Numerical investigation of CPU cooling with micro-pin–fin heat sink in different shapes


KOCA F., Güder T. B.

European Physical Journal Plus, cilt.137, sa.11, 2022 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 137 Sayı: 11
  • Basım Tarihi: 2022
  • Doi Numarası: 10.1140/epjp/s13360-022-03489-7
  • Dergi Adı: European Physical Journal Plus
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, INSPEC
  • Sivas Cumhuriyet Üniversitesi Adresli: Evet

Özet

© 2022, The Author(s), under exclusive licence to Società Italiana di Fisica and Springer-Verlag GmbH Germany, part of Springer Nature.The fact that today's computers transmit too much information at effective speeds causes high heat concentrations. These heat concentrations increase the temperature of the computer processor and cause it to deteriorate its consistency and reduce its performance. In the scope of the study offered, the heat transfer performance of the circular, square and triangular fins placed on the heat source of the models created were examined in the range 500 ≤ Re ≤ 10,000. A constant heat flux of 2000 W/m2 was applied to the bottom surface of the heat source. The inlet temperature of the air used as an operating fluid was 300 K, and the atmospheric pressure was accepted as an outlet boundary condition. Analyses were made with Ansys Fluent program. In the study, 4 different models were used as finless, circular finned, square finned and triangular finned models. Nusselt numbers, temperature distributions, pressure drops, friction factor and thermal performance factor values were examined and analyzed for 7 different Reynolds number. As a result, it was revealed that the best work in terms of heat transfer in the examined models was between Re = 5000 and Re = 6500. At Re = 5000, the PEC number, which is the thermal performance factor of the circular finned model, was 17.96% and 18.30% higher, respectively, than the PEC numbers of the square and triangular finned models.