Jet Impingement Boiling Heat Transfer Research Articles

Jet impingement boiling heat transfer performance of refrigerant HP-1 in micro-pin-finned surfaces for high-power chips

In this paper, jet impingement boiling heat transfer performance was studied using a novel new environmentally friendly refrigerant medium (HP-1) based on a high-power chip heat sink. A series of experimental investigations of jet impingement boiling were conducted on smooth and micro-pin-finned surfaces (PF0.5–0.5–1, PF0.5–0.5–2 and PF0.3–0.3–1). Heat transfer performances were evaluated over the saturated pressures from 450 to 590 kPa, liquid subcoolings from 2 to 15 K, nozzle sizes from 1 to 2 mm, mass fluxes from 71 to 858 kg/(m2s). The results show that critical heat flux (CHF) and heat transfer coefficient (HTC) increases with the increase of mass flux and saturated pressure. At ΔTsub=15 K, the CHF of the micro-pin-finned surfaces can be increased by 63 % compared with smooth surface, in which the CHF of PF0.5–0.5–2 surface is up to 1532 kW/m2. Boiling heat transfer coefficient first increases and then decreases with the increase of heat flux. Due to the increase of the inlet liquid subcooling, leading to a large temperature difference, so the heat transfer coefficient is smaller at a high liquid subcooling. Several factors, such as mass flux, saturated pressure and liquid subcooling, have a great impact on pressure drop. With the increase of vapor quality, the shear action between the vapor-liquid phase is enhanced, and the pressure drop will gradually increase. Bubble behaviors have also been captured by a high-speed camera to help analyze the mechanism of jet impingement boiling heat transfer in a confined space for high-power chips.

Experimental investigation on confined jet impingement boiling heat transfer characteristics on metal foam cover

Metal foam has great potential for promoting boiling heat transfer due to its complex porous structure and extraordinarily large specific surface area. The boiling heat transfer characteristics of confined single jet impingement on a metal foam covered surface were experimentally investigated for the first time, and the new data were compared with those on the smooth surface. The experimental results show that, the heat transfer performance of metal foam is affected by jet velocity and subcooled degree in the nucleate boiling region due to the significant contribution of single-phase convection heat transfer; critical heat flux of the metal foam covered surface is 10 %–36 % higher than that of the smooth surface, representing a maximal of 206.7 W·cm−2; the highest heat transfer coefficient is up to 39.9 kW·m−2·K−1, 31 % greater than that of the smooth surface; at the same surface superheated degree, the heat flux of the metal foam covered surface is 9 %–166 % greater than that of the smooth surface; at the same heat flux, the heat transfer coefficient on the metal foam covered surface is 14 %–90 % higher than that of the smooth surface; the metal foam cover reduces surface superheated degree by up to 16.8 K at the onset of nucleate boiling.

Computational investigation of single and multi-jet array impingement boiling

Jet impingement boiling has been studied extensively and has been identified as one of the most promising thermal management techniques for high heat flux applications. Unfortunately, only a few numerical studies have been reported in literature and they are mostly limited to single jets. In the present study, both submerged single round jets and confined multi-jet arrays are investigated numerically, using the Eulerian multiphase framework with the Rensselaer Polytechnic Institute (RPI) boiling model to predict heat transfer. The numerical results of the single jet case correlate well with reported experimental data and previously reported numerical results. The numerical results of the multi-jet array correlate well with experimental data reported in the literature, proving that the RPI boiling model can successfully predict the heat transfer of jet array boiling. The effect of conjugate heat transfer in jet impingement boiling heat transfer is also investigated for both single and multiple jet cases. The single-jet results agree with previous reported numerical studies. To improve numerical convergence, especially for higher heat fluxes, use was made of a hydrostatic pressure gradient at the outlet. This allowed for significant improvement in the convergence of the continuity equation. Finally, parametric analyses were conducted for both single and multi-jet arrays in the fully developed nucleate boiling regimes. Parameters included jet-to-surface spacing, Reynolds number and subcooling. The results for the single jet correlate well with the observations of experiments reported in the literature. The results for the multi-jet array showed less sensitivity to changes in jet velocity at low jet-to-surface spacing than the single-jet case. Both single and multi-jet cases showed that reducing the subcooling resulted in an onset of nucleate boiling at lower heat fluxes and that the boiling curve shifted to the left in the nucleate boiling regime.

