Advances in Thermal Dispersion Mass Flow Meters - Part 2 Installation and Flow Calibration
Part 1 of this two-part series describes the principle of operation of a new microprocessor-based thermal dispersion mass flow meter that uses four temperature sensing elements in its flow sensor and has a dry velocity sensor. Part 1 shows how this unique system provides automatic management of changes in gas selection, gas temperature, gas pressure, and outside temperature. This Part 2 describes advances in the installation and flow calibration of thermal dispersion mass flow meters, including a built-in flow conditioner for in-line flow meters that greatly reduces upstream straight-pipe requirements and a high performance flow calibration facility that accommodates a range of gases, temperatures, and pressures.
For over forty years industrial-grade thermal dispersion mass flow meters have solved the wide range of general industry’s needs for directly measuring the mass flow rate of air, natural gas, and other gases in pipes and ducts in rugged industrial applications. Thermal dispersion mass flow meters measure the heat convected into the boundary layer of the gas flowing over the surface of a heated velocity sensor immersed in the flow. Since it is the mass-bearing molecules of the gas that carry away the heat, thermal dispersion mass flow meters directly measure mass flow rate. In acknowledgement of the wide acceptance of thermal dispersion mass flow meters for industrial applications, the American Society of Mechanical Engineers (ASME) has published a new national standard for this kind of flow meter .
Figure 1 shows the two primary configurations of thermal dispersion mass flow meters: in-line and insertion. The flow sensor of both configurations has a heated velocity sensor and a separate temperature sensor that are mounted side-by-side in rugged stainless-steel tubular sheaths and immersed in the flow stream. This kind of robust construction is responsible for transforming the predecessor lightduty thermal anemometer  into an industrial-grade flow meter.
Traditional thermal dispersion mass flow meters have two temperature sensing platinum RTD elements, one electrically self-heated element in the tip of the velocity sensor and one in the tip of the temperature sensor that measures the gas temperature T. Typically both elements are potted into their sheaths with a potting compound, such as ceramic cement or epoxy. The sensor drive in the transmitter electronics delivers an electrical current I1 to the velocity sensor such that it is self-heated to an average temperature T1 that is elevated above the gas temperature. In the constant temperature differential mode of operation, the flow sensor drive maintains at a constant value the temperature difference ΔT = T1 –T. The output signal is the electrical power W supplied to the heated velocity sensor that is required to keep ΔT constant.
Stem conduction and skin resistance are the two phenomena in traditional thermal dispersion mass flow meters most responsible for degrading accuracy and stability. Part 1 of this two-part series describes an advanced flow sensor that facilitates solving these two problems. This new flow sensor employs a total of four platinum RTD temperature sensing elements, instead of the traditional two. The velocity sensor in this “four-temperature” flow sensor has the heated T1 temperature sensing element in its tip as before, but now has a second element in its stem. The temperature sensor has the temperature sensing element in its tip as before, but now also has a second element in its stem. The T1 element is a wire-wound platinum RTD and the remaining three elements are thin-film platinum RTDs.
The two temperature sensing elements in the velocity sensor together act as a heat flux gauge that measures the fraction of heat conducted down the stem of the velocity sensor. The two elements in the temperature sensor perform the same function for the temperature sensor. This is how the four-temperature flow sensor facilitates correction for stem conduction. Additionally, the construction of the velocity sensor avoids altogether the use of any potting materials between the T1 element and the internal surface of the sheath by means of tightly fitting, as in swaging or press fitting, the cylindrical wire-wound T1 element into the sheath. Such velocity sensors are known as “dry” sensors. Because it has no potting materials and because the mating parts have the same coefficient of thermal expansion, the skin resistance of the velocity sensor is minimized and is stable. This is how the four-temperature flow sensor facilitates solving the skin resistance problem.
Part 1 describes the principles of operation embedded in the microprocessor based electronics that operates the physical layer with the four-temperature flow sensor incorporating a dry velocity sensor. This advanced system provides automatic management of changes in gas selection, gas temperature, gas pressure, and outside temperature, as well as automatic correction for stem conduction and skin resistance. This Part 2 describes some advances in the installation and flow calibration of this four-temperature microprocessor-based system, but has general applicability to all thermal dispersion mass flow meters.