Hot water cooled electronics: Exergy analysis and waste heat reuse feasibility

We report an experimental study on exergetically efficient electronics cooling using hot water as coolant. It is shown that water temperatures as high as 60 °C are sufficient to cool microprocessors with over 90% first law (energy based) efficiency. The chip used in our experiment is kept at temperatures of 80 °C or below so as not to exceed any allowable industrial specifications for maximum microprocessor chip temperature. The use of hot water as coolant will eliminate the requirement for chillers typically used in air-cooled data centers and, therefore, significantly reduce the power consumption. An exergy analysis shows that a six fold rise in second law (exergy based) efficiency is achieved by switching the water inlet temperature from 30 °C to 60 °C. The resulting high exergy at the heat sink outlet is a measure of the potential usefulness of the waste heat of data centers, thereby helping to design data centers with minimal carbon footprint. A new metric for the economic value of the recovered heat, based on costs for electricity and fossil fuels, heat recovery efficiency and an application specific utility function, is introduced to underscore the benefits of hot water cooling. This new concept shows that the economic value of the heat recovered from data centers can be much higher than its thermodynamic value.     

Experimental analysis of flow and heat transfer in a miniature porous heat sink for high heat flux application

A novel miniature porous heat sink system was presented for dissipating high heat fluxes of electronic device, and its operational principle and characteristics were analyzed. The flow and heat transfer of miniature porous heat sink was experimentally investigated at high heat fluxes. It was observed that the heat load of up to 280 W (heat flux of 140 W/cm²) was removed by the heat sink with the coolant pressure drop of about 34 kPa across the heat sink system and the heater junction temperature of 62.9 °C at the coolant flow rate of 6.2 cm³/s. Nu number of heat sink increased with the increase of Re number, and maximum value of 323 for Nu was achieved at highest Re of 518. The overall heat transfer coefficient of heat sink increased with the increase of coolant flow rate and heat load, and the maximal heat transfer coefficient was 36.8 kW(m² °C)^?1 in the experiment. The minimum value of 0.16 °C/W for the whole thermal resistance of heat sink was achieved at flow rate of 6.2 cm³/s, and increasing coolant flow rate and heat fluxes could lead to the decrease in thermal resistance. The micro heat sink has good performance for electronics cooling at high heat fluxes, and it can improve the reliability and lifetime of electronic device.     

Saturated flow boiling heat transfer and associated bubble characteristics of FC-72 on a heated micro-pin-finned silicon chip

An experiment is carried out here to investigate flow boiling heat transfer and associated bubble characteristics of FC-72 on a heated micro-pin-finned silicon chip flush-mounted in the bottom of a horizontal rectangular channel. Besides, three different micro-structures of the chip surface are examined, namely, the smooth, pin-finned 200 and pin-finned 100 surfaces. The pin-finned 200 and 100 surfaces, respectively, contain micro-pin-fins of size 200 μm × 200 μm × 70 μm (width × length × height) and 100 μm × 100 μm × 70 μm. The pitch of the fins is equal to the fin width for both surfaces. The effects of the FC-72 mass flux, imposed heat flux, and surface micro-structures of the silicon chip on the FC-72 saturated flow boiling characteristics are examined in detail. The experimental data show that an increase in the FC-72 mass flux causes a delay in the boiling incipience. However, the flow boiling heat transfer coefficient is not affected by the coolant mass flux. But adding the micro-pin-fin structures to the chip surfaces can effectively enhance the single-phase convection and flow boiling heat transfer. Moreover, the mean bubble departure diameter and active nucleation site density are reduced for a rise in the FC-72 mass flux. A higher coolant mass flux results in a higher mean bubble departure frequency. Furthermore, larger bubble departure diameter, higher bubble departure frequency, and higher active nucleation site density are observed at a higher imposed heat flux. We also note that adding the micro-pin-fins to the chips decrease the bubble departure diameter and increase the bubble departure frequency. However, the departing bubbles are larger for the pin-finned 100 surface than the pin-finned 200 surface but the bubble departure frequency exhibits an opposite trend. Finally, empirical equations to correlate the present data for the FC-72 single-phase liquid convection and saturated flow boiling heat transfer coefficients and for the bubble characteristics are provided.     