Jet impingement boiling heat transfer from rock to liquid nitrogen during cryogenic quenching

Jet impingement quenching heat transfer of LN2 on rock surface was experimentally investigated. The test plate was circular in shape and was 16 cm in diameter and 2 cm in thickness. An aluminum alloy specimen with the same configuration as the rock plate was also used to highlight the characteristics of rock material in affecting the LN2 jet impingement boiling heat transfer. Comparing to the aluminum, rock has a low thermal conductivity and the rock surface was distributed with micro pores and cavities. This surface character enabled a 100 °C increase of the rewetting temperature. An attempt was made to quantitatively explain the rewetting temperature enhancement using the local cold spot theory. The stagnation point on rock surface was directly wetted by LN2 jet without film boiling, for which the high rewetting temperature and fast temperature drop were proposed as the reasons. The maximum heat flux and wetting front velocity were investigated for both quenching samples and the results were compared to correlations in the literature.

Heat transfer correlations for jet impingement boiling over micro-pin-finned surface

Heat transfer performance of submerged jet impingement boiling over staggered micro-pin-finned surfaces was investigated using air-dissolved FC-72. The dimension of the silicon chips is 10 × 10 × 0.5 mm3 (length × width × thickness) on staggered micro-pin-fins with four dimensions of 30 × 30 × 60 μm3, 50 × 50 × 60 μm3, 30 × 30 × 120 μm3 and 50 × 50 × 120 μm3 (width × thickness × height, named S-PF30-60, S-PF50-60, S-PF30-120, and S-PF50-120) were fabricated by using the dry etching technique. The effects of micro-pin-fins, jet-to-target distance (H = 3, 6, and 9 mm), and jet Reynolds number (Re = 2853, 5707, and 8560) on jet impingement boiling heat transfer performance were explored. For comparison, experiments with jet impinging on a smooth surface were also conducted. The results showed that all micro-pin-finned surfaces show better heat transfer performance than that of a smooth surface. The largest Nusselt number is 1367, corresponding to a heat transfer coefficient of 26387 W·m−2·K−1 with S-PF30-120 at Re = 8560, H/d = 2, and q = 151 W·cm−2, which is approximately twice the largest Nusselt number of Chip S. In the single-phase heat-transfer-dominant region, the Nusselt number (Nu) is mainly influenced by several dimensionless numbers, including Reynolds number (Re), boiling number (Bo), the ratio of jet-to-target distance to jet diameter (H/d), the ratio of micro-pin-finned surface area to smooth surface area A/AS, and a dimensionless number corresponding to flow resistance Dh/Lh. Correlations to predict Nu in both single-phase heat-transfer-dominant region and two-phase heat-transfer-dominant region for smooth and micro-pin-finned surfaces were proposed. The results show that most data (96%) in the single-phase heat-transfer-dominant region and most data (96%) in the two-phase heat-transfer-dominant region were predicted within ±13% and ±15%, respectively. In addition, CHF correlations for smooth and micro-pin-finned surfaces were also proposed, and most data (95%) are predicted within ±20% for a smooth surface and all the data within ±5% for the micro-pin-finned surfaces.

Confined jet impingement boiling in a chamber with staggered pillars

The jet impingement boiling heat transfer is investigated experimentally and numerically in a cylindrical chamber with the dense circular pillars, which is a simplified model of the shell-side chamber of the phase-change heat exchanger in a thermoacoustic Stirling engine. The saturated or subcooled liquid water is impinged onto the horizontal pillars from the bottom inlet of the boiling chamber, whereas the two-phase flow is exhausted through the outlet on the top of the chamber. Aside from the measurements of the wall heat flux and wall temperature, the boiling induced bubbly flow patterns are optically observed with a high-speed camera through the transparent window. In the simulation, the boiling heat transfer is predicted with Rensselaer Polytechnic Institute (RPI) boiling model, in which the wall heat flux is composed of three parts. The predicted boiling curve is compared with the measured one, and the calculated distributions of the vapor volume fraction adjacent to the boiling wall are compared with the captured images. The reasonable agreements have been reached. Three different regimes are identified based on the captured images. The jet velocity plays less significant role in the low heat flux cases for the saturated jets, but more in the high heat flux scenarios for the subcooled jet impingement boiling. The latent heat of phase change weakens the effect of jet velocity when crossing the boiling point. Moreover, a higher velocity of the saturated jet results in a later transition to the fully established boiling regime with a higher wall superheat. After the transition point, the predicted wall heat flux components are almost independent of wall heat flux or wall superheat, indicating that they are the good indicators to determine the boiling regime in the simulation.

