Ultrasonic Flow Meter Research
Metering Research Facility
SwRI engineers inspect a meter installed in the HPL test section
Current Ultrasonic Flow Meter Research Goals and Topics of Study
Ultrasonic flow meter research at Southwest Research Institute (SwRI) has provided necessary test data to the American Gas Association (AGA) Transmission Measurement Committee (TMC) Task Group working to revise AGA Report No. 9 – Measurement of Gas by Multipath Ultrasonic Meters.
Recent ultrasonic research at SwRI has been conducted on the following topics:
- Effects of diameter mismatch on meter performance. Past MRF research focused on testing the effect of mismatched pipe and meter diameters within the range allowed by AGA Report No. 9. The results indicated that meters were not sensitive to a 1% diameter mismatch and that flange alignment was not critical.
- Effects of line pressure variations on meter bias. Depending on meter type, early research results showed that bias errors could be as large as 0.5% for an 800-psi difference in static pressure between calibration conditions and operating conditions. More recently, flow calibrations were performed on both ultrasonic meters with wetted transducers mounted in spool pieces, and clamp-on ultrasonic flow meters. The pressure effect on three ultrasonic meters with wetted transducers was between 0.1% and 0.2% over a pressure range from 400 psi to 1,000 psi. Calibrations of four clamp-on type ultrasonic meters gave different calibration curves at 200 psi and 400 psi than at 1,000 psi.
- Effects of low flow velocities on meter calibration. Recent tests investigated the effect of low flow velocity between 0.5 ft/s and 5 ft/s on ultrasonic meter calibration. Tests showed that at mean velocities less than 1 to 2 ft/s, the ultrasonic meter factor, path velocities, and sound speeds may diverge from values typical of operation at higher flow rates.
- Effects of skid outlet header flow arrangement on ultrasonic meter calibration. Tests of an 8-inch and a 6-inch diameter ultrasonic flow meter, installed both in and out of a meter skid, found that with proper flow conditioning, effects of a skid outlet header flow pattern were 0.2% or less in measurement error.