Assessing Capabilities and Limitations of Air- and Liquid-Cooling for Low Profile Servers

Author(s):  
Albert Chan ◽  
Jie Wei

As microprocessor power exceeds 100W, adequate heat removal by convective air-flow through a heatsink increasingly becomes more challenging. This is especially true for low-profile servers with very limited volume for air-flow. It is therefore useful to have an idea of the limitations of air-cooling for such servers. In this paper, three case studies serve to illustrate the capability of typical air-cooled solutions for low-profile servers. These studies show the inherent limitations of air-cooled solutions for volume-constrained computer systems. Liquid-cooling has been used in cooling mainframe processors packaged in MCM format. Its use in low-cost servers is extremely limited. This paper will deal with issues that hinder widespread application of liquid-cooling in commercial servers. The most important issue is cost, followed by lack of commodity components suitable for liquid-cooled systems. One method to reduce cost is to use fabricate the cold plate using heatsink manufacturing techniques. Case studies are presented to show liquid-cooling with these lower cost cold plates can provide performance that exceeds air-cooling solutions. Finally, suggestions are offered for facilitating the introduction of liquid-cooling systems for future low-profile servers.   This paper was also originally published as part of the Proceedings of the ASME 2005 Heat Transfer Summer Conference.

2020 ◽  
Vol 15 (1) ◽  
pp. 73-86
Author(s):  
Masa Noguchi ◽  
Koon Beng Ooi

ABSTRACT Fuel poverty is one of the global concerns affecting not only users' financial capacity or affordability for maintaining housing operation but also the occupants' health and wellbeing. Space heating and cooling require a relatively large amount of domestic energy use in housing. Therefore, this study was formed with the aim to propose an innovative approach to utilising free, clean renewable sources of energy applicable to the space heating and cooling of housing in both cold and hot regions. Accordingly, housing test facilities based in Melbourne, Australia, and Kuching, Malaysia, were selected and used for this study that examined the thermal performance of a proposed ‘hydronic radiator’ (HR) system through simulation and onsite measurements. The geothermal heat capacity of a ‘vertical ground heat exchanger’ (VGHE) installed in the house in Melbourne was examined previously by the authors and the VGHE measured data was also applied to this HR performance simulation. The water that circulates through the HRs is heated by sunlight and VGHE or cooled by night sky radiation. This study drew conclusions that the sole utilisation of renewable sources through these proposed HR space heating and cooling systems can provide thermally accessible or comfortable indoor living environments in both heating or cooling dominant regions. Thus, fuel poverty issues may be alleviated through HR system application. The HRs can remove a ‘sensible’ portion of metabolic heat, but they cannot effectively contribute to the ‘latent’ heat removal. Thus, the future potential use or effect of ‘flow-through’ HRs, which are integrated into a underfloor air distribution (UFAD) plenum, was also dsicussed in this study. In the test house located in Melbourne, the flow-through HR UFAD system is currently under development. Therefore, the performance will be measured once the system has come into operation for further testing.


Open Physics ◽  
2020 ◽  
Vol 18 (1) ◽  
pp. 241-254
Author(s):  
Jonathan Rodriguez Andrade ◽  
Ernst Kussul ◽  
Tetyana Baydyk

AbstractIn this study, we propose a new design for a microchannel filter. The closed indoor environments with which we interact daily are sources of diseases for the respiratory system of human beings. Recommendations for the design of microchannel filters for indoor air purification are proposed, implementing low-cost microequipment technology (MET) for the manufacture of the elements that constitute a microfiltration system. For the microchannel filter production, we proposed to use MET, which is a miniaturization technology and can reduce manufacturing costs. The microchannel filter was 3.75 cm in radius with a thickness of 3 mm. It had a triangular profile and a helical trajectory. It was designed, manufactured, and tested for two profile dimensions. The main purpose was to reduce the pressure drop of the air flow through the filter. We described the air flow simulation for the microchannel filter using SolidWorks. A prototype microchannel filter was constructed, which underwent manufacturing tests. It is possible to clean the microchannel using water flow, which allows us to maintain the filtration quality within an optimum range of contaminant removal.