Subcooled flow boiling heat transfer and associated bubble characteristics of FC-72 on a heated micro-pin-finned silicon chip

Experiments are conducted here to investigate subcooled flow boiling heat transfer and associated bubble characteristics of FC-72 on a heated micro-pin-finned silicon chip flush-mounted on the bottom of a horizontal rectangular channel. In the experiments the mass flux is varied from 287 to 431 kg/m² s, coolant inlet subcooling from 2.3 to 4.3 °C, and imposed heat flux from 1 to 10 W/cm². Besides, the silicon chips contain three different geometries of micro-structures, namely, the smooth, pin-finned 200 and pin-finned 100 surfaces. The pin-finned 200 and 100 surfaces, respectively, contain micro-pin-fins of size 200 μm × 200 μm × 70 μm (width × length × height) and 100 μm × 100 μm × 70 μm. The measured data show that the subcooled flow boiling heat transfer coefficient is reduced at increasing inlet liquid subcooling but is little affected by the coolant mass flux. Besides, adding the micro-pin-fin structures to the chip surface can effectively raise the single-phase convection and flow boiling heat transfer coefficients. Moreover, the mean bubble departure diameter and active nucleation site density are reduced for rises in the FC-72 mass flux and inlet liquid subcooling. Increasing coolant mass flux or reducing inlet liquid subcooling results in a higher mean bubble departure frequency. Furthermore, larger bubble departure diameter, higher bubble departure frequency, and higher active nucleation site density are observed as the imposed heat flux is increased. Finally, empirical correlations for the present data for the heat transfer and bubble characteristics in the FC-72 subcooled flow boiling are proposed.     

Numerical study of laminar heat transfer and pressure drop characteristics in a water-cooled minichannel heat sink

With the rapid development of the information technology (IT) industry, the heat flux in integrated circuit (IC) chips cooled by air has almost reached its limit about 100 W/cm². Some applications in high technologies require heat fluxes well beyond such a limitation. Therefore the search of a more efficient cooling technology becomes one of the bottleneck problems of the further development of IT industry. The microchannel flow geometry offers large surface area of heat transfer and a high convective heat transfer coefficient. However, it has been hard to implement because of its very high pressure head required to pump the coolant fluid though the channels. A normal channel could not give high heat flux although the pressure drop is very small. A minichannel can be used in heat sink with a quite high heat flux and a mild pressure loss. A minichannel heat sink with bottom size of 20 mm × 20 mm is analyzed numerically for the single-phase laminar flow of water as coolant through small hydraulic diameters and a constant heat flux boundary condition is assumed. The effects of channel dimensions, channel wall thickness, bottom thickness and inlet velocity on the pressure drop, thermal resistance and the maximum allowable heat flux are presented. The results indicate that a narrow and deep channel with thin bottom thickness and relatively thin channel wall thickness results in improved heat transfer performance with a relatively high but acceptable pressure drop. A nearly-optimized configuration of heat sink is found which can cool a chip with heat flux of 256 W/cm² at the pumping power of 0.205 W. The nearly-optimized configuration is verified by an orthogonal design. The simulated thermal resistance agrees quite well with the result of conventional correlations method with the maximum difference of 12%.     

Entropy generation of vapor condensation in the presence of a non-condensable gas in a shell and tube condenser

This article investigates the entropy production of condensation of a vapor in the presence of a non-condensable gas in a counter-current baffled shell and one-pass tube condenser. The non-dimensional entropy number is derived with respect to heat exchange between the bulk fluid and condensate, as well as heat exchange between the condensate and coolant. Numerical results show that heat transfer from the condensate to the coolant has a dominant role in generating entropy. For example, at an air mass flow rate of 330 kg/h, 93.4% of the total entropy generation is due to this source. The resultant profiles during the condensation process indicate that a higher air mass flow rate leads to a lower rate of entropy production. For example, as the air mass flow rate increases from 330 kg/h to 660 kg/h and 990 kg/h, the total entropy generation decreases from 976 J/s K to 904 and 857.2 J/s K, respectively. By introducing a new parameter called the condensation effectiveness, a correlation is also developed for predictions of the entropy number, and an illustrative example is presented.     

