Cooling performance in a minichannel heat sink with different triangular pin-fins configurations

Kai Zhao, Xiaoxia Sun, Yuanqing Xia, Qiangqiang Li, Lili Shen, Min LinFrontiers in Energy Research 2023 · DOI

TL;DR

This paper investigates how the orientation (rotation angle) of triangular pin-fins within minichannel heat sinks affects the cooling of high-power electronics like IGBTs in electric vehicles. By using CFD simulations and the "Omega" vortex identification criterion, the researchers discovered that a $120^\circ$ rotation configuration significantly outperforms others by breaking the thermal boundary layer while minimizing efficiency-killing stagnation zones.

What Problem Does This Paper Address?

As electric vehicles (EVs) become more powerful, managing the heat generated by Insulated Gate Bipolar Transistors (IGBTs)—the "brains" of the power inverter—is critical for reliability. Traditional smooth minichannels are limited because a "boundary layer" of slow-moving fluid builds up along the walls, acting as an insulating blanket that prevents heat from reaching the coolant.

Engineers use pin-fins to break up this layer and increase surface area, but the specific geometry and orientation of these fins create a trade-off: they improve heat transfer (Nusselt number) but also increase flow resistance (friction factor). This paper specifically tackles the lack of quantitative understanding regarding how the rotation angle of triangular pin-fins influences vortex generation, stagnation zones, and overall cooling efficiency.

Methods

The authors employed a numerical simulation approach using Computational Fluid Dynamics (CFD):

  • Turbulence Modeling: They used the SST $k-\omega$ model, a standard but robust choice for capturing flow near walls and in separation regions.
  • Parametric Study: They designed 64 different configurations by varying the "front" and "rear" rotation angles of two triangular pin-fins from $15^\circ$ to $120^\circ$ in $15^\circ$ increments.
  • Vortex Identification: A highlight of the study is the use of the Omega ($\Omega$) criterion. Unlike older methods, $\Omega$ is better at identifying both strong and weak vortices, allowing the team to visualize exactly how the fins stir the fluid.
  • Validation: The model was validated against existing experimental data, showing a small error margin (under 9.1%), ensuring the simulations reflect physical reality.

Datasets

This is a simulation-based study rather than a machine learning paper, so it does not use a traditional training dataset. Instead, it generates a synthetic dataset of performance metrics across 64 geometric configurations. The "benchmark" used for comparison is a smooth minichannel without pin-fins, and the validation data was sourced from previously published experimental work by Chen et al. (2022b).

Key Insights and Ideas

  • The "Jet" Effect: The space between the triangular prism and the channel wall creates a high-velocity "jet." This jet impacts the side walls, effectively "scrubbing" away the boundary layer and boosting heat transfer.
  • The Stagnation Trap: The biggest enemy of efficiency is the stagnation zone—areas in corners or directly behind fins where fluid barely moves. In these zones, heat accumulates rather than being carried away.
  • Optimal Funneling: The $120^\circ$ rotation (where the triangle points "away" from the incoming flow) creates a funnel-like exit. This suppresses the backflow zone behind the prism and forces more intense mixing with the main flow.
  • Reynolds Number Sensitivity: The "best" configuration isn't universal; it changes with flow speed. As the Reynolds number (Re) increased, the optimal setup shifted from a $90^\circ–120^\circ$ configuration to a symmetrical $120^\circ–120^\circ$ one.

Key Results

  • Cooling Boost: The smooth minichannel had a maximum wall temperature 10 K higher than the pin-finned versions, proving the effectiveness of the fins.
  • Performance Evaluation Criterion (PEC): The PEC, which balances heat transfer gains against pressure drop penalties, peaked at a Reynolds number of approximately 553.
  • Winners and Losers: At Re = 663, the $120^\circ–120^\circ$ structure was the top performer (PEC ≈ 1.49). Conversely, the $60^\circ–60^\circ$ structure was the worst, largely due to an extended "slit exit" that created a massive, inefficient backflow zone.
  • Vortex Concentration: High-performing configurations showed vortices concentrated at the rear of the prism, promoting secondary flow that redistributed temperature more evenly across the channel.

Limitations and Future Work

  • Computational Cost: The authors note that simulating a large number of pin-fins is time-consuming. They suggest that Machine Learning could be used in future work to model these relationships and find optimal designs more quickly.
  • Flow Regime: The study focused on laminar and low-Reynolds number flows. Performance might differ significantly in fully turbulent regimes.
  • Dimensionless Metrics: There is currently no single dimensionless index to describe the "rotation angle" effect, making it difficult to generalize these findings to different scales without further simulation.

Summary updated

Abstract

With the continuous progress of automotive new energy technology, the motor has become an important part of the power system, and the heat dissipation of insulated-gate bipolar transistors (IGBT) determines the reliability of the power system. Minichannel structure can be added to the thermal management system of new energy vehicles to improve the heat transfer capacity. Due to the growth of the boundary layer in the smooth minichannel flow channel, the cooling performance improvement was limited. Pin-fins and rib structures were used to break the boundary layer and increased the heat transfer area to enhance the heat transfer capacity. In this study, a numerical simulation model of minichannel with triangular pin-fins with different rotation angles was established and calculated using the SST k-omega method. The temperature field, velocity field, pressure, and vortex distribution under different configurations were discussed in detail. The jet area formed by the prism wall and the side wall of the minichannel would impact the wall and reduce the growth of the boundary layer. However, the stagnation area generated in the center and corner will reduce the improvement of heat transfer capacity. The thermo-hydraulic characteristics of different configurations at different Reynolds numbers (Re), such as Nusselt number (Nu), Darcy friction resistance coefficient ( f ), and performance evaluation criterion (PEC), were analyzed. As Re increased, the best and worst configurations changed, the best configuration changed from the 90°–120° structure to the 120°–120° structure, and the worst configuration changed from the 75°–60° to the 60°–60° structure. When the Re = 663, the influence of the front and rear rotation angle on the cooling performance was explored. When the rotation angle was closer to 60°, the cooling performance of the minichannel was better. And the closer the rotation angle was to 120°, the cooling performance was better. This has a reference effect on the design of minichannel heat sinks.
BibTeX
@inproceedings{zhao2023cooling,
  author = {Kai Zhao and Xiaoxia Sun and Yuanqing Xia and Qiangqiang Li and Lili Shen and Min Lin},
  title = {{Cooling performance in a minichannel heat sink with different triangular pin-fins configurations}},
  booktitle = {Frontiers in Energy Research},
  year = {2023},
  doi = {10.3389/fenrg.2023.1087501},
}
Figure