0
Research Papers

Development and Characterization of an Air-Cooled Loop Heat Pipe With a Wick in the Condenser

[+] Author and Article Information
Evelyn N. Wang

e-mail: enwang@mit.edu
Department of Mechanical Engineering,
Massachusetts Institute of Technology,
77 Massachusetts Avenue,
Cambridge, MA 02139

1Present address: Combustion Research Facility, Sandia, 7011 East Avenue, Livermore, CA 94550.

2Corresponding author.

Manuscript received February 2, 2013; final manuscript received April 27, 2013; published online October 25, 2013. Assoc. Editor: Hongbin Ma.

J. Thermal Sci. Eng. Appl 6(1), 011010 (Oct 25, 2013) (9 pages) Paper No: TSEA-13-1024; doi: 10.1115/1.4025049 History: Received February 02, 2013; Revised April 27, 2013

Thermal management of modern electronics is rapidly becoming a critical bottleneck of their computational performance. Air-cooled heat sinks offer ease and flexibility in installation and are currently the most widely used solution for cooling electronics. We report the characterization of a novel loop heat pipe (LHP) with a wick in the condenser, developed for the integration into an air-cooled heat sink. The evaporator and condenser are planar (102 mm × 102 mm footprint) and allow for potential integration of multiple, stacked condensers. The condenser wick is used to separate the liquid and vapor phases during condensation by capillary menisci and enables the use of multiple condensers with equal condensation behavior and performance. In this paper, the thermal–fluidic cycle is outlined, and the requirements to generate capillary pressure in the condenser are discussed. The LHP design to fulfill the requirements is then described, and the experimental characterization of a single-condenser version of the LHP is reported. The thermal performance was dependent on the fan speed and the volume of the working fluid; a thermal resistance of 0.177  °C/W was demonstrated at a heat load of 200 W, fan speed of 5000 rpm and fluid volume of 67 mL. When the LHP was filled with the working fluid to the proper volume, capillary pressure in the condenser was confirmed for all heat loads tested, with a maximum of 3.5 kPa at 200 W. When overfilled with the working fluid, the condenser was flooded with liquid, preventing the formation of capillary pressure and significantly increasing the LHP thermal resistance. This study provides the detailed thermal–fluidic considerations needed to generate capillary pressure in the condenser for controlling the condensation behavior and serves as the basis of developing multiple-condenser LHPs with low thermal resistance.

FIGURES IN THIS ARTICLE
<>
Copyright © 2014 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 3

Schematic representation of the integrated single-condenser LHP, showing the key design features of the evaporator and condenser through cross-sections perpendicular and parallel, respectively, to the plane of the evaporator base. The evaporator utilizes a three-layer wick, consisting of a high thermal conductivity and capillary pressure wick at the base (wick 1), a low thermal conductivity wick (wick 2), and a high permeability wick (wick 3). The condenser consists of three regions: the central condensation space, the subcooler, and the liquid return channel.

Grahic Jump Location
Fig. 2

Thermal-fluidic cycle of the LHP. A) Schematic of the heat pipe cycle. The hatched areas in the condenser and evaporator represent the wicks. The compensation chamber and vapor channels are interconnected network of channels at the top and bottom of the evaporator, respectively. B) Schematic of the pressure and temperature states of the cycle. The dashed sloped curve is the saturation curve, and the dotted vertical portions represent the capillary pressure differentials.

Grahic Jump Location
Fig. 1

Schematic of the air-cooled heat sink design. A LHP constitutes the fin structure. The bottom layer is the evaporator, and the remaining stacked array is the condensers. A low-profile motor is mounted on top of the structure, driving a shaft that spins the impeller blades that rotate between the condensers. Rotation of the impellers draws ambient air from the top of the device, into a central core, and across the individual condensers for convective cooling.

Grahic Jump Location
Fig. 11

Temperature difference between the vapor and ambient air plotted against heat load for different fan speeds. Data for 3300 and 5000 rpm were taken at a working fluid filling volume of 67 mL and data for 4200 rpm includes the volumes of 63, 65, and 67 mL. The dotted lines are linear fits to the data, and the slopes of these fits are shown.

Grahic Jump Location
Fig. 4

Expected temperatures ( °C) for operation with a 1000 -W heat load and a base temperature of 80 °C, at a cross-section 3 cm away from the edge of the evaporator. A significant fraction of the thermal gradient is maintained across the insulating wick layer. Modeling was performed with COSMOL Multiphysics.

Grahic Jump Location
Fig. 5

Cross-sectional image of a prototype evaporator, showing the three-layer sintered wick structure. The centerline of the evaporator is shown; an insert is made in the evaporator frame for the impeller shaft bearing. The insulating wick (<44 μm, Monel 400) separates the vapor channels and compensation chamber, which are formed as channels within the high conductivity (5–15 μm, copper) and high permeability (40–75 μm, copper) wicks, respectively.

Grahic Jump Location
Fig. 6

Expected temperatures (°C) on the outer surface of the condenser. The central, condensation space is uniform in temperature, while a significant thermal gradient occurs across the subcooler regions. Total heat removed per condenser face is 61.5 W for the modeled case. Modeling was performed with COSMOL Multiphysics.

Grahic Jump Location
Fig. 7

Image of a symmetric half of a condenser. The vapor ports access the central vapor/condensation space, and the liquid ports connect to the liquid return channels. The subcooler is shown as a partition between the vapor and liquid regions.

Grahic Jump Location
Fig. 8

Experimental setup. (a) Single-condenser LHP. The condenser and the top surface of the evaporator are cooled with two impellers. The reservoir was added for possible operation as a capillary pumped loop; this was not tested in this study. (b) Single-condenser LHP with the instrumented filling manifold. Two pressure transducers and a thermistor measure the vapor and liquid pressures and the vapor temperature, respectively. The four valves on the manifold allow the LHP to be isolated from the filling station for testing.

Grahic Jump Location
Fig. 9

Schematic representation of experimental instrumentation. “P” and “T” indicate pressure and temperature measurements, and “V”, “L,” and “S” indicate vapor, liquid, and surface measurements, respectively. The liquid pressure was obtained 100 mm above the condenser and 112 mm above the compensation chamber in the evaporator. The liquid pressures at the respective locations were calculated by adding the appropriate hydrostatic heads to the measured pressures.

Grahic Jump Location
Fig. 10

Startup characteristics of the LHP showing the vapor and liquid pressures plotted as a function of time, at a fan speed of 4200 RPM. The applied heat load (dashed) is also shown schematically for reference (right axis).

Grahic Jump Location
Fig. 12

Pressure difference as a function of the heat load for different fan speeds. The dotted line indicates the expected pressure drop across the condenser wick. The (receding) capillary pressure is the difference between the vapor–liquid pressure difference (experimental data) and the pressure drop across the wick (calculated predictions). Data for 3300 and 5000 rpm were taken at the filling volume of 67 mL; data for 4200 rpm were taken at 69 mL.

Grahic Jump Location
Fig. 13

Temperature difference between the vapor and ambient air plotted as a function of the filling volumes for varying heat loads. Data were taken at a fan speed of 4200 rpm.

Grahic Jump Location
Fig. 14

Vapor and liquid pressures, plotted as a function of filling volume. The subscripts “V” and “L” indicate vapor and liquid, respectively. The vapor and liquid pressures are equal when the heat pipe is flooded (>78 mL). Data were taken at a fan speed of 4200 rpm.

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In