Assessment of thermoelectric module with nanofluid heat exchanger

For applications such as cooling of electronic devices, it is a common practice to sandwich the thermoelectric module between an integrated chip and a heat exchanger, with the cold-side of the module attached to the chip. This configuration results thermal contact resistances in series between the chip, module, and heat exchanger. In this paper, an appraisal of thermal augmentation of thermoelectric module using nanofluid-based heat exchanger is presented. The system under consideration uses commercially available thermoelectric module, 27 nm Al₂O₃–H₂O nanofluid, and a heat source to replicate the chip. The volume fraction of nanofluid is varied between 0% and 2%. At optimum input current conditions, experimental simulations were performed to measure the transient and steady-state thermal response of the module to imposed isoflux conditions. Data collected from the nanofluid-based exchanger is compared with that of deionized water.Results show that there exist a lag-time in thermal response between the module and the heat exchanger. This is attributed to thermal contact resistance between the two components. A comparison of nanofluid and deionized water data reveals that the temperature difference between the hot- and cold-side, ΔT = T_h ? T_c ≈ 0, is almost zero for nanofluid whereas ΔT > 0 for water. When ΔT ≈ 0, the contribution of Fourier effect to the overall heating is approximately zero hence enhancing the module cooling capacity. Experimental evidence further shows that temperature gradient across the thermal paste that bonds the chip and heat exchanger is much lower for the nanofluid than for deionized water. Low temperature gradient results in low resistance to the flow of heat across the thermal paste. The average thermal contact resistance, R = ΔT/Q, is 0.18 and 0.12 °C/W, respectively for the deionized water and nanofluid. For the range of optimum current, 1.2 ? current ? 4.1 A, considered in this study, the COP ranges between 1.96 and 0.68.     

Sintered porous heat sink for cooling of high-powered microprocessors for server applications

This paper experimentally investigates the sintered porous heat sink for the cooling of the high-powered compact microprocessors for server applications. Heat sink cold plate consisted of rectangular channel with sintered porous copper insert of 40% porosity and 1.44 × 10^?11 m² permeability. Forced convection heat transfer and pressure drop through the porous structure were studied at Re ? 408 with water as the coolant medium. In the study, heat fluxes of up to 2.9 MW/m² were successfully removed at the source with the coolant pressure drop of 34 kPa across the porous sample while maintaining the heater junction temperature below the permissible limit of 100 ± 5 °C for chipsets. The minimum value of 0.48 °C/W for cold plate thermal resistance (R_cp) was achieved at maximum flow rate of 4.2 cm³/s in the experiment. For the designed heat sink, different components of the cold plate thermal resistance (R_cp) from the thermal footprint of source to the coolant were identified and it was found that contact resistance at the interface of source and cold plate makes up 44% of R_cp and proved to be the main component. Convection resistance from heated channel wall with porous insert to coolant accounts for 37% of the R_cp. With forced convection of water at Re = 408 through porous copper media, maximum values of 20 kW/m² K for heat transfer coefficient and 126 for Nusselt number were recorded. The measured effective thermal conductivity of the water saturated porous copper was as high as 32 W/m K that supported the superior heat augmentation characteristics of the copper–water based sintered porous heat sink. The present investigation helps to classify the sintered porous heat sink as a potential thermal management device for high-end microprocessors.     

Constructal cooling channels for micro-channel heat sinks

This paper documents the geometric optimisation of a three-dimensional micro-channel heat sink. The objective is to minimise the peak temperature from the walls to the coolant fluid. The optimisation is performed numerically by using the finite volume method. The numerical simulation was carried out on a unit cell with volume ranging from 0.1 mm³ to 0.9 mm³ and pressure drop between 10 kPa and 75 kPa. The axial length of the micro-channel heat sink was fixed at 10 mm. The cross-sectional area of the micro-channel heat sink is free to morph with respect to the degree of freedoms provided by the aspect ratio and the solid volume fraction. The effect of the total solid volume fraction and the pressure drop on the aspect ratio, channel hydraulic diameter and peak temperature is investigated. The numerical results show that the degrees of freedom have a strong effect on the peak temperature and the maximum thermal conductance. The optimal geometric characteristics obtained numerically (the aspect ratio and the optimal channel shape (hydraulic diameter)) are reported and compared with those obtained from approximate relationships using scale analysis. The predicted trends are found to be in good agreement with the numerical results.     

