A Study on Flow Regime Identification in Microgravity Gas-Liquid Flows Using a Capacitance Sensor

by Lowe, Devin C.

Microgravity two-phase flow is becoming increasingly important in space-based technologies. Research into microgravity two-phase flows, in addition to providing data on the behavior of such flows in a near weightless environment, serves to further research into normal gravity two-phase flow by changing one of the major variables affecting the flow, namely gravity. Gravitational force plays a major role in two-phase flows, and has a great influence on the flow regimes. Depending on the liquid and gas flow rates, the fluid properties, the tube size, and the gravity field, the flow will assume one of the following flow regimes: bubbly flow, where the gas flows as discrete bubbles in the liquid medium; slug flow, where the gas is contained in long Taylor bubbles separated by liquid slugs; transitional flow, where the gas flows in the center of the tube and the liquid generally flows along the circumference of the tube with frequent bridging; and annular flow, where the liquid forms an annulus around the circumference of the tube and the gas flows uninterrupted through the core. One of the important parameters of two-phase flow is the void fraction, which is the ratio of the volume of gas flowing in the tube to the total flow volume. This thesis investigates the use of void fraction signals to determine a more objective method of micro gravity flow regime identification than what is possible through examining video images.

The first objective of this study was to design, build, and test a void fraction sensor for measuring micro gravity two-phase flows. To this end, a capacitance type sensor was built. It has a two tube diameter sensing length as a compromise between obtaining a local measurement and the greater sensitivity achieved with a longer sensor. Calibration of the sensor was done using quick closing valves.

Void fraction data was collected for water-air two-phase flow in a 3/8 in. (9.525 mm) tube on board the NASA Lewis DC-9 microgravity aircraft. Void fraction data was compared against that of Elkow (1995) and Bousman (1995), and was found to be in excellent agreement.

The microgravity void fraction data was used in the form of probability density functions to determine a more objective method of identifying the flow regimes and their transitions than what is possible from examining video images. This, in turn, minimizes the errors and complexity associated with the use of subjective methods such as two dimensional video images.

Once the flow regimes were identified, the data was compared against microgravity flow regime transition models in the literature. The Drift-Flux model, originally developed by Zuber and Findlay (1965) and later modified for micro gravity flows by Bousman (1995), separated the bubbly and slug flow regimes well when evaluated with the transition void fraction determined from this data set. The slug-to-annular transition model suggested by Bousman (1995) resulted in a transition line which fell in the transitional flow region. The Weber number model for the transition from slug to transitional and from transitional to annular flow, originally suggested by Zhao and Rezkallah (1993), separated the slug and transitional flows perfectly. The transitional-to-annular boundary was slightly over-predicted. It was concluded that the Weber number model is of more use than the slug-to-annular transition model suggested by Bousman (1995) since it delineates both the slug-to-transitional and transitional-to-annular flow regime boundaries.