A miniature loop heat pipe (LHP) with polytetrafluoroethylene (PTFE) wicks was fabricated and its evaporator thermal performance was investigated with parametric experiments. The variables considered were the clearance between the cylindrical evaporator casing and the wick, the working fluid inventory, the properties of the working fluids, and the sink temperature. Micro-gaps between the outer surface of the wick and inner surface of the evaporator casing were included in the experiments with variables of clearance to investigate their effect on the evaporator heat-transfer coefficient. Ethanol, acetone, and R134a were charged in the LHP to evaluate the effect of the properties of the working fluids. The LHP tests were conducted with several W/cm 2 of applied heat flux to the evaporator under controlled sink temperature. The clearance seriously affected the evaporator heat-transfer coefficient, with a gap of 20 μm between the wick and casing having the best effect on the evaporator heat transfer across a range of tested heat fluxes. The effects of the working fluid inventory and fluid properties on the evaporator heat transfer were also clarified.Finally, the developed LHP was tested using ethanol as a working fluid under a stepwise heat load and sink temperature.

A transient mathematical model is developed to study the transient response and analyze the distribution of heat load in a loop heat pipe. The model is based on the one-dimensional and time-dependent conservation equations for heat and fluid flow. The momentum and energy conservation equations for each of the loop heat pipe components are solved. The model results are compared against the data obtained from two miniature loop heat pipes using polytetrafluoroethylene wicks, ethanol, and acetone as working fluids. The mathematical model satisfactorily predicts the dynamic behavior of the loop heat pipe unit. It is shown that the percentage of heat leak across the wick decreases and the ratio of latent heat increases with increasing heat load. Some temperature overshoots observed in the calculation results are not observed in the experimental data. When a new power is applied, no time lag is observed in the loop heat pipe response between the simulation and experimental results. Nomenclature A = cross section area, m 2 A s = surface area, m 2 C = heat capacity, J∕kg c = constant in Chisholm correlation c p = specific heat at constant pressure, J∕kg K D = diameter, m f = Darcy's friction coefficient G A B = thermal conductance between A and B, W∕K Gr = Grashof number g = gravity, m∕s 2 h = heat transfer coefficient, W∕m 2 K h lat = latent heat, J∕kg k = thermal conductivity, W∕m K L = length, m _ m = mass flow rate, kg∕s Nu = Nusselt number Pr = Prandtl number p = pressure, Pa _ Q A B = rate of heat transfer from A to B, W _ Q apply = heat load, W q = amount of heat transfer per volume, W∕m 3 Re = Reynolds number T = temperature,°C u = velocity, m∕s V = volume, m 3 X = ratio between vapor and liquid frictional pressure loss Greek β = coefficient of volume expansion, 1∕K ε = porosity ν = kinematic viscosity coefficient, m 2 ∕s ρ = density, kg∕m 3 τ w = wall shear stress, Pa Φ i = square root of ratio between the frictional pressure loss in single phase i and two-phase flow Subscript amb = ambient bay = bayonet tube cc = compensation chamber e = evaporator eff = effective fc = forced convection gr = groove hb = heater block int = interface l = liquid nc = natural convection sat = saturation sub = subcooled v = vapor 2f = two phase

The heat and mass transfer in a porous wick within a capillary evaporator were analyzed in three dimensions, using a pore network model to simulate immiscible liquid-vapor flow and phase changes in the capillary structure. Characteristics of the porous structure, such as pore radius distribution, permeability and porosity were obtained from measurements of an actual wick and were included in the calculations. Scenarios involving both a fully liquid-saturated and an unsaturated wick containing liquid and vapor were examined under steady state conditions. The location at which the initial vapor phase was generated was identified based on classical nucleation theory. The transition from a saturated to an unsaturated wick was assessed by modeling the distribution of the liquid-vapor interface, and the wicks were compared based on color 3D renderings of temperature and pressure distributions and meniscus curvatures. The maximum temperature of the unsaturated wick exceeded that of the saturated wick although the area of the liquid-vapor interface in the unsaturated wick was five times that in the saturated wick. Since a distribution of pore radii was considered in these calculations, the interface within the wick was not smooth but exhibited asperity. The distribution of meniscus curvatures in the unsaturated wick was much wider compared with the saturated wick. The results obtained in the case of an unsaturated wick demonstrated the occurrence of the heat pipe effect, induced by a distribution of capillary pressures.

No abstract