Home | Technical Information | Wiki | Outreach | Media | Sponsors | Links | Contact Us
Passive Separation of Two-Phase Flows using Conduit Geometry
The study of two-phase flows is essential to fluids management in both spacecraft and micro scale terrestrial applications. In spacecraft such as the new Crew Exploration Vehicle, numerous applications for fluids processing of human-rated flight systems necessitate the study of this flow. In a low-g environment, fluids such as water, fuels, and coolants are often controlled by the inescapable effects of surface tension. The goal of this experiment is to obtain reliable empirical data that is valuable to NASA and the scientific community at large by implementing a new, elegant and repeatable experiment to study two-phase flow. This research will utilize specific fluid properties, flow properties, wettability characteristics, and conduit geometry as a method of passive flow separation. Optimal parameters will be determined analytically and verified with experimental data. To provide quantitative data regarding the success of the separation of the liquid and gas phases, the experimental apparatus will contain high speed video cameras and flow meters. All of the components of the apparatus will operate in a closed liquid loop and be double contained for safe operation. Data collected aboard NASA’s low-g aircraft will be analyzed, compared to other published work, and published if merited offering insight to the design of certain spacecraft fluids systems.
1. Passive system for separation of a bubbly flow, compared to heavy, expensive centrifuges (current system) that require power to operate.
2. Experimentally determine the minimum and maximum flow rates for successful separation in this specific conduit.
3. Quantitatively compare the effect two different interior corner angles have on the flow rate of the exiting fluid.
The core idea driving our experiment is the utilization of the fact that a flow can be dominated by surface forces (such as capillary action) under the right circumstances. Surface forces normally are dwarfed by body forces, typically gravity; with exception given to two cases:
• The flow is happening in low gravity and in the absence of other body forces.
• The flow has small length scales.
The non-dimensional Bond number dictates whether the flow is driven by body forces (such as gravity) or capillary forces (surface tension). It is may be written as:
where ρ is the liquid density, R is a characteristic length scale, and σ is the surface tension. This number is a ratio of body forces (g may be replaced or superimposed with other acceleration) to surface tension and generally indicates a surface tension dominated flow when sufficiently smaller then one.
Note that for a given fluid, two parameters control the Bond number: an appropriate length scale (R) and gravity. For example, the bond number of water inside a 0.1 mm glass tube on earth is 0.0013, comparable to the bond number of a cup of water aboard a space shuttle.
Low bond number flow can be easily achieved in the presence of terrestrial gravity without requiring the length scale to be absurdly small, this is seen (for example) in the transport of sap in trees. However, it is very difficult to produce and video tape bubbles in sufficiently small flow, hence the need for reduced gravity.
The mathematics involved with this project hardly stops with the Bond number. Mark Weislogel's research, primarly dealing with low Bond number flows inside interior corners, gives us ways to understand many aspects of the flows and laid foundations for the design of the two carefully designed conduits sections that form the core of our experiment.
Some progress throughout working with the theory portion of the project:
Calculation of the area and perimeter of a teardrop section, these expressions are in turn used to calculate a crucial number called capillary rise.
A screenshot of a spreadsheet used to figure out some of the aspects of the flows.
A screenshot of a spreadsheet used to help conduit selection and characterization.
Graphs of steady state flow inside 10° and 15° isosceles triangle conduits.
The following design was selected:
1. The working liquid is 5 centistoke silicone oil, selected for high viscosity (slower flow is easier to capture on video), low Bond numbers, and safety.
2. Two transparent polycarbonate test conduits of teardrop cross section, one with a 15° corner and the other with 10°.
3. Peristaltic pump to deliver continuous, closed loop fluid flow through test conduits and secondary separation chamber.
Our test equipment is, in (crude) summary, a fluid circuit and some electronics inside of an aluminum and polycabonate case. Per NASA's requirements, the apparatus must be able to stand 9 g in any direction with a minimum safety factor of two.
A simplified overview of the apparatus.
The video system.
The complete fluid circuit.
Here are some early renderings of the apparatus:
Here are some photos showing the actual developement of the apparatus (a few of these were taken in Houston):
Data was logged in video form, a total of 128 tests were done in 64 parabolas (microgravity sessions) of about 23 seconds each. Several tests were done without bubbles, the resulting flows agree very closely with theoretic prediction (at least where theory holds). The example below shows a frame picked from a bubble free test alongside a graph showing the digitized curve with a predicted profile.
Experiments with bubbles in the flow generally showed migration of the bubbles toward the surface. The bubbles did not usually burst or coalesce, except for a few tests, apparently this is a sensitive phenomenon. It seemed like relatively slow flows with less then usual liquid volumes gave the best results, allowing the bubbles much time to dry out and burst. Excessive squeezing of the bubbles in the lower parts of the interior corners seemed to be detrimental, so that the cell with the sharper (10°) corner did poorly compared to the other (15°); this is entirely contrary to what was expected.
Example of complete separation.
Example of no separation.
Another interesting test result is that all of us can stand having gravity switch from 0 g to 1.8 g 32 times over about an hour without puking. Compares well with students from Michigan, Cornell, and Witchita.
Concus, Paul, et. al. "Measurement of Critical Contact Angle in a Microgravity Experiment." Experiments in Fluids 28 (2000): 197-205.
Dreyer, Michael E., Jens Gerstmann, Michael Stange, Uwe Rosendahl, Gerrit Wölk, and Hans J. Rath. "Capillary Effects under Low Gravity." Space Forum 3 2(1998): 87-136.
McQuillen, J., C. Colin and J. Fabre. "Ground-based Gas-Liquid Flow Research in Microgravity Conditions: State of Knowledge." Space Forum 3 (1998): 165-203.
Munson, Bruce R., Donald F. Young, and Theodore H. Okiishi. Fundamentals of Fluid Mechanics. 5th ed. New York: Wiley, 2006.
Pais, Salvatore Cezar. "Bubble Generation in a Continuous Liquid Flow Under Reduced Gravity Conditions." NASA CR-1999-209170 (1999).
Weislogel, Mark M. "Fluid Interface Phenomena in a Low-Gravity Environment: Recent Results from Drop Tower Experimentation." Space Forum 3 (1998): 59-86.
Weislogel, Mark M. "Some Analytical Tools for Fluids Management in Space: Isothermal Capillary Flows Along Interior Corners." Adv. Space Res. 32 (2003): 163-170.
White, Frank M. Fluid Mechanics. 3rd ed. New York: McGraw-Hill, 1994.
