US20080218970A1

US20080218970A1 - Thermal Management for a Ruggedized Electronics Enclosure

Info

Publication number: US20080218970A1
Application number: US12/020,487
Authority: US
Inventors: William E. Kehret; Dennis H. Smith
Original assignee: Themis Computer
Current assignee: Themis Computer
Priority date: 2002-08-30
Filing date: 2008-01-25
Publication date: 2008-09-11
Also published as: US6765793B2; US20070177348A1; US20040042175A1; WO2004021401A3; JP2005537667A; US6944022B1; EP1552736A2; AU2003265552A8; EP1552736A4; WO2004021401A2; AU2003265552A1

Abstract

The present invention relates to a liquid cooling assembly for cooling electronic components. The liquid cooling assembly contains a heat spreader plate for providing mechanical support and thermal dissipation; a fluid channel for directing a cooling fluid in the plane of the heat spreader; and a bottom plate for protecting against destructive shock events and for providing thermal dissipation. The present invention also provides a maze structure in the liquid cooling assembly to increase structural stability against destructive shock events. The present invention also relates to a ruggedized electronics enclosure for housing electronic components. A top compartment contains a first electronics layer and a second electronics layer adjacent to said first electronics layer and a cooling assembly. A thermal shunt is configured to channel heat from the first and second electronics layers to the cooling assembly and to provide additional mechanical support to protect against potentially destructive shock events.

Description

RELATED APPLICATIONS

This application is a divisional of U.S. Ser. No. 11/207,672 that was filed on Aug. 19, 2005, which in turn is a continuation-in-part of U.S. Ser. No. 11/203,005 that was filed on Aug. 11, 2005, which in turn is a continuation of U.S. Ser. No. 10/850,523 that was filed on May 19, 2004 and has now issued as U.S. Pat. No. 6,944,022, issued Sep. 13, 2005 which in turn is a continuation of U.S. Ser. No. 10/232,915 that was filed on Aug. 30, 2002 and has now issued as U.S. Pat. No. 6,765,793, issued Jul. 20, 2004, which applications are incorporated herein by reference.