Author(s):  
Yogen Utturkar ◽  
Mehmet Arik ◽  
Tunc Icoz

Thermal management is currently one of the key limitations in the design of electronic systems. Parallel to the advancements in the electronics industry and increase in power dissipation the development of effective, low-cost, compact heat removal solutions become extremely critical to ensure a failsafe and reliable operation. While liquid cooling is poised to provide the cooling capability for next generation electronics, its use in present-day products is less prevalent due to risks associated with condensation, leakage, and pumping power. Consequently, air-cooling strategies still continue to vie for near-term cooling needs in the electronic industry. In cohort with these trends, an advanced air-cooling solution in form of a synthetic jet assisted heat sink has been investigated in the present study. The study focuses on key design aspect of the heat sink fin design, synthetic jet design and characterization, and the interaction of unsteady air jets with the heat sink fins. Numerical simulations are employed to investigate 3D unsteady flow dynamics and experimental setup is designed and built for validation. The paper systematically presents the design trade-offs associated with the number of jets in the thermal solution and the jet driving conditions (voltage and frequency), in terms of the thermal performance and the cost. Overall, the synthetic jet integrated heat sink has demonstrably been shown to dissipate up to 4.7 times better than conventional natural convection heat sink with a COP value of greater than 40 within a volume of 25 in3.


Author(s):  
Taeyoung Han ◽  
Bahram Khalighi ◽  
Erik C. Yen ◽  
Shailendra Kaushik

Abstract The Li-ion battery operation life is strongly dependent on the operating temperature and the temperature variation that occurs within each individual cell. Liquid-cooling is very effective in removing substantial amounts of heat with relatively low flow rates. On the other hand, air-cooling is simpler, lighter, and easier to maintain. However, for achieving similar cooling performance, a much higher volumetric air flow rate is required due to its lower heat capacity. This paper describes the fundamental differences between air-cooling and liquid-cooling applications in terms of basic flow and heat transfer parameters for Li-ion battery packs in terms of QITD (inlet temperature difference). For air-cooling concepts with high QITD, one must focus on heat transfer devices with relatively high heat transfer coefficients (100–150 W/m2/K) at air flow rates of 300–400 m3/h, low flow induced noise, and low-pressure drops. This can be achieved by using turbulators, such as delta winglets. The results show that the design concepts based on delta winglets can achieve QITD of greater than 150 W/K.


Author(s):  
Богдан Сергійович Портной

It is proposed to determine the rational velocity of air flow through the air coolers of a stepped a waste heat-recovery absorption-ejector chiller utilizing the heat of exhaust gases of a gas turbine unit to cool the air at the inlet, by computer simulation of air processes processing. Whereas the result of air cooling depends on the efficiency of the air coolers at the inlet of the gas turbine unit, it is proposed to determine it as an increase in the specific fuel economy, which consider both the cooling depth (the magnitude of the temperature decrease) of the air and the air resistance of the air cooler, which significantly affects the efficiency of operation cooling devices. On the example of air cooling at the inlet of a gas turbine unit has been analyzed the value of specific fuel economy by cooling the air at the inlet to a temperature of 10 °C in a two-stage absorption-ejector chiller, depending on the rational airflow rate through the cooling units (air coolers). The efficiency of the air coolers at different air flow rates has been analyzed.It is shown that proceeding from the different rate of increment in the specific fuel economy caused by the change in the rational velocity of air flow through the air coolers of chillers, it is necessary to choose a design (rational) the rational velocity of air flow that ensures the achievement of a maximum or close to the maximum increase in the specific fuel economy at relatively high rates increments. In order to determine the established the rational velocity of air flow through the air coolers, which provides the maximum increment of the specific fuel economy, the dependence of the increment of the specific fuel economy on the airflow velocity is analyzed. Based on the results of modeling air cooling processes at the inlet of the gas turbine unit, using software from firms that produce heat exchange equipment, it is proposed to determine the rational velocity of air through the air coolers, which ensures a close maximum specific fuel economy at relatively high rates of its increment


