shariar motakef, matthew r. overholt
capesym, inc.
natick, massachusetts
removal of high heat fluxes is increasingly becoming the technology bottleneck in a wide variety of applications such as solid state lasers and diodes, telecommunication systems, and consumer electronics. fundamental considerations indicate that in nearly all of these applications liquid cooling is the only solution. none of the traditional solutions, however, are suitable for removal of heat fluxes larger than 500 kw/cm2.
in a series of upcoming articles we will discuss the use of micro-fabrication technologies to meet the challenges of high heat flux systems. a brief outline of these articles is given below. white papers on these technologies can be currently obtained by contacting [email protected]
in the first article we will discuss heat removal by an array of impinging micro-jets, with jet diameters as low as 300 microns. impingement is well known to provide very high single phase heat transfer coefficients. in general, heat transfer coefficient varies inversely with the jet diameter, motivating the move to smaller diameter jets. the problem with using a high density array of microjets (100’s per sq. cm.) is manifolding the return flow of the fluid after impingement, so that the neighboring jets do not interfere with each other. otherwise, the performance of a microjet array degenerates to that of a macro-jet. using the micro-fabrication technology liga, a novel 3d structure has been developed that provides for local exhausting of the microjets. a photograph of the patent-pending mjca (micro-jet cooling array) produced by international mezzo technologies is shown in figure 1. mjca is based on a hollow honey-comb structure, which is 1.0-3mm thick, contains a large array of circular flow channels to direct the microjets to the target, and is generally placed within a few hundred microns of the target surface. after impingement, the fluid is collected through small diameter holes (approx. 200 microns) located on the lower face of the honey comb, around the jet entrance. the fluid flows through the circular channels, impinges on the target as a submerged jet, and flows into the honeycomb towards an exhaust manifold. in this approach the microjets are effectively decoupled from each other and the high performance of a single microjet is achieved over large areas. we will discuss the fundamentals of micro-jet impingement cooling, and will report measured heat transfer coefficients as high as 500,000 w/m2k. this fielded system has been shown to easily handle extremely high heat fluxes, well above 1kw/cm2.
in the second article, we will discuss air-cooling with mjca. here, we will report on the measured performance of mjca-air, which is surprisingly high. we have measured heat transfer coefficients as high as 20,000 w/m2k, with no visible limits to achieving higher values. this capability allows us to extend air-cooling to the lower range of high heat flux applications, which otherwise would require liquid cooling.
in the third article we will focus on a central element of the liquid-cooling loop, namely the air-liquid heat exchanger needed to remove heat from the coolant to the environment. off the shelf, ultra-compact heat exchangers are still fairly bulky to be used in most of the high heat flux applications such as electronic cooling. we will discuss how micro-fabrication can be used to reduce the air-side thermal resistance in cross-flow heat exchangers, and will demonstrate the micro-channel cross-flow gas-liquid heat exchanger developed by international mezzo technologies, figure 2. by reducing the dimension of the air flow passage to 100’s of microns, very high heat transfer coefficients are achieved, allowing for significant reduction of the volume and weight of the heat exchanger. these fielded heat exchangers exhibit a one order of magnitude (x10) better performance (per unit volume) than the best off-the-shelf cross flow heat exchangers, without suffering from any appreciable increases in the pressure drops.
|
figure 1. a 10x10x1.7 mm mjca chip
figure 2. micro-fabricated cross-flow heat exchanger panel. the performance per volume of this panel is close to one order of magnitude better than the best off-the-shelf air-liquid heat exchangers.
about the author
dr. shariar motakef is the president of capesym, and its predecessor cape simulations. he has more than 25 years of experience in various applications of thermo-fluid engineering such as processing of electronic materials, crystal growth, and high heat flux systems.
|