Processor coolant design and analysis

With the development of technology and the rapid increase in usage areas, heating problems in electronic devices have become even more important. In cases where passive cooling and active cooling processes are not sufficient to improve overheating in electronic devices, high temperature formation is prevented by liquid cooling if appropriate. In this thesis, the effect of nanofluid use on heat transfer to improve the thermal performance of liquid cooling systems was investigated with computational fluid dynamics from numerical methods. For different cooler designs, both single-phase and multi-phase modeling methods were used in fluid modeling in the study involving forced convection and heat transfer in laminar and turbulent flow. The mathematical models used were validated on different experimental and numerical studies. It is possible to numerically model nanofluids as single phase and multiphase. In the single-phase modeling approach, the nanofluid material was modeled as mixed with H_2 O at certain concentration ratios. Thermophysical properties of Al〗_2 O_(3 ) were calculated by adhering to certain laws. Single-phase modeling was performed for pure H_2 O, 〖%1 Al〗_2 O_(3 ), 〖%5 Al〗_2 O_(3 )nanofluids at different Reynolds numbers. Analyzes were performed at Reynolds values of 500, 750, 1000, 2500, 5000 and 7000, and the increase in Reynolds value decreases the maximum temperature read at the outlet. In the multiphase modeling method, the nanofluid 〖 Al〗_2 O_(3 ), was defined as a separate phase suspended in different sizes into the base fluid H_2 O, and two different models were used. The use of nanofluids increased the thermal conductivity in both methods compared to the use of pure H_2 O. At the same time, the use of nanofluids at 5% concentration gave better results than 1% concentration. When the temperature distributions are examined, the Mixture model, one of the single-phase and multi-phase modeling methods, gives very similar distributions. The effect of nanofluids on thermal performance varies depending on many parameters. One of the most important of these is the size of the suspended nanoparticle. In this study, different particle sizes, 20 nm, 30 nm and 40 nm, were studied in both Mixture and Eulerian modeling approaches. Two different design coolers were used in the studies. In these designs, the flow is distributed in parallel and serpentine. Flow distribution and pressure drop also had an effect on thermal performance. Temperatures were read lower in the serpentine block, which provides a more regular flow distribution. All studies were carried out using ANSYS Fluent, a CFD software.