Numerical simulation on the flow and heat transfer characteristics in dimple/protrusion enhanced air handling unit in data center

Zixing Wang, Hao Ding, Le Lei, Nan Li, Wen Quan Tao*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)

Abstract

Air handling unit (AHU) is the heat exchanger used for data center cooling. This article developed a detailed dimple/protrusion enhanced AHU modeling process and studied the flow and thermal performance of AHU in the Reynold number range from 5 × 103 to 3.53 × 104. Elliptic cylindrical dimples (ECD) and spherical crown dimples (SCD) are applied to enhance heat transfer. The entire channel (EC) and typical unit (TU) simulation domains are compared. The EC simulation domain is more reliable as the TU causes the Nu and f prediction relative deviations as large as 20.53% and 24.03%, respectively. The flow patterns in the channels are analyzed. The results show that the bigger SCD depth and smaller ECD depth make the mainstream bends closer to the “S” shape and the v velocity near the dimple/protrusion wall larger. Also, the second flow vortex distribution pattern becomes more complicated, and the velocity gradient near the wall is increased. These flow patterns enhance heat transfer. The dimpled surface has smaller local convective heat transfer coefficient compared with the protrusion surface in the same channel. With the decrease of ECD depth and the increase of SCD depth, the area-average convective heat transfer coefficient is increased.

Original languageEnglish
Article number107710
JournalInternational Communications in Heat and Mass Transfer
Volume157
DOIs
Publication statusPublished - Sept 2024
Externally publishedYes

Keywords

  • Air handling unit
  • Data center
  • Dimple/protrusion channel
  • Heat transfer enhancement
  • Turbulent flow

Fingerprint

Dive into the research topics of 'Numerical simulation on the flow and heat transfer characteristics in dimple/protrusion enhanced air handling unit in data center'. Together they form a unique fingerprint.

Cite this