Development of a multiple layer vacuum insulation chip

Advent of micro thermal devices such as lab-on-a-chip and micro heat pump necessitates development of highly effective insulation chips or layers. This paper reports the development of a vacuum insulation chip (VIC) having very low effective thermal conductivity and very small thickness. Fifty nanometer thickness metal coating on both sides of an LCD glass chip and 5 μm vacuum gap are stacked in a series to decrease the heat transfer by radiation. An array of support legs is necessary to maintain the structure under the atmospheric pressure. Design of VIC involves trade-offs between the heat conduction through the multi-layer structure and the mechanical strength. A model to determine the actual design values is proposed. The results are in reasonable agreement with the more refined results using commercial numerical codes. Based on these results, a VIC of 32 × 32 × 1.88 mm³ is manufactured, and the effective thermal conductivity is measured by guarded hot plate method. The chip shows effective thermal conductivities of 0.0015 and 0.001 W/m K at vacuum levels of 1.33 and 0.24 Pa (N/m²), respectively.     

Enhanced flow boiling heat transfer of FC-72 on micro-pin-finned surfaces

For the purpose of cooling electronic components with high heat flux efficiently, some experiments were conducted to study the flow boiling heat transfer performance of FC-72 on silicon chips. Micro-pin-fins were fabricated on the chip surface using a dry etching technique to enhance boiling heat transfer. Three different fluid velocities (0.5, 1 and 2 m/s) and three different liquid subcoolings (15, 25 and 35 K) were performed, respectively. A smooth chip (chip S) and four micro-pin-finned chips with the same fin thickness of 30 μm and different fin heights of 60 μm (chip PF30–60) and 120 μm (chip PF30–120), respectively, were tested. All the micro-pin-finned surfaces show a considerable heat transfer enhancement compared to the smooth one, and the critical heat flux increases in the order of chip S, PF30–60 and PF30–120. For a lower ratio of fin height to fin pitch and/or higher fluid velocity, the fluid velocity has a positive effect on the nucleate boiling curves for the micro-pin-finned surfaces. At the velocities lower than 1 m/s, the micro-pin-finned surfaces show a sharp increase in heat flux with increasing wall superheat, and the wall temperature at the critical heat flux (CHF) is less than the upper limit, 85 °C, for the reliable operation of LSI chips. The CHF values for all surfaces increase with fluid velocity and subcooling. The maximum CHF can reach nearly 150 W/cm² for chip PF30–120 at the fluid velocity of 2 m/s and the liquid subcooling of 35 K.     

Enhancement of heat transfer coefficients by actuation against an impinging jet

Recent technological developments have lead to significant increase in the generated heat by electronic and optical components. The removal of high heat fluxes can be successfully treated by several methods, e.g. impinging jets. Further improvement is offered by incorporating arrays of jets or causing jets to pulsate. The research reported herein introduces a new method which is based on actuation of a slab against a two dimensional steady, impinging, laminar, liquid micro-jet. This leads to enhanced heat transfer in the wall region of the jet. An experimental setup which included a piezoelectric (PZT) actuator, a dedicated silicon chip and a steady, slot, impinging jet, was assembled. Using a high speed infrared (IR) radiometer, the cooling process of the chip was recorded and the heat transfer enhancement values were determined for normalized actuation amplitudes, Reynolds and Strouhal numbers in the ranges of 0.45 < δ < 0.75, 756 < Re < 1260 and 0 < St < 0.052, respectively. It was experimentally found that heat transfer coefficients were enhanced by up to 34%.     

Study on a wire-type Joule Thomson microcooler with a concentric heat exchanger

This study examines the performance of a wire-type Joule Thomson microcooler utilizing a flexible concentric counterflow heat exchanger. Three gases: C₂H₄, CO₂ and N₂ were used separately for trials conducted at inlet pressures ranging from 0.5 MPa to 5 MPa with C₂H₄ having the best performance. During unloaded tests at an inlet pressure of 2.0 MPa, C₂H₄ obtained a minimum temperature of 225 K while CO₂ obtained a minimum temperature of 232 K. Using CO₂ the microcooler was able to maintain a temperature of 273 K at 100 mW heat input and 2 MPa inlet pressure. An inlet pressure of 3 MPa allowed a 550 mW heat input at 273 K. Theoretical performance calculations were conducted and compared to experimental results revealing considerable reduction of microcooler performance due to the presence of heat in-leak. Results have displayed that the JT coefficient of the coolant gas is a more dominant factor than heat transfer properties in determining the performance of the coolant. Due to the microscale of the device, relevant scaling effects were evaluated, particularly entrance effects, surface roughness and axial conduction.     

