A predictive model for fin array boiling heat transfer performance under two-phase immersion cooling
Abstract
Heat sinks with extended surface area can enhance the pool boiling performance for two-phase immersion cooling of electronic devices. However, even for a simple heat sink geometry with an array of longitudinal fins, predicting performance during boiling is challenging. Individual fins can be modeled as extended surfaces with a superheat-dependent heat transfer coefficient based on flat surface boiling performance, but this approach fails in the limit of closely spaced fin arrays because of fin-vapor interactions. Recent studies have identified the fluid capillary length Lb as the key length scale at which such vapor confinement effects must be considered to accurately predict the performance of finned heat sinks in pool boiling. In this study, we propose a predictive model for the pool boiling heat transfer performance of a fin array heat sink, valid across all dimensions above and below the capillary length. The model follows a fin analysis with a constant base superheat and a heat transfer coefficient that is dependent on local fin surface superheat. This fin-specific function hfin(ΔT) is determined from an empirically calibrated function hflat(ΔT) obtained from a flat surface boiling test with the same surface characteristics. For fin spacing above the capillary length (S > Lb), hflat(ΔT) can be directly applied as hfin(ΔT). Whereas for fin spacing below the capillary length (S < Lb), vapor confinement effects enhance boiling heat transfer at lower heat fluxes and deteriorate performance at higher heat fluxes; these confinement effects are incorporated as mechanistic modifications to the function hflat(ΔT). Boiling tests are conducted to validate the prediction accuracy of the model. In particular, the vapor confinement effects on fin array heat sinks with tight fin spacing are well-captured by the model. Furthermore, the effects of fin height and spacing on the boiling performance of fin array heat sinks are explored by applying the experimentally validated model. The predictive model provides a tool for designing and optimizing the heat sinks for two-phase immersion cooling applications and provides insight into the dimensional scaling effects on the boiling performance of fin arrays.
Date of this Version
12-7-2024
Published in:
Y. Huang, B. Sarma, and J.A. Weibel, A predictive model for fin array boiling heat transfer performance under two-phase immersion cooling, International Journal of Heat and Mass Transfer 239, 126513, 2025.
Link to original published article:
https://doi.org/10.1016/j.ijheatmasstransfer.2024.126513
Comments
This is the author-accepted manuscript of Y. Huang, B. Sarma, and J.A. Weibel, A predictive model for fin array boiling heat transfer performance under two-phase immersion cooling, International Journal of Heat and Mass Transfer 239, 126513, 2025. Copyright Elsevier, it's made available here CC-BY-NC-ND, and the version of record is available at DOI: 10.1016/j.ijheatmasstransfer.2024.126513