The jet impingement boiling heat transfer with ad hoc wall thermal boundary conditions

In a two-phase heat exchanger, the thermal boundary condition at the boiling wall plays an import role. To investigate the characteristics of its flow and heat transfer, it is required to solve a three-domain conjugation heat transfer problem which takes into account of boiling, conduction, and air convection. In the current design, the saturated water flows into a cylindrical chamber with a tube array, whereas the hot air travels outside of chamber and boils the water inside. The effects of the water inlet velocity and hot air inlet mass flow rate are measured in the experiments. A simulation tool with Graphical User Interface code is developed for predicting the three-domain conjugation heat transfer. The boiling heat transfer in the complex case is showed to be well explained in the approach of combining the Rensselaer Polytechnic Institute (RPI) boiling model. The experiments indicate that the wall temperature on the solid-air interface and the transferred energy are independent of water inlet velocity but significantly depend on the air inlet mass flow rate. The wall temperatures in the centre core area (tube array region) are relatively uniform, whereas a huge temperature gradient is measured in the peripheral area. The maximum temperature difference in the core region is only around 17.7% or 24.6% of the core-to-peripheral temperature difference in the cases with the high or low air inlet velocity. The experimental observations have been reproduced in the simulation. The heat transfers on the hot air side and the water boiling side significantly influence each other. Prominent variations of wall temperature and heat flux result in a co-existence of single-phase water convection and the water boiling flow. It demonstrates that the conjugation has to be considered with applying an ad hoc thermal boundary conditions in the cases.

Confined Submerged Jet Impingement Boiling of Subcooled FC-72 over Micro-Pin-Finned Surfaces

Heat transfer characteristics of confined submerged jet impingement boiling of air-dissolved FC-72 on heated micro-pin-finned surfaces are presented. The dimension of the silicon chips is 10 × 10 × 0.5 mm3 (length × width × thickness) on micro-pin-fins with the four dimensions of 30 × 30 × 60 μm3, 50 × 50 × 60 μm3, 30 × 30 × 120 μm3, and 50 × 50 × 120 μm3 fabricated by using the dry etching technique. For comparison, experiments of jet impinging on a smooth surface were also conducted. The results have shown that submerged jet impingement boiling gives a large heat transfer enhancement compared with pool boiling, and all micro-pin-fins showed better heat transfer performance than a smooth surface. The effects of jet Reynolds number, jet inlet subcooling, micro-pin-fins, and nozzle-to-surface distance on jet impingement boiling heat transfer were explored. For micro-pin-fins, the maximum allowable heat flux increases with jet Reynolds number and subcooling. The largest value of the maximum allowable heat flux of micro-pin-fins by submerged jet impingement boiling is 157 W/cm2, which is about 8.3 times as large as that for the smooth surface in pool boiling. Also, Nusselt number has a strong dependence on Reynolds number.

Effect of conjugation on jet impingement boiling heat transfer

Boiling involves significant bubble induced flow mixing and phase change which give rise to a phenomenal heat transfer rate. The jet impingement produces a strong convection and influences the nucleate boiling heat transfer and burn-out characteristics. Experimental studies on jet impingement boiling heat transfer have been conducted, whereas only limited numerical simulations have been reported. In the current study, three jet impingement configurations are studied with a numerical approach in which the Rensselaer Polytechnic Institute (RPI) boiling model is applied. The simulation results agree well with the reported experimental data. The RPI boiling model is able to simulate the subcooled jet impingement boiling heat transfer in submerged and confined configurations. Finally, the effect of conjugation on jet impingement boiling heat transfer is investigated. The conjugation makes a great difference in this case. The results suggest that the data obtained in the experiments with the different heating schemes (thick copper block or thin-film heater, corresponding to with or without conjugation), although the jet parameters are same, can be significantly different and incomparable. Attention should be paid on heating scheme in the experimental and numerical studies.

Comparison of Deionized Water and FC-72 in Pool and Jet Impingement Boiling Thermal Management

High heat fluxes and stringent constraints on surface temperature and its uniformity during thermal management of electrical and electronic components often necessitate use of boiling heat transfer. This paper compares the pool and jet impingement boiling heat transfer characteristics of deionized water and FC-72 at an equivalent fluid saturation temperature of 57°C and for identical experimental conditions. To lower the saturation temperature of water down to 57°C, experiments with water are performed at a reduced absolute system pressure of 0.176 bar. Despite the reduction in pressure, pool boiling critical heat flux with deionized water is found to be 3.6 times larger than with FC-72. Furthermore, jet impingement is seen to enhance boiling heat transport more significantly for water than for FC-72. Consequently, heat transfer coefficients during jet impingement boiling are as much as 3.9 times larger for water compared to FC-72 at identical Reynolds numbers and surface temperatures. The heat transfer advantage of using water is mainly associated with the superior thermophysical properties of this fluid. However, in addition to the large fluid saturation temperature at atmospheric conditions, direct cooling of electronics is frequently not possible using water due to the incompatibility of the fluid with electrical components. To assess the practical utility of subatmospheric deionized water through indirect cooling of electronics, a 1-D heat sink analysis is performed on a multichip module geometry. The overall thermal resistance of the heat sink using water is determined to be around two times lower than that of direct cooling of FC-72 on a silicon substrate. Under saturation condition of the working fluids, dissipation of heat fluxes in excess of ~45 W/cm2 with the multichip module using water is constrained by the chip surface temperature limit.

