1. Field of the Invention
The subject invention relates to a cooling assembly for cooling an electronic device such as a microprocessor or a computer chip.
2. Description of the Prior Art
These electronic devices generate a high concentration of heat, typically a power density in the range of 5 to 35 W/cm2. Accordingly, research activities have focused on developing more efficient cooling assemblies capable of efficiently dissipating the heat generated from such electronic devices, while occupying a minimum of space.
A forced air cooling assembly typically includes a heat exchanger and a heat sink, and cools the electronic device by natural or forced convection cooling methods. The electronic device is attached to the heat sink and transfers heat thereto. The heat exchanger typically uses air to directly remove the heat from the heat sink. However, air has a relatively low heat capacity. Such forced air cooling assemblies are suitable for removing heat from relatively low power heat sources with a power density in the range of 5 to 15 W/cm2. However, the increased computing speeds have resulted in a corresponding increase in the power density of the electronic devices in the order of 20 to 35 W/cm2, thus requiring more effective cooling assemblies.
In response to the increased heat produced by the electronic devices, liquid-cooled cooling assemblies, commonly referred to as liquid cooled units (“LCUs”) were developed. The LCUs employ a heat sink in conjunction with a high heat capacity cooling fluid, like water or water-glycol solutions, to remove heat from these types of higher power density heat sources. One type of LCU circulates the cooling fluid through the heat sink to remove the heat absorbed from the heat source affixed thereto. The cooling fluid is then transferred to a remote location where the heat is easily dissipated into a flowing air stream with the use of a liquid-to-air heat exchanger and an air moving device such as a fan or a blower. These types of LCUs are characterized as indirect cooling units since they remove heat form the heat source indirectly by a secondary working fluid. Generally, a single-phase liquid first removes heat from the heat sink and then dissipates it into the air stream flowing through the remotely located liquid-to-air heat exchanger. Such LCUs are satisfactory for a moderate heat flux less than 35 to 45 W/cm2.
The U.S. Pat. No. 5,304,846, issued to Azer et. al., and the U.S. Pat. No. 6,422,307, issued to Bhatti et. al., each disclose a typical heat sink assembly used in a LCU. The heat sink assemblies include a base plate with a plurality of fins having smooth sidewalls extending upwardly from the base plate. In operation, the base plate absorbs the heat from the electronic device and transfers the heat to the fins. A cooling fluid flows past the smooth walled fins, drawing the heat from the fins, thereby removing the heat from the heat sink. The flow of cooling fluid may be directed parallel to the fins or impinged thereon.
The U.S. Pat. No. 5,019,880, issued to Higgins, discloses a heat sink that includes a circular base with a central flow diverter having a conical shape extending upwardly from the base. A plurality of planar fins is disposed radially about the circumference of the flow diverter and extend upwardly from the base to a lid. An inlet is disposed above the lid for directing a flow of cooling fluid perpendicularly onto the flow diverter. The flow of cooling fluid then circulates radially outward to the outer periphery of the base through a plurality of flow channels defined between the planar fins.
The amount of heat transferred between the fins and the cooling fluid is dependent on a heat transfer coefficient therebetween. The heat transfer coefficient is dependent on a thermal boundary layer, which is a layer of stagnant cooling fluid adjacent each of the fins. The thermal boundary layer acts as an insulator, limiting the heat transfer coefficient. As the cooling fluid flows past the fins uninterrupted, the thermal boundary layer becomes thicker, decreasing the heat transfer coefficient and thereby decreasing the effectiveness of the heat sink assembly. Additionally, the amount of heat stored in each of the fins varies according to the distance between each of the fins and the heat source, with the highest concentration of heat occurring directly above the heat source, with the fins disposed farther from the heat source absorbing less heat. Therefore, the heat transfer to the cooling fluid at the outer periphery of the heat sink is less efficient than the heat transfer to the cooling fluid directly above the heat source.
The subject invention provides a heat sink assembly for removing heat from an electronic device. The heat sink assembly includes a base having a top surface and a lid having a bottom surface in spaced relationship with and parallel to the top surface of the base. A flow diverter extends upwardly from the top surface of the base toward the bottom surface of the lid. The lid defines an inlet aligned with the flow diverter for impinging the flow of cooling fluid on the flow diverter. A spiral wall extends between the top surface of the base and the bottom surface of the lid and is disposed in an increasing spiral from the flow diverter to define a spiral channel having an outlet for directing the flow of cooling fluid radially relative to the flow diverter.
