Spray Liquid-film Flow Theory And Experimental Study Of Spray Cooling Of Electronic Devices

With the heat flux of electronic devices becoming higher and higher, it has been an important problem to discharge the heat produced, which could decrease the reliability of the devices. As one of the three promising solutions for cooling of high heat flux applications, spray cooling technology has received much attention and been widely studied in recent years. Much experimental work has been done on spray cooling. However, the experimental results do not always agree with each another, and theoretical understandings on spray cooling are still at their early stage due to the intrinsic complexity of the mechanisms involved.Based on the state of the art of spray cooling, the following topics have been studied:As to a cone-axisymmetric spray cooling systems, the spray area was divided, and the slip-flow boundary condition was introduced. Based on an assumption of the coniform spray, an equivalent continuum model of the average spray density was established. Based on the variable separation method (VSM), the governing equations for the film thickness problem have been solved. A reasonable consistent condition of the momentum equations has been found and solved. Based on the methods ofmathematical physics, series expansion solution of the velocity filed in the film has been derived. A boundary condition at the boundary of the spray area was given and the velocity slip phenomenon has been considered. The undeterminde coefficients were derived, and the functions of the film thickness and the fluid field of the film within the range of the impact area have been established. Simulation results indicate that: 1. if the spray velocity uniformly distributes in the area, the film is even, while if the spray velocity distributes unevenly, the film can be either central-raised or central-cupped; 2. the structure of the liquid film could be approximated as a two-layer liquid layer. The velocity of the liquid in bottom layer is lager than that in the top layer, and there are two methods for a thinner liquid film: increase spray velocity and decrease spray density; 3. the slip-flow boundary condition has little effect on the film thickness and fluid, which could be ignored as a result.A reasonable CHF (critical heat flux) criterion was analyzed and ascertained, which states that the maximum CHF can be achieved when the spray is configured on the condition that the spray impact area inscribes the square test surface. The tracks of nozzles with different cone angles were studied with different inclination angles, based on the CHF criterion aforementioned. When the nozzle is a rectangular one, there are two kinds of inclination modes, namely, the axis-incline and the diagonal-incline modes. Simulation results indicate that, circular nozzles are better if the heated surface is a circular one because of its larger coverage area. Rectangular nozzles are better when the spray cooling surface is a rectangular one, and the axis-incline model is better than the diagonal-incline model when there is an inclination.Based on the data obtained by Visaria et al in their work, a new CHF model was established and new expression of CHF value was derived based on the nozzle track model. The effects of the spray inclination angle on CHF were predicted. Simulation results indicate that CHF increases slightly as the inclination angle increases.An experimental system was constructed, which includes nozzles, a heater, a pump and thermocouples. The characteristics (the droplet speed and diameter) of the system were estimated by using correlative theory considering the effect of the drag force.The relationships of the pressure drop across the three DANFOSS nozzles and the pressure in the outlet of the pump were conducted, linear relations were derived. A separate study was conducted to find the relationships of the total volumetric flow rate Q and the pressure drop across the nozzles ?P. It shows that the experimental total volumetric flow rates are well approximated by a ?P0.5 power law. More studies were conducted to find the special volumetric flow rate at the spray cone centerline, and then get the best working conditions of the three nozzles.Different commercial pressurized full cone DANFOSS nozzles were used to spray onto a 30×30mm2 square copper heated horizontal surface without inclination angle, and the optimal distance between the nozzles and the heated surface were achieved. An optimal heat transfer criterion (called H criterion) was proposed, which means that the optimal heat transfer appears when the region outside the impellent thin spray film inscribes in the square heated surface.Based on the H criterion aforementioned, two DANFOSS nozzles of the three with inclined spraying angles were used to experimentally study the temperature distribution in the heated surface. Distilled water was used to spray on the square surface afore- mentioned and a heated circular copper surface with diameter of 30mm, respectively. The results indicate that the increasing grad of the surface temperature is more important than the decreasing heat transfer capability while there are inclination angles.A new concept called dynamic spray cooling was proposed based on the concept of intermittent spray cooling. Electromagnetic actuated devices were chosen for a dynamic spray cooling system. The pull-in phenomena and its state of the art of the electromagnetic actuated devices were discussed. Based on the classification of the electromagnetic actuated devices, the quasi-static, dynamic, and modulated dynamic pull-in characteristics for parallel-plate and torsional magnetostatic devices with step input magneto motive force were analyzed.