Here is the abstract you requested from the Thermal_2018 technical program page. This is the original abstract submitted by the author. Any changes to the technical content of the final manuscript published by IMAPS or the presentation that is given during the event is done by the author, not IMAPS.
|Experimental Studies of Electronics Cooling Capabilities at High Altitude|
|Keywords: Altitude, Electronics cooling, Airflow|
|Cooling is one of the most costly factors in data centers regarding the overall power consumption. Over the past few decades, various cooling approaches have been raised to maintain facility��s equipment operation. However, with the increasing demands of data requirement, networking and computation, data centers are now facing substantial challenges in saving energy and reducing the power usage effectiveness (PUE). As a result, many innovative strategies start to bubble up to the surface to respond to those demands. Some approaches relate to the change of working gas density. That is to say, the locations of data center are not limited to somewhere close to sea level anymore. For example, the world��s highest data center �V ALMA correlator, which is located at Andes of northern Chile, more than 5000 meters above the sea. Another example is Facebook��s new data center at Los Lunas, NM, of which altitude is 1480 meters above sea level. In the case of ALMA correlator, the air density is only 60 % compared to the air density at sea level, which can enormously affect the cooling capabilities since the electronics cooling is related to the density of the working gas. Humidity and the types of working can affect both fan performance and electronics cooling capabilities as well. One example is Microsoft��s current project ��Natick��, a self-sufficient underwater data center. The emergence of the lights-out data center indicates that using working gas other than air could be a feasible approach. They sucked all the oxygen and moisture out, and filled the vessel with one atmospheric pressure of dry nitrogen to prevent corrosion. Therefore, the working gas for data center can be not limited to air in the future. Theoretically, the effect of working gas density on electronics cooling capabilities can be predicted by ideal gas law, fan law, and heat convection equation. However, few experiments have examined and compared to the theoretical models. For practical case studies, it will be necessary to verify the effect of change in gas density on electronics cooling performance rather than just using correction factor to predict the results. In this paper, we will discuss two topics: (a) Heat sink performance at a constant volumetric air flow rate under various pressure conditions. (b) Fan performance under various pressure conditions. A pressure chamber was used to simulate the environmental conditions of different altitude levels. All the experiments were conducted in the chamber. In the first part, we set up a heatsink and a fan connected serially in a duct. A heatsink was mounted to a dummy heater heated with constant power, and a fan was set at constant fan speed, while being monitored during the testing period. The correlation between case-to- ambient thermal resistance and pressure conditions was studied. In the second part, we took the heatsink out of the chamber, and an AMCA 210 airflow bench was placed into the chamber. This AMCA 210 airflow bench was served as a flow measurement standard to determine the aerodynamic characteristics of the fan. We recorded the chamber conditions and fan characteristics to determine the fan performance under various pressure conditions.|
|Hedy H.Y. Jhang, Engineer
Long Win Science and Technology Corporation
Taoyuan, Non US