BACKGROUND OF THE INVENTION

1. Field of the Invention
This invention is related to enclosures for electronic circuits and particularly to the thermal management of ruggedized enclosures for use in installations subjected to hostile environments, including destructive shock events and destructive vibration events.
2. Description of the Related Art
Conventional ruggedized electronics enclosures are often employed in military applications. The environments in which military electronic circuits must be able to operate typically present conditions outside of a commercial electronic circuit's operational parameters. Examples of such conditions include excessive moisture, salt, heat, vibrations, and mechanical shock. Historically, military electronic equipment was custom made to provide the required survivability in the hostile environments. While effective in surviving the environment, custom equipment is often significantly more expensive than commercial systems, and is typically difficult if not impossible to upgrade to the latest technologies. Therefore, a current trend in conventional military hardware is to adapt commercially available electronics for use in military applications. These systems are typically known as Commercial Off The Shelf systems, or COTS.
The COTS design philosophy has allowed the military to keep current with technological innovations in computers and electronics, without requiring specialized and dedicated electronic circuit board assemblies. The COTS design methodology is attractive because of the rapidly increasing computational power of commercially available, general-purpose computers. Since the components in a COTS system are commercially available, though usually modified to some extent, the military can maintain an upgrade path similar to that of a commercial PC user. Thus the COTS philosophy allows the military to integrate the most potent electronic components available into their current hardware systems.
While COTS systems have allowed the military to reduce the cost of equipment and to make more frequent upgrades to existing equipment, there are inherent disadvantages to COTS systems. As noted above, military applications must be able to withstand various environmental extremes, including humidity, temperature, shock and vibration. These conditions are typically outside of the operating parameters of commercial electronics and, thus, added precautions and modifications to the physical structures of the equipment must be made to ensure reliability of operation in these environments. Conventional COTS systems typically use two specialized modifications to maintain reliability. These approaches may be used separately, or in combination.
To deploy COTS equipment in hazardous environments, COTS components are housed in a complex ruggedized enclosure or case. One approach, sometimes referred to as “cocooning” places a smaller, isolated equipment rack within a larger, hard mounted enclosure. With this approach shock, vibration and other environmental extremes are attenuated by the isolation system to a level that is compatible with COTS equipment. Another approach, sometimes called Rugged, Off The Shelf (ROTS) seeks to “harden” the COTS equipment, in a manner such as to make it immune to the rigors of the extended environmental conditions to which it is exposed. This later approach strengthens the equipment's enclosure and provides added support for internal components. Both cocooning and ROTS design methodologies must also improve cooling efficiency to accommodate higher operating ambient temperatures. Both approaches suffer from added complexity, size, weight and cost.
Commercial systems are typically designed around three main criteria, cost, time-to-market and easy expansion. To deliver on all three design goals, the assumption is that the environment for the system will not be exposed to extreme environmental conditions. Cost is the primary motivator to keeping the packaging simple and inexpensive. The package support structures may have a low cost to keep the system cost from escalating. Keeping costs down to a minimum is counter to the requirements of making a system robust enough to survive a military environment.
To easily accommodate system expansion, computer manufacturers try to simplify the installation of peripheral cards, memory and storage. The idea of having a minimum number of fasteners (i.e., a snap-in-place design) allows the customer easy access and installation of peripherals. The design's modularity preserves the customer's investment. When you couple the commercial constraints with the requirements of the military environment, the design requires a different approach, typically moving the structural changes to the system enclosure and it's attachments. The usual cocooning approach is to design the enclosure to absorb as much of the shock as possible to allow the incumbent system to survive the environment. In practice, this is not easily achieved, especially when using larger and heavier computer systems. Thus, the idea of completely isolating a commercial system from the rigors of the military environment is difficult to achieve and adds a large cost premium because the rack is the item being modified. The current solution to supporting COTS technology in a military environment described above, adds significant complexity to the system.
