Research PaperCharacterization of a dual taper thermosiphon loop for CPU cooling in data centers
-
Add time:08/27/2019 Source:sciencedirect.com
The inefficient cooling processes in data centers consume a large amount of energy making them very expensive. This study is focuses on using a thermosiphon loop as an efficient replacement of currently used air cooling and water cooling techniques for dissipating large heat loads from the CPUs. A symmetric dual taper configuration is introduced in the evaporator to generate a highly efficient and stable two-phase flow in the system using HFE7000. Three different taper angles in the evaporator – 2°, 2.5°, and 3° – are studied, and the respective heat transfer performances are evaluated. The evaporator has 200 µm square microchannels machined on a 34.5 mm × 32 mm copper heat sink. Prior to CPU testing, the dual taper evaporator performance is evaluated in a benchtop thermosiphon loop in which the evaporator is heated by electric heaters. The loop is able to dissipate 280 W without reaching the critical flux point with a heat transfer coefficient of 26 kW/m2 °C. The thermosiphon loop is then tested for cooling a data center server with an i7-930 processor with thermal design power (TDP) of 130 W. The cooling performance of the thermosiphon loop is compared with commercial air based and water based coolers in both server and benchtop configurations.
We also recommend Trading Suppliers and Manufacturers of CPU 57 (cas 132836-34-1). Pls Click Website Link as below: cas 132836-34-1 suppliers
Prev:ppXen: A hypervisor CPU scheduler for mitigating performance variability in virtualized clouds
Next:Novel heat pipe radiator for vertical CPU cooling and its experimental study) - 【Back】【Close 】【Print】【Add to favorite 】
- Related Information
- Parallel ant colony optimization on multi-core SIMD CPUs08/30/2019
- Experimental study on heat sink with porous copper as conductive material for CPU cooling08/29/2019
- Novel heat pipe radiator for vertical CPU cooling and its experimental study08/28/2019
- ppXen: A hypervisor CPU scheduler for mitigating performance variability in virtualized clouds08/26/2019
- Short noteWang–Landau sampling: Saving CPU time08/25/2019


