0
Research Papers

Analysis of Parallel Microchannels for Flow Control and Hot Spot Cooling

[+] Author and Article Information
Stephen A. Solovitz

School of Engineering and Computer Science,
Washington State University,
Vancouver 14204 NE Salmon Creek Ave.,
Vancouver, WA 98686
e-mail: stevesol@vancouver.wsu.edu

Manuscript received December 11, 2012; final manuscript received March 6, 2013; published online September 27, 2013. Assoc. Editor: Mehmet Arik.

J. Thermal Sci. Eng. Appl 5(4), 041007 (Sep 27, 2013) (13 pages) Paper No: TSEA-12-1227; doi: 10.1115/1.4024021 History: Received December 11, 2012; Revised March 06, 2013

Microchannel heat transfer is commonly applied in the thermal management of high-power electronics. Most designs involve a series of parallel microchannels, which are typically analyzed by assuming a uniform flow distribution. However, many devices have a nonuniform thermal distribution, with hot spots producing much higher heat fluxes and temperatures than the baseline. Although solutions have been developed to improve local heat transfer, these are advanced methods using embedded cooling devices. As an alternative, a passive solution is developed here using analytical methods to optimize the channel geometry for a desired, nonuniform flow distribution. This results in a simple power law for the passage diameter, which may be useful for many microfluidic systems, including electronics cooling devices. Computational simulations are then applied to demonstrate the effectiveness of the power law for laminar conditions. At low Reynolds numbers, the flow distribution can be controlled to good accuracy, matching the desired distribution to within less than 1%. Further simulations consider the control of hot spots in laminar developing flow. Under these circumstances, temperatures can be made uniform to within 2 °C over a range of Reynolds numbers (60 to 300), demonstrating the capability of this power law solution.

FIGURES IN THIS ARTICLE
<>
Copyright © 2013 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 2

Schematic of flow rates for microchannel structure with five channels, where the flow is biased towards channel 2

Grahic Jump Location
Fig. 1

Schematic of a typical microchannel structure with five channels [18]

Grahic Jump Location
Fig. 3

Schematic of a microchannel structure with five channels of varying diameter. In this image, channel 2 is larger to bias the flow.

Grahic Jump Location
Fig. 4

Schematic of computational model, which was based on the topology depicted in Fig. 3. The model shown here had a biased flow in the first channel.

Grahic Jump Location
Fig. 6

Channel flow rates Qi, normalized by the design flow rate in each of the unbiased channels Qun for systems biased towards each of the five channels. The designs used the approximate correlation of Eq. (8), with a target bias k of five times the unbiased level. The Reynolds number ReDh was 50.

Grahic Jump Location
Fig. 7

Channel flow rates Qi, normalized by the total flow rate Qtot for a randomly selected, nonuniform distribution. The design used the approximate correlation of Eq. (8). The Reynolds number ReDh was 50.

Grahic Jump Location
Fig. 8

Schematic of the modified computational model, which included five small planar heaters to produce local hot spots. The model shown here had a biased flow in the third channel. (a) Top view; (b) isometric view.

Grahic Jump Location
Fig. 9

Maximum heater temperatures (in °C) for laminar developing flow at ReDh = 60 and 300. The third heater had a 2 W output, while the other four heaters generated a baseline level of 1 W.

Grahic Jump Location
Fig. 10

Maximum heater temperature rise (in °C) due to convection for a design biased towards the third heater. The biased heater power level was varied to examine the effectiveness of the design. The Reynolds number ReDh was 300.

Grahic Jump Location
Fig. 5

Channel flow rates Qi, normalized by the design flow rate in each of the unbiased channels Qun for a system biased towards the first channel. The model used the exact correlation of Eq. (7), with a target bias, k, of five times the unbiased level. Reynolds numbers ReDh varied from 5 to 500.

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In