Johnson     1  

EXPERIMENTAL INVESTIGATION OF AIRFOIL THERMOSYPHONS FOR THERMAL MANAGEMENT OF NEXT GENERATION AIRCRAFT

Christina A. Johnson

Department of Mechanical and Aerospace Engineering, University of Virginia, Charlottesville VA

Advisor: Pamela Norris

Abstract Airfoil-shaped thermosyphons were designed, manufactured, and tested to investigate their performance (rate of heat transfer) for cooling applications. One thermosyphon had a cylindrical cavity and the other had a slot-shaped cavity. The thermosyphon material was copper, and the working fluid was deionized water. The fill volumes (as a percentage of the entire cavity volume) tested were 0%, 5%, and 20%. The condenser section was air-cooled in a wind tunnel with wind speeds of 100 mph and subjected to a range of evaporator temperatures. The rate of heat transfer with no working fluid was 156.5 W at the evaporator temperature design point of 315 F. The rate of heat transfer at the design point was the highest at the 5% fill volume at approximately 323.1 W. The highest thermosyphon performance was at 5% fill volume for all evaporator temperatures. As fluid is added to the thermosyphon, the surface temperature rises and the temperature distribution becomes more isothermal, explaining the observed increase in performance.

Introduction Boeing, the aerospace company and aircraft manufacturer, is continually exploring approaches to improve existing aircraft technology. One recent initiative is to upgrade future aircraft systems, such as cabin air conditioning, to run on electrical power. This will require replacing the current two-generator configuration with four more powerful electric generators. These added power sources will increase the amount of waste heat that must be expelled from the aircraft. At present, one small brick-shaped heat exchanger per jet engine is used for this heat transfer as shown in figure 1.

 Figure 1: OGV configuration.

This heat exchanger is positioned inside the jet engine and down-wind from flow straighteners, termed outlet guide vanes (OGVs), that straighten the bypass air flow to transfer heat into the airstream via forced convection. The geometry and size of this heat exchanger make this technology inadequate for Boeing’s intended upgrades for two reasons: limited ability to dissipate excess heat and decreased engine thrust if scaled to a larger size. Prior work has determined that the most viable solution has been determined to involve heat pipe technology, specifically wickless heat pipes called thermosyphons1. The internal operation of a thermosyphon is illustrated in figure 2. A thermosyphon is a closed vessel containing a fluid, and its primary function is to rapidly transfer heat from one location to another via latent heat of vaporization. At the heat source, the fluid vaporizes and moves rapidly toward the heat sink where it re-condenses and then is returned to the evaporator with the aid of gravity2.

Johnson     2  

 Figure 2: Cutaway of a thermosyphon.

Therefore, the OGVs, whose only current function is to straighten the flow of bypass air, will be hybridized to function as both flow straighteners and thermosyphons. To simulate operating conditions, the condenser section of the thermosyphons is air-cooled in a wind tunnel. The conceptual idea for testing is shown in figure 3. When the experiment has settled to steady state, the energy dissipated into the air stream is equivalent to the electrical power put into the thermosyphon. Therefore, for this experiment the thermosyphon performance is quantified by the rate of heat transfer to the airstream as measured by the electrical power input under steady state conditions.

 Figure 3: Conceptual idea for steady state testing.

Methods

The ultimate goal of this work is to transition from previous work dealing with cylindrical-shaped thermosyphons3 to develop a design for an airfoil-shaped thermosyphon that may be utilized as a hybrid flow straightener and heat exchanger and test its performance. Improvements to the design must be made until the rate of heat transfer is at a level such that an array of these thermosyphons will meet the heat dissipation needs of Boeing’s increased generator configuration. Construction Several iterations of finite element analysis (FEA) were performed to predict the wall displacements for an airfoil-shaped shell, with varied cavity shapes and sizes, subjected to high internal pressures. Due to the size of the OGVs (10.5” span, 2.5” chord, and 0.25” maximum thickness), the cavity is necessarily small in size. Adequate space in the cavity is crucial to avoid entrainment of the condensate into the rising vapor so that the thermosyphon functions properly. Both entrainment considerations and the results of the FEA give opposing constraints on the design. With this in mind, two different cavity shapes, a cylinder (strong structure but small cross-sectional area) and a slot (weaker structure but larger cross-sectional area), were chosen to be appropriate for testing. The airfoil cross-sections with cavities are shown in figure 4. These are NACA0010 airfoils with a 2.5” chord length and 0.25” maximum thickness.

 Figure 4: Airfoil cross-section cavity shapes.

