Quantification Of The Effects Of Dirty Upstream Pipe On Ultrasonic Meter Performance

The flow measurement performance of two commercially available ultrasonic flow meters has been evaluated under conditions in which there were various levels of contaminate-like coating applied to the inside wall of the pipe upstream of the meter.

The tests were performed under controlled conditions in the Southwest Research Institute Metering Research Facility (MRF) High Pressure Loop (HPL) and were designed to compare the baseline (clean pipe) flow measurement performance of each meter to the performance attained with various levels of coating on the upstream pipe.

The tests used 8-inch multipath ultrasonic meters and were performed at nominal operating conditions of 800 psia at 70°F using transmission-grade natural gas. The baseline performance of each meter was established by testing with honed pipe upstream. Meter tests were then repeated after the internal surface of the upstream pipe had been coated to simulate pipeline contamination. The meters were not coated for these tests.

Test Meters

Daniel and Instromet each provided an ultrasonic gas flow meter for use in this program. The meter path configurations are shown in Figure 1. Inspection of the meter parameters suggested that the meters were previously calibrated; however, no additional adjustments were made to the meters for the testing described in this article. It should be noted that other meter manufacturers were willing to provide meters for testing, but the Pipeline Research Council International (PRCI) committee members decided to focus on the Daniel and Instromet meters.

Figure 1: Meter Path Configurations.

Data Collection And Processing

The meters were tested with transmission-grade natural gas (i.e., approximately 95% methane gas mixture) furnished by the local utility company. Data were collected simultaneously from the test meter and from the MRF HPL critical flow nozzle bank, which served as the flow rate reference. The five binary-weighted critical flow nozzles in the HPL nozzle bank were previously calibrated in-situ, at different line pressures, against the HPL weigh tank system.

The volumetric flow rate, as well as the individual path status, flow velocity, and speed of sound data reported by each ultrasonic meter were digitally acquired during each flow point. Digital data were read from the test meter using the UNIFORM protocol for the Instromet meter and the MODBUS protocol for the Daniel meter.

Meter Installation

Each test meter was installed in the HPL header area with upstream piping that consisted of the vane portion of the Gallagher flow conditioner installed immediately downstream of a 12-inch by 8 inch reducing section, followed by a 5D pipe spool and a Gallagher perforated plate flow conditioner. An additional 10D of pipe was installed downstream of the flow conditioner prior to the two 10D sections of honed pipe specifically constructed for this testing. The upstream piping configuration was identical for testing performed with both meters; however, because of differences in the meter lengths, the downstream piping differed.

The testing required that the meters and the upstream piping be repeatedly removed and re-installed. To ensure repeatability of the setup, the upstream bore of the meter was carefully aligned with the upstream piping (through visual inspection) prior to installing the downstream piping. All of the piping was marked to maintain the same orientation from test to test.