Accordingly, the subject invention provides a heat sink with a flow diverter for absorbing heat generated from the electronic device. The flow of cooling fluid removes the heat stored in the flow diverter, with the spiral wall directing the flow of cooling fluid away from the flow diverter. The flow diverter provides a large mass directly above the heat source for absorbing a significant amount of the heat generated thereby. The spiral wall absorbs additional heat from the heat source, which is transmitted radially outward from the heat source through the base plate. The direct impingement of the flow of cooling fluid on the flow diverter removes the majority of heat stored therein, with the remaining heat stored in the spiral wall and the base removed by the flow of cooling fluid as the cooling fluid circulates through the spiral channel. The subject invention, therefore, provides a more efficient heat sink for a cooling assembly.
Other advantages of the present invention will be readily appreciated, as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings wherein:
Referring to the Figures, wherein like numerals indicates like or corresponding parts throughout the several views, a heat sink assembly is generally shown at 10.
Referring to
Referring to
The base 12 and the lid 16 are generally circular with the outlet 32 extending tangentially therefrom. An outer wall 34 extends between the base 12 and the lid 16 about an outer periphery of the base 12 and the lid 16 for encapsulating the heat sink 10, so that the flow of cooling fluid enters through the inlet tube 26 and is discharged through the outlet 32.
The spiral wall 28 includes a plurality of louvers 38 for creating turbulence in the flow of cooling fluid. Preferably, as best shown in
Referring to
Referring to
Referring to
Referring to
where “a” is the radius of the circular base 50 and “Bid” is a dimensionless quantity called the Biot number for the flow diverter 20 defined by the equation:
where “kd” is the thermal conductivity of the flow diverter material and “h” is the heat transfer coefficient of the cooling fluid surrounding the flow diverter 20.
The desired values of the radius (a) of the circular base 50 are in the range 0.2 in≦a≦0.4 in (5.1 mm≦a≦10.2 mm) and those of the Biot number (Bid) in the range 0.0005≦Bid≦0.05 corresponding to the preferred values of c/a, which are in the range of 1.22≦c/a≦14.12.
The efficiency (φ) of the flow diverter 20 is defined as the ratio of the heat flux ({dot over (q)}d″) on the surface of the flow diverter 20 to the heat flux ({dot over (q)}b″) at the circular base 50 of the flow diverter 20. Since heat flux is the heat dissipation rate per unit area, the efficiency is defined by the equation:
where “{dot over (q)}d” is the heat dissipation rate from the surface of the flow diverter 20, “{dot over (q)}b” is the heat dissipation rate from the circular base 50 of the flow diverter 20, “Sd” is the surface area of the flow diverter 20, and “Sb” is the area of the circular base 50.
When the flow diverter 20 having a hyperbolic cross section and the flow diverter 120 having a conical cross section are compared, wherein each of the flow diverters 20, 120 is intended to dissipate the same amount of heat and include the same circular base area but define a different surface area, the efficiency ratio between the two flow diverters 20, 120 is defined by the equation:
where “φhyperbolic” is the efficiency of the hyperbolic-shaped flow diverter 20, “φconical” is the efficiency of the conical-shaped flow diverter 120, “Shyperbolic” is the surface area of the hyperbolic-shaped flow diverter 20, and “Sconical” is the surface area of the conical-shaped flow diverter 120.
Since the surface area (Shyperbolic) of the hyperbolic-shaped flow diverter 20 is less than the surface area (Sconical) of the conical-shaped flow diverter 120, each having the same base area, it follows from Equation (4) that the efficiency (φhyperbolic) of the hyperbolic-shaped flow diverter 20 is greater than the efficiency (φconical) of the conical-shaped flow diverter 120. Presented in Table 1 are the numerical values of the surface area (Shyperbolic) of the hyperbolic-shaped flow diverter 20 and the surface area (Sconical) of the conical-shaped flow diverter 120 normalized by the area of the circular base 50 over a range of the ratio of the height (c) of the flow diverters 20, 120 to the radius (a) of the circular base 50. Also included in Table 1 are the numerical values of the flow diverter efficiency ratios suggested by Equation (4). The tabular results show that for the entire range of the c/a ratio, the efficiency of the hyperbolic-shaped flow diverter 20 is 17% to 49% higher than that of the conical shaped flow diverter 120.
Referring to
where “r” is the local radius of the spiral and “θ” its angle of rotation in radians from the beginning of the spiral wall 28 at the flow diverter 20.
