A New Use for an Old Standby

A New Use for an Old Standby, SwRI researchers are testing Coriolis flow meters for natural gas industry applications

Terry Grimley     image of PDF button

photo of Grimley

Terry Grimley is manager of the Flow Measurement Section in the Fluids Engineering Department of SwRI’s Mechanical Engineering Division. Grimley, whose specialties include heat transfer, fluid mechanics and two-phase flow, oversees operation of SwRI’s Metering Research Facility.

photo of Coriolis meter

The flow capacity of a Coriolis meter allows it to be installed in line sizes larger than the actual meter size. Here a 3-inch diameter meter is being tested downstream of a 6-inch diameter valve, filter and reducer; this installation is typical of a field installation.

image of two tubes in parallel

As the drive system vibrates the flow tube, the U-tube configuration creates a difference in the direction of the Coriolis forces from the tube inlet to the tube outlet, creating a twist in the flow tube that is measured as a phase difference. By using two tubes in parallel, the meter can be balanced.

image of a top view of a dual-tube Coriolis meter

A top view of a dual-tube Coriolis meter shows an exaggerated twist in the flow tubes resulting from the Coriolis force generated by flow through the tubes in combination with the electromechanically driven tube vibration. When there is no flow, the tubes vibrate without being twisted.

image of a high-pressure 8-inch-diameter Coriolis meter

This high-pressure 8-inch-diameter Coriolis meter has capacity sufficient to allow it to be used in many gas transmission pipeline applications. The elimination of secondary instrumentation and surrounding piping requirements can make the installation cost-competitive with other measurement technologies.

image of a 3-inch meter

To prove insensitivity to adverse effects related to the upstream and downstream piping configuration, this 3-inch meter is installed downstream of a 6-inch by 3-inch diameter reducer and two 6-inch diameter elbows that are oriented such that they generate flow swirl at the meter inlet.

Accurate measurement of any fluid commodity is important because at some point money will change hands based on the quantity measured. In the oil and gas industry, the value and quantity of the products exchanged are enormous. According to the Energy Information Agency, natural gas production in the United States alone is more than 80 billion cubic feet per day.

Although fluid quantity measurement can be as simple as determining the height of a liquid in a cylindrical tank, gas measurements typically involve measuring flowing fluid quantity as a function of time to compute a total quantity. For decades, the most common flow meter in industrial operations was the orifice meter, which uses a plate with a hole smaller than the pipe diameter to create a differential pressure that is measured and used to compute the flow rate. Numerous other types of flow meters rely on measuring differential pressure or the rotation speed of a turbine wheel, or more recently, the measure of the transit time of ultrasonic energy as it crosses the flow stream. These meters do not directly measure the mass flow rate of the fluid, but must rely on additional physical measurements and fluid property correlations to determine the total mass (or “standard” volume) of fluid that is used for monetary transactions. This limitation has led the natural gas industry to show new interest in Coriolis flow meters because of their ability to determine more directly the mass flow rate of a fluid.

The conceptual basis for Coriolis meters is not new. A paper published in 1835 by French engineer Gaspard-Gustave de Coriolis discussed forces that exist in a rotating frame of reference. One component of these forces eventually became known as the Coriolis force.

The first patents for the concept of a Coriolis flow meter were filed in the 1950s, and by the 1970s they were commercially introduced for use with liquids and other fluids. However, it wasn’t until the 1990s that developments in digital electronics made it possible for Coriolis meters to function properly in natural gas applications. Digital electronics provide the increased sensitivity needed to measure the signals produced as a result of gas mass flow rates which are a fraction of that for liquid flowing through the same size flow tube. Today, partially based on research by Southwest Research Institute (SwRI) engineers at the Institute’s Metering Research Facility (MRF), Coriolis meters are gaining wide acceptance within the natural gas industry.

How they work

Coriolis flow meters rely on the inertial force imparted on a flow tube that results from electromechanically oscillating the flow tube at its natural frequency. Using a U-shaped tube for the flow path, the Coriolis forces act in opposite directions at the inlet and outlet of the flow tube, causing the tube to twist slightly. Sensors located near each end of the tube measure the twist via a change in the phase relationship of the tube motion. The measured time shift is proportional to the mass flow through the sensor. That measurement is independent of the type of fluid flowing through the meter tubes. In addition, the change in the natural frequency of the tube/fluid combination can be used to measure the fluid density. For natural gas measurements, the resulting frequency shift typically is not sufficient to provide accurate gas density measurement. Instead, the tube frequency becomes a diagnostic measurement for the operational condition of the meter.

