a reference fluid of a known acoustic absorption coefficient. The ultrasonic attenuation of fresh water as the reference fluid and mineral oil as the object fluid were measured along with the sound speed and effective frequency. The empirical equation for the spreading loss in the reference fluid is determined by high-order polynomial fitting. To estimate the shear viscosity of the mineral oil, a linear fit is applied to the total loss difference between the two fluids, whose slope (the absorption coefficient) is combined with an assumed shear-to-volume viscosity relation. The empirical method predicted the viscosities of two types of the mineral oil with a maximum statistical uncertainty of 8.8% and a maximum systematic error of 12.5% compared with directly measured viscosity using a glass-type viscometer. The validity of this method was examined by comparison with the results from theoretical far-field spreading.