Two of the most difficult conditions to design for are vibration and mechanical shock. Mechanical shock and vibration may over time destroy electronic equipment by deforming or fracturing enclosures and internal support structures and by causing electrical connectors, circuit card assemblies and other components to fail. In military applications, as well as in commercial avionics and the automotive industry, electronics must be able to operate while being subjected to constant vibrational forces generated by the vehicle engines, or waves, as well as being subjected to sudden, and often drastic, shocks. Examples of such shocks are those generated by bombs, missiles, depth charges, air pockets, potholes, and other impacts typically encountered by military or commercial vessels. Furthermore, these conditions may also be seen in the operating conditions of a network or telephone server during an earthquake. While providing some protection from shock and vibration, the conventional ruggedized enclosure operating alone cannot provide adequate protection for mission-critical electrical components and circuits.
In order to provide additional protection against shock and vibration, conventional COTS systems mount the ruggedized enclosures described above in a mechanically isolated cocoon. FIG. 1 illustrates a conventional mechanically isolated cocoon. As illustrated in FIG. 1, a cocoon 100 is provided to house the various ruggedized enclosures 110. The cocoon 100 may be attached to a floor 130 and/or a wall 140 of its surroundings. Commonly this includes the fuselage or deck plate of a military vehicle. The cocoon 100 is attached to the surroundings 130, 140 via mechanical isolators 120. A particularly advanced mechanical isolator 120 is the polymer isolator illustrated in FIG. 1, though conventional systems may use any spring-like apparatus to provide the isolation. By attaching the cocoon 100 to its surroundings 130, 140 via mechanical isolators 120, the cocoon 100 is allowed limited movement with five degrees of freedom. This limited movement helps to dampen the effects of shock and vibration.
There are several drawbacks to using the mechanically isolated cocoon 100. The size and complexity of the cocoon 100 exacerbates the need for efficient heat-removal from the enclosure. Often complex heat flow routes must be devised in order to maintain a desirable operating temperature of electronic components within the cocoon 100. Taken together, these design considerations drastically increase the cost and complexity of such an enclosure.
Some conventional electronics enclosures, like the cocoon 100, rely on a liquid cooling system for stabilizing the internal operating temperatures of mounted circuit boards. Conventional liquid cooled enclosures are provided with a heavy cold plate containing bored channels for a liquid cooling assembly to pass through. The cold plate can be manufactured from a variety of thermally conductive materials to assist in dissipating the heat generated from electronic circuit boards. However, the reliance of conventional liquid cooling systems upon the traditional cold plate arrangement drives up the cost and overall weight of the assembly.
Various heat-removing methods are known to industries outside of the ruggedized electronics markets. In a typical semiconductor device heat management arrangement, a material with a moderately high thermal conductivity, like aluminum, is deposited upon a lower thermally conductive substrate like silicon. A highly thermally conductive layer, like pyrolytic graphite or copper, is then deposited on top of the moderately high thermally conductive layer. Finally, a layer of semiconducting material (or active material) is deposited on top of the highly thermally conductive material to complete the semiconducting device. The three layers of thermally conductive material underneath the heat generating active layer provide adequate heat spreading throughout the conductive layers. In some conventional heat sinking techniques, a diamond pin is embedded within the pyrolytic graphite such that heat can dissipate away from the active layer in a direction different from the direction of heat dissipated by the thermally conductive layers.
However, several drawbacks arise when applying conventional semiconductor device heat dissipation techniques to large area electronics encompassing multiple stacks of electronic layers. Even more drawbacks are present when applying these heat dissipation techniques to large area electronics operating in an environment conducive to destructive shock events and destructive vibration events. Relying upon a conventional heat management system is too expensive because of the need for multiple thermal layers, each with their own unique thermal conductivity to surround the electronic board. Also, introducing multiple thermal layers would increase the weight and reduce the structural integrity of a ruggedized electronics enclosure.
What is needed is a ruggedized enclosure for use in hostile environments which is capable of efficiently dissipating heat generated by enclosed electronic circuitry through the use of a lightweight, cost-effective, structurally sound liquid cooling assembly.
In addition, what is needed is a ruggedized enclosure for use in hostile environments which: is 1) lightweight; 2) cost-effective; 3) capable of providing a structurally sound housing for packaged electronic layers; and 4) capable of efficiently dissipating heat generated by multiple layers of electronics.