Two-phase cooling characteristics of a saturated free falling circular jet of HFE7100 on a heated disk: Effect of jet length

Direct jet impingement boiling heat transfer operating at low flow rates is of great interest for the localized moderate heat fluxes from the targets with delicate mechanical structure, where the aggressive techniques such as high-speed jets are not suitable. Boiling heat transfer from an upward facing disk targeted by a falling jet was studied experimentally at different volumetric flow rates and various jet lengths. The working fluid was chosen to be the dielectric liquid HFE7100 and the heated spot was an 8-mm diameter disk. Using previous CHF correlations in their original form, valid at very low volumetric flow rates, results in large disagreements since it was found that variation in the jet length changes the boiling characteristics. It is demonstrated that although the circular hydraulic jump formation within the heater diameter may suppress the heat transfer under certain conditions, moving the jet closer to the target may significantly improve the boiling curves at the critical heat flux (CHF) regime. At low flow rates, the CHF increases as the jet length decreases while for moderate and high flow rates the boiling curves show approximately a universal behavior for different jet lengths. For such low flow rates, the effect of jet length on boiling curves was shown to be related to the variation of the cross section of the falling jet and the formation of hydraulic jump at radial distances smaller than the heater diameter. The current CHF results for different jet lengths are correlated by including the effect of jet length in the previous correlation proposed by Sharan and Lienhard.

Local jet impingement boiling heat transfer with R113

An experimental study was performed to characterize the boiling heat transfer of impinging circular submerged jets on simulated microelectronic chips with a nominal area of 5 mm × 5 mm. The heat transfer modes included natural convection, partially developed nucleate boiling, fully developed nucleate boiling and critical heat flux. The study included the effects of jet parameters and fluid subcooling on the nucleate boiling. The results showed that the nucleate boiling data varied only with fluid subcooling regardless of jet parameters and that both the pool and impingement nucleate boiling curves at the same subcooling condition were well correlated. The high heat flux portions of the boiling curves with jet exit velocities greater than 10 m/s were corrected for the elevated saturation temperature. A new expression was developed with an interpolation method to construct the partially developed nucleate boiling curve.

Local heat transfer from a hot plate to a water jet

Jet impingement boiling is very efficient in cooling of hot surfaces as a part of the impinging liquid evaporates. Because of its importance to many cooling procedures, investigations on basic mechanisms of jet impingement boiling heat transfer are needed. Until now, most of the experimental studies, carried out under steady-state conditions, used a heat flux controlled system and were limited by the critical heat flux (CHF). The present study focuses on steady-state experiments along the entire boiling curve for hot plate temperatures of up to 700°C. A test section has been built up simulating a hot plate. It is divided into 8 independently heated modules of 10 mm length to enable local heat transfer measurements. By means of temperature controlled systems for each module local steady-state experiments in the whole range between single phase heat transfer and film boiling are possible. By solving the two dimensional inverse heat conduction problem, the local heat flux and the corresponding wall temperature on the surface of each module can be computed. The measurements show important differences between boiling curves measured at the stagnation line and those obtained in the parallel flow region. At the stagnation line, the transition boiling regime is characterised by very high heat fluxes, extended to large wall superheats. Inversely, boiling curves in the parallel flow region are very near to classical ones obtained for forced convection boiling. The analysis of temperature fluctuations measured at a depth of 0.8 mm from the boiling surface enables some conclusions on the boiling mechanism in the different boiling regimes.

Jet Impingement Boiling From a Circular Free-Surface Jet During Quenching: Part 2—Two-Phase Jet

A proposed technique for controlling jet impingement boiling heat transfer involves injection of gas into the liquid jet. This paper reports results from an experimental study of boiling heat transfer during quenching of a cylindrical copper specimen, initially at a uniform temperature exceeding the temperature corresponding to maximum heat flux, by a two-phase (water-air), circular, free-surface jet. The second phase is introduced as small bubbles into the jet upstream of the nozzle exit. Data are presented for single-phase convective heat transfer at the stagnation point, as well as in the form of boiling curves, maximum heat fluxes, and minimum film boiling temperatures at locations extending from the stagnation point to a radius of ten nozzle diameters. For void fractions ranging from 0.0 to 0.4 and liquid-only velocities between 2.0 and 4.0 m/s 11,300⩽Red,fo⩽22,600, several significant effects are associated with introduction of the gas bubbles into the jet. As well as enhancing single-phase convective heat transfer by up to a factor of 2.1 in the stagnation region, addition of the bubbles increases the wall superheat in nucleate boiling and eliminates the temperature excursion associated with cessation of boiling. The maximum heat flux is unaffected by changes in the void fraction, while minimum film boiling temperatures increase and film boiling heat transfer decreases with increasing void fraction. A companion paper (Hall et al., 2001) details corresponding results from the single-phase jet.