 

Johnson     3  

The airfoil bodies were constructed from a solid piece of copper and the internal cavities were created using a method called wire electrical discharge machining (EDM). The end caps were also made out of copper and were attached to the airfoil body using a metal-filled epoxy. The structural integrity of the complete thermosyphon assemblies was tested using a hydrostatic pressure test where they were subjected to internal pressures of 100 psig. This was done to ensure that the thermosyphons would not fail if the cavity pressurized at high temperatures. Any large leaks were sealed using more of the same epoxy, and smaller leaks were sealed using a high vacuum sealant spray. Experiment To simulate operating conditions, i.e. the generator oil loop heating the base of the thermosyphon, the evaporator section is electrically heated with cartridge heaters with the condenser section exposed to the airflow in the test section of the wind tunnel. By using a temperature controller to set a chosen heater temperature and waiting until the experiment settled to steady state, the energy dissipated into the air stream is measured from the electrical power input to the cartridge heaters. The surface temperature of the thermosyphon is measured by thermocouples placed along the length of thermosyphon at the location of the cavity. The internal pressure is measured using a pressure transducer attached to the top end cap of the thermosyphon. An experimental set-up was created to allow for both the evacuation of the cavity and the thermosyphon charging while it is positioned in the wind tunnel test section. An image of one of the thermosyphons placed in the test-section is shown in figure 5.

 Figure 5: Slot-shaped cavity thermosyphon in wind

tunnel test section. To date, only the slot-shaped thermosyphon has been wind tunnel tested. The thermosyphon with the slot-shaped cavity was tested with three different fill volumes: 0%, 5%, and 20%. These fill volumes are given as a percentage of the total cavity volume. For each fill volume, the thermosyphon’s rate of heat transfer was measured at five different cartridge heater (evaporator) temperatures: 250F, 275F, 300F, 315F, and 325F, with the design point being 315 F.

Results The rates of heat transfer for the thermosyphon with the slot-shaped cavity at several evaporator temperatures are shown in figure 6 below. The 0% set of rate of heat transfer data ranges from 100 to 150 W. The 5% and 20% data sets have rates of heat transfer that are over twice as large as the rates for when the thermosyphon contains no working fluid.

Johnson     4  

 Figure 6: Rates of heat transfer for slot-shaped

cavity thermosyphon.  

The  surface  temperatures  measured  by   the   thermocouples   placed   over   the  cavity   and   along   the   length   of   the  thermosyphon   (with   the   smallest   vertical  position  being  closest  to  the  evaporator)  at  the  design  point  evaporator  temperature  of  315  F  are  plotted  in  figure  7  below.  The  5%  and   20%   data   sets   are   more   isothermal  than   the   0%   data   set   by   a   considerable  amount.   This   difference   illustrates   the  effect  that  the  addition  of  a  small  amount  of  working  fluid  has  on  heat  transfer.    

 Figure 7: Surface temperatures along length of

thermosyphon for slot-shaped cavity thermosyphon at the design point of 315 F.

Discussion

The rate of heat transfer of the thermosyphon when there is no water inside was by far the lowest and is considered to be

the base level of performance. That is to say, this is the poorest performance that may be expected. The 5% fill volume set of data had the best performance for all evaporator temperatures, though the difference between the 5% and 20% data sets was slight. The surface temperature distribution for the 0% fill volume is distinctly non-isothermal due to the fact that heat transfer up the thermosyphon is due strictly to conduction. In contrast to this, the surface temperature distributions at 5% and 20% are more constant, and the exhibition of somewhat isothermal surface temperatures implies typical thermosyphon behavior is taking place.

Conclusion Airfoil-shaped thermosyphons were designed, manufactured, and tested to investigate their performance (rate of heat transfer) for cooling applications. One thermosyphon had a cylindrical cavity and the other had a slot-shaped cavity. The thermosyphon material was copper, and the working fluid was deionized water. The fill volumes (as a percentage of the entire cavity volume) tested were 0%, 5%, and 20%. The condenser section was air-cooled in a wind tunnel with wind speeds of 100 mph and subjected to a range of evaporator temperatures. The rate of heat transfer with no working fluid was 156.5 W at the evaporator temperature design point of 315 F. The rate of heat transfer at the design point was the highest at the 5% fill volume at approximately 323.1 W. The highest thermosyphon performance was at 5% fill volume for all evaporator temperatures. As fluid is added to the thermosyphon, the surface temperature rises and temperature distribution becomes more isothermal, explaining the observed increase in performance.

Acknowledgment The author gratefully acknowledges financial support from The Boeing Company

Johnson     5  

and from the Department of Defense (DoD) through the National Defense Science & Engineering Graduate Fellowship (NDSEG) Program. The author also acknowledges academic and moral support from her advisors Pam Norris and Don Jordan.

References 1. Randolph   T.   L.,   2008,   “Thermosyphon  

Technology   for   Heat   Management   in  High-­‐Bypass   Jet   Engines   Aboard   Next  Generation   Boeing   737   Aircraft”,   M.S.  Thesis  University  of  Virginia.

2. A. Faghri, Heat Pipe Science and Technology, Taylor & Francis, Washington, DC, 1995.

3. DeCecchis, P. M., 2010, “Investigation  of  the   Effect   of   Fill   Volume   on   Heat  Transfer   from   Air-­‐Cooled   Low   Aspect  Ratio   Thermosyphons”, M.S. Thesis, University of Virginia