SUMMARY OF THE INVENTION

The present invention overcomes the limitations and disadvantages of conventional thermal management techniques used in electronics enclosures that operate in harsh environments.
According to one embodiment, the present invention provides a layer of foam or foam-like structure surrounding the fluid channels of a liquid cooling assembly for adaptation to a ruggedized enclosure. Grooves are bored through an upper portion of the foam structure to hold the fluid channels. In an embodiment, support structures surround the foam structure and can be reinforced with carbon fiber or other high tensile strength materials to provide the cooling assembly with a mechanically rigid “skin.” The foam structure of the present invention provides both mechanical support and thermal heat dissipation.
According to one embodiment, the present invention includes a maze type structure surrounding the fluid channels of a liquid cooling assembly. This maze or support structure includes grooves through an upper portion of the maze structure such that fluid channels can be secured to the upper portion of the maze structure. According to one embodiment, the maze structure includes a matrix of cells fabricated from high tensile strength material. The maze structure of the present invention provides both mechanical support and thermal heat dissipation.
In one embodiment, a ruggedized electronics enclosure is provided with a first electronics layer placed adjacent to a cooling assembly. The ruggedized electronics enclosure contains a first and second electronics layer, a first and second thermal interposer, a thermal shunt, and a cooling assembly. The first thermal interposer is placed adjacent to the first electronics layer. The second electronics layer is placed adjacent to the first thermal interposer. The second thermal interposer is placed adjacent to the second electronics layer. The first and second thermal interposers provide heat dissipation away from the first and second electronics layers. The thermal shunt provides a thermal connection between the first electronics layer, the second electronics layer, and the cooling assembly. In an embodiment, the thermal shunt is bored through the first electronics layer, the first thermal interposer, and the second electronics layer.
In one embodiment, the electronics enclosure includes a top compartment for housing the electronic circuit, and a cooling assembly attached thereto. The top compartment may be sealed to further protect the electronic circuit from moisture and unwanted particles in the air. The cooling assembly includes a rigid truss plate structure which forms a structural member for rigidifying the enclosure, and also forms an efficient heat radiator for removing heat from the electronic circuit. The truss plate structure achieves it's high strength to weight ratio in a manner similar to conventional “honey-comb” or sandwich structures. The truss plate structure converts bending mode forces, applied to opposing plates, into compression and extension mode forces. However, unlike conventional “honey-comb” or sandwich constructions, the present invention provides ducts or passage ways through which cooling air (or other cooling fluid) is allowed to flow to aid in the efficient removal of heat from the top compartment. In an alternate embodiment, the truss plate structure is a honey-comb truss structure that provides passages through which cooling air (or other cooling fluid) is allowed to flow.
In one embodiment, the rigid truss plate structure is formed from a passive radiator coupled between a heat spreader plate and a bottom plate. The heat spreader plate also forms the bottom of the top enclosure and provides both mechanical and thermal coupling between the top compartment and the cooling assembly. In one embodiment, the passive radiator may be comprised of a corrugated fin. In another embodiment, the passive radiator is comprised of triangularly shaped fins (an A-frame structure). Both the corrugated fin and the triangular fin structure may provide additional protection against destructive shear and twisting of the enclosure. In another embodiment, the passive radiator is comprised of a pin-style heatsink. In one embodiment the pin-style heatsink is arranged according to a pin density pattern to create a turbulence gradient for the cooling assembly.
In one embodiment, the enclosure is rigidified by the truss plate structure in order to protect the electronic circuit against an anticipated destructive shock event. In one embodiment, the enclosure and circuit can withstand and survive a 60 G shock event. In alternate embodiments the enclosure is designed based upon various criteria such that a particular enclosure and enclosed device (e.g., circuit) is designed to withstand and survive shock events in the range of 20 G to at least 60 G depending upon these design criteria. In another embodiment, the enclosure's resonant frequency is raised above an anticipated destructive vibration event. In one embodiment, of special interest for land vehicle or aircraft applications, the enclosure and circuit have a resonant frequency in the range of 200 Hz to at least 1 kHz. In another embodiment, of special interest for shipboard applications, the enclosure and circuit have a resonant frequency in the range of 20 to 40 Hz. The listed ranges are merely exemplary, and alternate embodiments may have a resonant frequency selected to be higher than a known destructive vibration event.
In one embodiment, the cooling assembly further provides heat pipes for drawing away additional heat from the electronic circuit and delivering it to an external heat exchanger. In one embodiment, the heat pipes cooperate with the passive radiator to provide an efficient heat exchanger.
In one embodiment, the electronic enclosure includes the use of microchips. These chips may be placed top-down on the heat spreader plate in order to provide a more efficient heat transfer from the chip to the cooling assembly.
A method for protecting and cooling an electronic circuit via a rigid truss plate structure is also provided.
The features and advantages described in the specification are not all inclusive, and particularly, many additional features and advantages will be apparent to one of ordinary skill in the art in view of the drawings, specification and claims herein. Moreover, it should be noted that the language used in the specification has been principally selected for readability and instructional purposes, and may not have been selected to delineate or circumscribe the inventive subject matter, resort to the claims being necessary to determine such inventive subject matter.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 illustrates a conventional mechanically isolated cocoon system.

FIG. 2 illustrates an exploded view of a ruggedized electronics enclosure according to one embodiment of the present invention.

FIG. 3 illustrates a cut-away structural detail of the assembled ruggedized electronics enclosure according to one embodiment of the present invention.

FIG. 4 illustrates a cut-away diagram of the ruggedized electronics enclosure showing heat and airflow related to the enclosure according to one embodiment of the present invention.

FIG. 5 illustrates a cooling assembly utilizing a triangular fin structure.

FIG. 6 illustrates a cooling assembly utilizing a pin-style heatsink.

FIG. 7 illustrates a cooling assembly utilizing a pin-style heatsink forming a turbulence gradient.

FIG. 8 illustrates a liquid cooling assembly utilizing structural foam in accordance with one embodiment of the present invention.

FIG. 9 illustrates a cross-sectional view of the liquid cooling assembly of FIG. 8 according to one embodiment of the present invention.

FIGS. 10A-10B illustrate a liquid cooling assembly utilizing a maze of walls structure in accordance with an embodiment of the present invention.

