Heat removal with microchannel heat sinks
With component heat dissipation levels reaching 500 W/cm2 and beyond, conventional air cooling systems are inadequate for removing excess heat. Research has intensified toward developing more innovative chip cooling techniques. The ultimate goal is to reduce thermal resistance from the chip junction to ambient, and keep the chip’s junction temperature as low as possible.
For high performance CPUs, graphics cards, power amplifiers and other devices, air-cooling has proven ineffective at dissipating high heat fluxes. Heat transfer methods such as heat pipes, vapor chambers, nanomaterials, liquid cooling and miniature refrigeration systems have been attracting more interest.
Liquid-cooled microchannel heat sinks and coolers have been shown to be a very effective way to remove high heat load. A large heat transfer coefficient can be achieved by reducing the channel hydraulic diameter. In a confined geometry the small flow rate within microchannels produces laminar (smooth) flow, which results in a heat transfer coefficient inversely proportional to the hydraulic diameter. In other words, the narrower the channels in the heat sink, the higher the heat transfer coefficient.
Figure 1. Microchannel Cooler Showing Heat Exchanger Zones, [1]
Forcing a fluid through a greater number of small channels vs. a smaller number of larger channels increases the level of heat transfer from the hot source, e.g. chips, cards, into the flowing liquid. There are more channels for a small hydraulic diameter than large ones. So the heat sink with a smaller hydraulic diameter has better overall thermal performance.
The use of microchannels as a viable cooling solution was first proposed in 1981 by Tuckerman and Pease, who designed and tested an integral, water-cooled heat sink by etching microscopic channels 50 µm wide and 300 µm deep on the silicon substrate [2]. They reported achieving a high heat flux of 790 W/cm2 with a temperature rise of 71°C above the inlet water temperature. Tuckerman’s work was well received by the electronics community, and many extensive studies have since been conducted on different aspects of microchannels in electronic cooling.
The concept behind microchannels relates to the definition of the Nusselt number, Nu, which relates to the heat transfer coefficient as follows: [Equ 1]
Where Kf is the fluid thermal conductivity and Dh is the hydraulic diameter. If the flow is laminar and fully developed because of the small hydraulic diameter, and the Nusselt number is a constant, assuming the classical channel flow, the small Dh of microchannels in the denominator should significantly enhance the heat transfer coefficient. For a fully developed laminar (smooth) flow in a square channel with constant wall temperature or constant wall heat flux, the Nusselt number is a constant.
The negative aspect of this concept is an increase in pressure drop. Assuming laminar flow: [Equ 2]
Where K is the loss coefficient and G is the volumetric flow rate. The above pressure drop equation shows that the smaller the hydraulic diameter, the higher the pressure drop (decrease in pressure from one point in the microchannel to another point downstream).
Qu and Mudawar [3] measured the critical heat flux for a water-cooled microchannel heat sink containing 21 parallel channels. The microchannel array was tested in a module as shown in Figure 2. Thermocouples were installed along the channel to monitor temperature. Deionized water was used as the working fluid.
Figure 2. Microchannel Heat Sink Test Module [3]
The authors observed that the heat flux increases rapidly with small wall temperature increments after the boiling starts. When the wall heat flux reached the CHF, the wall temperature began to jump.
Even though the use of microchannel devices seems very promising, they require a significant amount of power to push the fluid through the channels in high heat flux applications. One solution is to use convective boiling heat transfer and two-phase flow in the microchannels. Use of boiling heat transfer could improve the efficiency of microchannels in two ways: it reduces the pumping power required to push the fluid through the channels, and at its time of phase change, the boiling coolant absorbs energy from the hot surface of the microchannel heat sink, substantially increasing the heat transfer coefficient.
The heat transfer ability attained with two-phase flow and boiling inside microchannels is limited only by its critical heat flux (CHF). The CHF of boiling at a surface refers to the heat flux at which the surface heat transfer coefficient suddenly drops. Exceeding the CHF will lead to a sudden jump in surface temperature which will result in the failure of an electronic component.
While using microchannels to cool electronics is attractive because of their very high heat transfer coefficient, their use and implementation is very challenging. Factors such as cost, manufacturing, pump selection, filtering requirements to prevent channel clogging, and space constraints must be evaluated before one commits to using microchannels in a system. The research in this area is ongoing and the implementation of this concept will become more widespread once the practical difficulties mentioned above are resolved.
Bahman Tavassoli, PhD, Advanced Thermal Solutions, Inc
References:
- Tuckerman, D. and Pease, R., High-Performance Heat Sinking for VLSI, IEEE Electron Device Letters, May 1981.
- Qu, W. and Mudawar, I., Measurement and Correlation of Critical Heat Flux in Two-Phase Microchannel Heat Sinks, Int. Journal Heat and Mass Transfer, 2004.
- Marston, K., Gaynes, M., Bezama, R. and Colgan, E., A Practical Implementation of Silicon Microchannel Coolers, Electronics Cooling Magazine, November 2007.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