Study on solar/waste heat driven multi-bed adsorption chiller with mass recovery

The study deals with a solar or waste heat driven three-bed adsorption cooling cycle employing mass recovery scheme. A cycle simulation computer program is developed to investigate the performance of the chiller. The innovative chiller is driven by exploiting solar/waste heat of temperatures between 60 and 90 °C with a cooling source at 30 °C for air-conditioning purpose. The performance of the three-bed adsorption chiller with mass recovery scheme was compared with that of the three-bed chiller without mass recovery. It is found that cooling effect as well as solar/waste heat recovery efficiency, η of the chiller with mass recovery scheme is superior to those of three-bed chiller without mass recovery for heat source temperatures between 60 and 90 °C. However, COP of the proposed chiller is higher than that of the three-bed chiller without mass recovery, when heat source temperature is below 65 °C.     

Optimum temperatures in a shell and tube condenser with respect to exergy

This paper focuses on evaluation of the optimum cooling water temperature during condensation of saturated water vapor within a shell and tube condenser, through minimization of exergy destruction. First, the relevant exergy destruction is mathematically derived and expressed as a function of operating temperatures and mass flow rates of both vapor and coolant. The optimization problem is defined subject to condensation of the entire vapor mass flow and it is solved based on the sequential quadratic programming (SQP) method. The optimization results are obtained at two different condensation temperatures of 46 °C and 54 °C for an industrial condenser. As the upstream steam mass flow rates increase, the optimal inlet cooling water temperature and exergy efficiency decrease, whereas exergy destruction increases. However, the results are higher for optimum values at a condensation temperature of 54 °C, compared to those when the condensation temperature is 46 °C. For example, when the steam mass flow rate is 1 kg/s and the condensation temperature increases from 46 °C to 54 °C, the optimal upstream coolant temperature increases from 16.78 °C to 25.17 °C. Also, assuming an ambient temperature of 15 °C, the exergy destruction decreases from 172.5 kW to 164.6 kW. A linear dependence of exergy efficiency on dimensionless temperature is described in terms of the ratio of the temperature difference between the inlet cooling water and the environment, to the temperature difference between condensation and environment.     

Experimental analysis and FEM simulation of finned U-shape multi heat pipe for desktop PC cooling

This paper presents the performance analysis of a finned U-shape heat pipe used for desktop PC-CPU cooling. The experiments are conducted by mounting the system vertically over a heat source situated inside a rectangular tunnel, and force convection is facilitated by means of a blower. The total thermal resistance (R_t) and heat transfer coefficient are estimated for both natural and forced convection modes under steady state condition, by varying the heat input from 4 W to 24 W, and the air velocity from 1 m/s to 4 m/s. The coolant velocity and heat input to achieve minimum R_t are found out and the corresponding effective thermal conductivity is calculated. The transient temperature distribution in the finned heat pipe is also observed. The experimental observations are verified by simulation using ANSYS 10. The results show that the air velocity, power input and heat pipe orientation have significant effects on the performance of finned heat pipes. As the heat input and air velocity increase, total thermal resistance decreases. The lowest value of the total thermal resistance obtained is 0.181 °C/W when heat input is 24 W and air velocity 3 m/s. The experimental and simulation results are found in good agreement.     

