Enhanced Cooling Efficiency of Immersion Systems Using Liquid Spinning Mechanisms and Mineral Oil
Sejun Gweon
Korea International School, Seongnam, South Korea
Publication date: November 20, 2025
Korea International School, Seongnam, South Korea
Publication date: November 20, 2025
DOI: http://doi.org/10.34614/JIYRC2025II05
ABSTRACT
This study investigates the impact of liquid spinning mechanisms on the cooling efficiency of mineral oil-based immersion systems. As air cooling approaches its thermal limits for high-performance electronics, immersion cooling presents a scalable alternative. Two identical tanks containing Apple M1 Mac Minis were tested: one passive and one with active magnetic stirring at 700 RPM. Temperature sensors and system monitoring tools recorded oil and CPU temperatures under a standardized 30-minute workload. Results showed that active stirring reduced peak oil temperature by 40% and CPU temperature by 10 °C compared to passive cooling. Additionally, the actively stirred system achieved significantly faster thermal recovery post-operation. These findings confirm that integrating a spinning mechanism enhances convective heat transfer, mitigates thermal stratification, and improves overall system efficiency. The method is cost-effective, scalable, and offers practical benefits for data centers and computational environments seeking sustainable thermal management. This study presents a proof-of-concept model demonstrating that mineral-oil immersion cooling with magnetic stirring can improve heat dissipation in small-scale computing systems.
This study investigates the impact of liquid spinning mechanisms on the cooling efficiency of mineral oil-based immersion systems. As air cooling approaches its thermal limits for high-performance electronics, immersion cooling presents a scalable alternative. Two identical tanks containing Apple M1 Mac Minis were tested: one passive and one with active magnetic stirring at 700 RPM. Temperature sensors and system monitoring tools recorded oil and CPU temperatures under a standardized 30-minute workload. Results showed that active stirring reduced peak oil temperature by 40% and CPU temperature by 10 °C compared to passive cooling. Additionally, the actively stirred system achieved significantly faster thermal recovery post-operation. These findings confirm that integrating a spinning mechanism enhances convective heat transfer, mitigates thermal stratification, and improves overall system efficiency. The method is cost-effective, scalable, and offers practical benefits for data centers and computational environments seeking sustainable thermal management. This study presents a proof-of-concept model demonstrating that mineral-oil immersion cooling with magnetic stirring can improve heat dissipation in small-scale computing systems.