Early testing of Coriolis flow meters at SwRI identified stability issues related to flow noise (turbulence and other velocity perturbations), which is inherent in any fluid flow but can be more significant in Coriolis flow meters because the flow noise can cause a meter response of a magnitude similar to the induced Coriolis force. Meter manufacturers overcame these problems by using the increased signal processing power in the meters’ electronics. Additional research to characterize the performance of meters from multiple manufacturers in a variety of piping configurations and over a wide range of operating conditions provided some of the base performance data that the industry needed to develop a generally accepted recommended practice for Coriolis meters. The American Gas Association (AGA) first published Report Number 11, “Measurement of Natural Gas by Coriolis Meters,” in 2003. It was revised earlier this year to reflect the evolution of Coriolis technology for natural gas applications.

Advantages of Coriolis meters

Most non-Coriolis flow meters require that a significant length of straight pipe be installed upstream and downstream from the meter to assure non-turbulent flow. Costs associated with installing this piping, or to install flow conditioners — devices placed in the flow path upstream of the meter that reduce the minimum length of straight pipe required — can be a large percentage of the overall cost of a meter station. Most Coriolis flow meters do not require this, making them an attractive cost-saving option.

Coriolis meters also do not require high-accuracy auxiliary pressure, temperature and flow stream composition measurements. While they must compensate for changes in tube stiffness resulting from changes in flow stream temperature and pressure, the accuracy requirements of those measurements are significantly lower than for volumetric-type flow measurement devices. Coriolis meters have internal temperature sensors that assess the tube temperature for the purpose of compensation. Depending on the meter size, the effect of flow-stream pressure can range from insignificant for small meters (2-inch diameter) to 0.1 percent per 100 psi change in line pressure for larger meters (8-inch diameter). Eliminating auxiliary measurements not only reduces the initial facility cost, but more importantly it also eliminates the number of auxiliary devices requiring periodic maintenance and calibration.

The main limitation of Coriolis flow meters is the pressure drop required to reach the manufacturer-specified flow capacity, so the meters must be installed where this is not an issue. Even with multiple flow tubes, the total flow area of a Coriolis meter is normally much smaller than the flow area of the surrounding pipe. Reducing the flow area provides increased velocity in the meter’s flow tubes that improves measurement sensitivity, but this also creates a pressure drop greater than other flow measurement methods because of the combination of increased velocity and the bent-tube geometry.

For a commercial client, SwRI researchers used the MRF to assess the performance of a meter installed downstream of some common field elements, such as filters, valves and elbows. The results allowed the company to develop standard installation practices, determine practical capacity limits and generate calibration requirements for installing Coriolis meter technology.

Evaluating the dynamics of larger meters

Until a few years ago, Coriolis meters for natural gas applications typically were limited to relatively small line diameters, generally less than 4 inches. Recently, however, the industry has introduced meters with flange diameters of 8 inches and larger based on better understanding of the fluid dynamics associated with Coriolis flow meters and the availability of the high-energy magnets used in the meters. As larger meters were being developed, the MRF served as a test bed for both manufacturers and end-users of Coriolis meters.

Verifying the accuracy performance of larger Coriolis flow meters is of particular interest to end-users, because even a small margin of error can translate into large amounts of money. For example, a 0.1-percent error in an 8-inch diameter Coriolis meter operating at 50-percent capacity with natural gas at a line pressure of 1,000 psia equals about $200,000 per year (assuming gas costs of $4 per thousand standard cubic feet).

Other applications

Besides natural gas, Coriolis flow meters are also commonly used to measure supercritical ethylene, which has a significantly higher monetary value than natural gas. Ethylene, one of the highest-volume chemicals produced, is primarily used for producing plastics. It is commonly stored at a pressure and temperature beyond the boundary where a fluid behaves as strictly a gas or a liquid. While orifice meters are commonly used for this application, Coriolis flow meters provide a less expensive system with more accurate measurement. Flow testing with natural gas can be used to verify the flow meter’s performance at pressures similar to those common in supercritical ethylene applications.

Whether for the natural gas industry or other markets, SwRI will continue to help its clients further this important metering technology.

Questions about this article? Contact Grimley at (210) 522-2353 or terry.grimley@swri.org.

image of the Metering Research Facility at SwRI

The Metering Research Facility at SwRI consists of two separate closed-loop natural gas flow loops used for flow measurement research projects as well as flow meter calibration. The flow loops allow the simulation of end-use operating conditions because of the ability to set and control the pressure, temperature and flow rate. The reference flow rate uncertainty is roughly 0.25 percent, and the repeatability is considerably lower, making the facility well-suited for performing meter calibrations and for studying the subtle changes in meter performance resulting from changes in the meter installation configuration, operating conditions and other parameters. The MRF was originally built in the early 1990s through an industry project sponsored by the Gas Research Institute (now the Gas Technology Institute) and was later purchased by SwRI. The facility has been providing flow measurement research and flow meter calibrations for more than 20 years.

Benefiting government, industry and the public through innovative science and technology
Southwest Research Institute® (SwRI®), headquartered in San Antonio, Texas, is a multidisciplinary, independent, nonprofit, applied engineering and physical sciences research and development organization with 9 technical divisions.