Experimental study on the thermo-physical properties of car engine coolant (water/ethylene glycol mixture type) based SiC nanofluids

The engine coolant (water/ethylene glycol mixture type) becomes one of the most commonly used commercial fluids in cooling system of automobiles. However, the heat transfer coefficient of this kind of engine coolant is limited. The rapid developments of nanotechnology have led to emerging of a relatively new class of fluids called nanofluids, which could offer the enhanced thermal conductivity (TC) compared with the conventional coolants. The present study reports the new findings on the thermal conductivity and viscosity of car engine coolants based silicon carbide (SiC) nanofluids. The homogeneous and stable nanofluids with volume fraction up to 0.5 vol.% were prepared by the two-step method with the addition of surfactant (oleic acid). It was found that the thermal conductivity of nanofluids increased with the volume fraction and temperature (10–50 °C), and the highest thermal conductivity enhancement was found to be 53.81% for 0.5 vol.% nanofluid at 50 °C. In addition, the overall effectiveness of the current nanofluids (0.2 vol.%) was found to be ~ 1.6, which indicated that the car engine coolant-based SiC nanofluid prepared in this paper was better compared to the car engine coolant used as base liquid in this study.     

Optimal geometry of L and C-shaped channels for maximum heat transfer rate in natural convection

The present paper documents the geometric optimization of L and C-shaped channels in laminar natural convection subject to global constraints. The objective is to maximize the heat transfer rate from the hot wall to the coolant fluid. Three different configurations were considered: (i) an L-shaped asymmetric vertical heated channel with an adiabatic horizontal inlet, (ii) an asymmetric vertical heated channel with an adiabatic vertical outlet, and finally, (iii) a C-shaped vertical channel with horizontal inlet and outlet. The two first configurations are free to morph according to two degrees of freedom: the wall-to-wall spacing and inlet (or outlet) height. The third configuration is optimized with respect to the wall-to-wall spacing, and the heights of the inlet and outlet ports. The effect of the inlet or outlet horizontal adiabatic duct lengths is also investigated. The optimization is performed numerically by using the finite element technique, in the range 10⁵ < Ra < 10⁷ for Pr = 0.7, where Ra is the Rayleigh number based on a fixed total height H of the channel. The numerical results show that optimization is relevant, since the three degrees of freedom considered have a strong effect on the heat transfer delivered from the hot wall to the fluid. The optimal geometric characteristics obtained numerically (i.e., optimal spacing, optimal height and lengths) are reported and correlated within a 7.5% maximal disagreement range.     

Cooling performance investigation of electronics cooling system using Al2O3–H2O nanofluid

In this study, the cooling performance of Al₂O₃–H₂O nanofluid was experimentally investigated as a much better developed alternative for the conventional coolant. For this purpose the nanofluid was passed through the custom-made copper minichannel heat sink which is normally attached with the electronic heat source. The thermal performance of the Al₂O₃–H₂O nanofluid was evaluated at different volume fraction of the nanoparticle as well as at different volume flow rate of the nanofluid. The volume fraction of the nanoparticle varied from 0.05 vol.% to 0.2 vol.% whereas the volume flow rate was increased from 0.50 L/min to 1.25 L/min. The experimental results showed that the nanofluid successfully has minimized the heat sink temperature compared to the conventional coolant. It was noticed also that the thermal entropy generation rate was reduced via using nanofluid instead of the normal water. Among the other functions of the nanofluid are to increase the frictional entropy generation rate and to drop the pressure which are insignificant compared to the normal coolant. Given the improved performance of the nanofluid, especially for high heat transportation capacity and low thermal entropy generation rate, it could be used as a better alternative coolant for the electronic cooling system instead of conventional pure water.     

Film cooling at hypersonic Mach numbers using forward facing array of micro-jets

Experimental investigations are carried out in the IISc hypersonic shock tunnel on film cooling effectiveness of a single jet (diameter 2 mm and 0.9 mm), and an array forward facing of micro-jets (diameter 300 μm each) of same effective area (corresponding to the respective single jet). The single jet and the corresponding micro-jets are injected from the stagnation zone of a blunt cone model (58° apex angle and nose radius of 35 mm). Nitrogen and Helium are injected as coolant gases. Experiments are performed at freestream Mach number 5.9, at 0° angle of attack, with a stagnation enthalpy of 1.84 MJ/kg, with and without injections. The ratios of the jet stagnation pressure to the freestream pitot pressure used in the present study are 1.2 and 1.45. Up to 50% reduction in surface heat transfer rate was observed with the array of micro-jets, compared to that of the respective single jet with nitrogen as the coolant, while the corresponding reduction was up to 37% for helium injection, with the schlieren flow visualizations showing no major change in the shock standoff distance, and thus no major changes in other aerodynamic aspects such as drag.