The spiral channel 30 includes a width (sθ) defined by the equation:
where “a” is the radius of the circular base 50 of the flow diverter 20. As the spiral angle (θ) increases, the width (sθ) of the spiral channel 30 becomes constant. The preferred values of sθ/a are in the range 2≦sθ/a≦3.
The spiral wall 28 includes a height (b) defined by the equation:
where “t” is the thickness of the spiral wall 28 and “Biw” is a dimensionless quantity called the Biot number for the spiral wall 28 defined by the equation:
where “kw” is the thermal conductivity of the spiral wall material and “h” is the heat transfer coefficient of the cooling fluid surrounding the spiral wall 28. The preferred values of the thickness (t) of the spiral wall 28 lie in the range of 0.001 in≦t≦0.006 in (0.025 mm≦t≦0.152 mm) and those of the Biot number (Biw) lie in the range of 0.000005≦Biw≦0.00005 corresponding to the preferred values of b/t, which are in the range of 29≦b/t≦290.
As the spiral angle (θ) increases, the spiral width (sθ) of the spiral channel 30 becomes constant to ensure that a flow velocity (ūθ) of the cooling fluid is maintained as the cooling fluid flows in a spiral fashion around the flow diverter 20 and through the spiral channel 30. The cooling fluid impinges onto the flow diverter 20 filling the inner spiral wall 28 spanning between 0 rad.≦θ≦2π rad. Once the inner spiral wall 28 is filled, the flow of cooling fluid is constrained to flow primarily in a spiral fashion around the flow diverter 20 within the spiral channel 30
The mean flow velocity (ūθ) of the flow of cooling fluid through the spiral channel 30 is defined by the equation:
where “{dot over (m)}” is the mass flow rate of the cooling fluid impinging on the flow diverter 20, “ρ” is the fluid density of the cooling fluid, “sθ” is the width of the spiral channel 30, and “b” is the height of the spiral wall 28.
Although the cooling fluid flows primarily in the θ-direction through the spiral channel 30, there is intrusion and extrusion of the fluid into the spiral channel 30 at each of the louvers 38 in the spiral wall 28 through the gaps created by the twisting of the louvers 38. The movement of the fluid at the louvers 38 serves to destabilize the thermal boundary layer on the spiral wall 28, thereby augmenting the heat transfer coefficient.
The heat transfer coefficient (h) between the spiral wall 28 and the flow of cooling fluid in the presence of the louvers 38 is defined by the equation:
where “ho” is the heat transfer coefficient in the absence of louvers 38, “n” is the linear density of the louvers 38 in the spiral wall 28, and “α” is the aspect ratio of the spiral channel 30. The aspect ratio (α) is defined by the equation:
The heat transfer coefficient (ho) in the absence of louvers 38 is defined for uniform wall temperature (UWT) boundary conditions by the equation:
and for uniform wall heat flux (UHF) boundary conditions by the equation:
where “b” is the height of the spiral wall 28, “k” is the thermal conductivity of the cooling fluid flowing through the spiral channel 30, and “α” is the aspect ratio of the spiral channel 30 defined by equation 10.
The preferred values of the louver density (n) are in the range of 15≦n≦35 louvers per inch along the spiral wall 28 (6≦n≦14 louvers per cm).
The foregoing invention has been described in accordance with the relevant legal standards; thus, the description is exemplary rather than limiting in nature. Variations and modifications to the disclosed embodiment may become apparent to those skilled in the art and do come within the scope of the invention. Accordingly, the scope of legal protection afforded this invention can only be determined by studying the following claims.
Number | Name | Date | Kind |
---|---|---|---|
4686606 | Yamada et al. | Aug 1987 | A |
5019880 | Higgins, III. | May 1991 | A |
5029636 | Kadle | Jul 1991 | A |
5144531 | Go et al. | Sep 1992 | A |
5304846 | Azar | Apr 1994 | A |
5394936 | Budelman | Mar 1995 | A |
5441102 | Burward-Hoy | Aug 1995 | A |
6059026 | Bailly et al. | May 2000 | A |
6148907 | Cheng | Nov 2000 | A |
6205803 | Scaringe | Mar 2001 | B1 |
6324058 | Hsiao | Nov 2001 | B1 |
6366462 | Chu et al. | Apr 2002 | B1 |
6422307 | Bhatti | Jul 2002 | B1 |
20030164231 | Goodson et al. | Sep 2003 | A1 |
20050103472 | Lofland et al. | May 2005 | A1 |