Author(s):  
Kim Yeow ◽  
Ho Teng ◽  
Marina Thelliez ◽  
Eugene Tan

A comparative study is conducted on the thermal behavior of three Li-ion battery modules with two cooled indirectly with air and one cooled indirectly with liquid. All three battery modules are stacked with the same twelve 8Ahr high-power pouch Li-ion battery cells. Heat generated from the cells is dissipated through 1-mm thick aluminum cooling plates sandwiched between two cells in the module. Each of the cooling plates has an extended surface for heat dissipation. The battery heat is dissipated through the cooling fins exposed in air flow channels in the case of air cooling, and through the extended cooling plate surfaces that are in contact with a liquid-cooled cold plate in the case of liquid cooling. The cell temperatures are analyzed using a simplified Finite Element Analysis (FEA) model for battery cooling. Simulation results show that with air cooling channels structured similar to that of compact heat exchangers, the air utilization and effectiveness of air cooling can be improved significantly. With proper design of the air cooling channels (i.e. with fin inserts in the air flow channels), indirect air cooling could reach a cooling condition comparable to that of indirect liquid cooling and obtain a higher gravimetric energy density with the same cooling-related parasitic volume in the battery system as long as the cell heat rejection is < 10 W/cell.


Author(s):  
Albert Chan ◽  
Jie Wei

Feasibility study on alternative cooling methods to air-cooling with heat sinks is provided in this paper. The study focuses on cooling of 64-bit microprocessor at 80nm technology node with projected heat dissipation of 200W. An example was presented to illustrate limitation of air-cooling for the 200W microprocessor using an all-Cu heat sink with tall fins. Three alternatives to air-cooling were studied in this work: liquid cooling, two-phase convective flow cooling and refrigeration cooling. Thermodynamic analysis was used to estimate operating conditions and fluid flow rates for each alternative. The information provides a preliminary basis for assessing capabilities and weaknesses among alternatives. Liquid and two-phase cooling simply transfer heat from high to low temperature. In contrast, refrigeration cooling operates as a heat pump, moving heat from low to high temperature. Refrigeration cooling offers capability to cool microprocessor (LSI) chip to temperatures below ambient or freezing. The drawback is more heat must be removed from the system. Liquid cooling operates at close to ambient pressure, while two-phase and refrigeration cooling operate at higher pressures. Challenges to implementation of all three alternatives include availability of low cost, miniature components (pumps or compressors, heat exchanger and condenser), designing for redundancy (or reliability) and ease of installation and field service. In terms of component availability and cost, liquid cooling is preferred choice, followed by two-phase and refrigeration cooling.


Author(s):  
Tianyi Gao ◽  
Bahgat Sammakia ◽  
Roger Schmidt

The heat dissipated by data center equipment continues to increase due to the growing server power year by year. Thermal management continually becomes a challenging problem both from the standpoint of maintaining the equipment within a proper and reliable operation environment, as well as from a total energy cost perspective. With the total heat load increasing, as well as local high density heat fluxes, more cooling air flow rates may be needed by the racks which cannot be sufficiently provided by CRAC units. This may occur during peak periods of operation and may also occur in isolated aisles in the data center. Under such circumstances, just changing the air flow rates to specific aisles may cause unanticipated hot air recirculation into cold aisles which may happen during the transients that arise as the load and air flow rates are changing; henceforth, it may even occur in steady state. This could reduce the reliability of the data center equipment. A hybrid cooling strategy, which uses water to air heat exchangers located at the rear of specific racks to assist the air cooling, can be very effective in solving this problem by removing heat from the rack before it is exhausted into the room. This study describes the impact of this heat exchanger strategy on the thermal management of a data center. In this analytical study, the heat removal percentage and working efficiency of the heat exchanger are presented under different water flowrates and water inlet temperatures. Different case studies are conducted, showing that this method can significantly assist the data center air cooling under scenarios such as, power increases or CRAC unit failure. Detailed function of the heat exchanger on impacting the rack inlet temperatures are also discussed in this paper.


Sign in / Sign up

Export Citation Format

Share Document