Operation of ultrasonic flow meters at conditions different than their calibration: Fid-Bau Portal

Operation of ultrasonic flow meters at conditions different than their calibration

Freund, W. / Zanker, K. / Goodson, D. / Hall, J.E. / Jamieson, A.W.

Currently, calibration of an ultrasonic flow meter for natural gas measurement is conducted under the conditions available at the flow calibration facility. Since almost all of these facilities utilize natural gas flowing in a pipeline, it is usually not possible to vary parameters such as temperature, pressure and gas composition, each of which affect the speed of sound. When the ultrasonic meters are then used in applications where these parameters are different from their calibration values, does the calibration still apply? In order to quantify the effect of changes in these parameters on the calibration of ultrasonic meters, a series of carefully controlled calibration have been performed. The first experiment consisted of calibrating a 200 mm (8 inch) and two 300 mm (12 inch) ultrasonic meters in the High Pressure Loop at Southwest Research Institute (SwRi) utilizing natural gas at 2.8 MPa (400 psi). As additional references, 200 mm and 300 mm turbine meters were also included in the calibration loop. The fluid was then changed to nitrogen, providing a 16 % change in the speed of sound, equivalent to a natural gas pressure of 4.6 MPa (667 psi). As a further test of potential pressure effects on an ultrasonic meter's calibration, a series of static measurements of the speed of sound was performed on the 300 mm meter at pressures between 1.4 MPa (200 psi) and 7 MPa (1000 psi) using nitrogen. The measured change in the speed of sound over this presswe range was within 0.03 % of the calculated value. In addition, further calibrations were run with changes in the speed of sound introduced through changes in the temperature of the fluid. Natural gas calibrations were performed at 21 % (70 deg F ) and 10 deg C (50 deg F) and nitrogen calibrations at 21 deg C (70 deg F) and 32 deg C (90 deg F). For each series of calibrations, the average calibration curves are compared to detmine the effed of the change. Within the expeded reproducibility of the facility and the meters, the calibration of the ultrasonic meters was insensitive to change in speed of sound, temperature and pressure. Evidently, the observed shifts in calibration on changing from natural gas to nitrogen are due to the equations of state used for these gases. These test results justify be procedure of calibration of an ultrasonic meter at one set of conditions and utilizing it under another, including using different gases.