FIG. 11 illustrates a ruggedized electronics enclosure with an arrangement of thermal shunts in accordance with an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

A preferred embodiment of the present invention is now described with reference to the figures where like reference numbers indicate identical or functionally similar elements. Also in the figures, the left most digit(s) of each reference number correspond(s) to the figure in which the reference number is first used.
The present invention relates to a ruggedized electronics enclosure for protecting electronic circuits that must be able to survive and operate under harsh conditions such as those in military and automotive environments. The enclosure must be able to protect the electronic circuits from severe vibration and shock, heat, moisture, dust particulate, and various other adverse conditions. Throughout this description, the word “destructive” will be used to indicate a force or event which may cause the enclosure or the electronic circuit to fail after a single occurrence of the event, or after repeated occurrences of the event between maintenance intervals. Specific destructive events will be discussed in more detail below.
FIG. 2 illustrates an exploded view of a ruggedized electronics enclosure 200 according to the present invention. As illustrated in FIG. 2, the enclosure 200 is configured to house and protect a compute element 210. The compute element 210 is chosen by way of example as illustrative of the primary features and operation of the enclosure 200, and one skilled in the art will recognize that the enclosure 200 may be configured to house and protect any electronic circuit. Examples of alternate electronic circuits include various other components used in a computer, ordinance guidance and communication boards, vehicle control modules, radio and communications equipment, radar equipment, etc. As will be discussed below, the enclosure 200 may most advantageously be used for any electronic circuit which may be formed having a low vertical profile, but may be used to add increased protection to any dimensioned electronic circuit.
The ruggedized electronics enclosure 200 includes a top compartment 220 for housing the electronic circuit 210 (illustrated as a compute element), and a cooling assembly 230 coupled to the bottom of the top compartment 220. As illustrated, the enclosure 200 is shaped as a rectangle, however any footprint shape may be used. Non-rectangular shapes may be preferred in applications where space is at a premium, such as in aircraft, or military ordinance.
The top compartment 220 includes a top cover 222, one or more thermal interposers 224, a pair of side walls 226, a front wall 227, a rear wall (not shown) and a heat spreader plate 240. In one embodiment, the side walls 226, front wall 227 and rear wall as well as the top cover 222 are formed from aluminum. Alternatively, these portions of the top compartment 220 may be may be formed of any rigid material including, but not limited to steel, and plastics. Preferably, side walls 226 are sized to extend the entire combined height of the top compartment 220 and cooling assembly 230. Front wall 227 and rear wall are preferably sized to extend the height of the top compartment 220. An upper portion of side walls 226, front wall 227, the back wall, top cover 22 and heat spreader plate 240 cooperate to form the sealed top compartment 220 for housing the electronic circuit 210. In another embodiment, the top compartment 220 may not be sealed, but may instead be open to the environment. The various parts which form the top compartment 220 may be coupled together using screws or other fastener types that may require special tools for removal. Additionally, the screw fasteners may be augmented by other self-aligning/locking mechanical components. By utilizing screw fasteners or other removable fasteners, the top compartment 220 may be opened as necessary to provide service to the electronics housed inside. Alternatively, the compartment structures, or a substructure therein, may be formed by milling or casting a single piece of material such as aluminum, steel or plastic. Another alternative includes welding the elements comprising the top compartment 220 together to form a solid enclosure. However, while welding may increase structural stability, it decreases the enclosure's 200 serviceability.
The cooling assembly 230 is coupled to the bottom of top compartment 220 and further includes a passive radiator 232 (here illustrated in one embodiment 232 a) and a bottom plate 234. The passive radiator 232 and bottom plate 234 are coupled to the cooling assembly 230 in order to draw heat away from the highest dissipation components (the top compartment 220) to a high efficiency heat exchanger (the passive radiator 232).
As illustrated, the passive radiator 232 may be formed from an aluminum corrugated fin 232 a. As will be discussed below, the use of an aluminum corrugated fin 232 a provides specific advantages over other passive radiators, however, one skilled in the art will recognize that other passive radiators may be used in place of the corrugated fin 232 a, as well as that the radiator 232 may be made from other material aside from aluminum. For example, the passive radiator may be formed from copper, carbon fiber, composite structures of aluminum and copper or plastic, and may additionally be used in conjunction with heat-pipes and cold plates. Additionally, other structures aside from a corrugated fin 232 a may be used. FIG. 5 illustrates a triangular fin, or A-frame, truss structure 232 b preferably formed from aluminum or steel. As will be discussed below, this embodiment of the passive radiator 232 is more difficult and more expensive to manufacture, but provides additional structural integrity to the enclosure 200. FIG. 6 illustrates another embodiment of the passive radiator 232 utilizing pin-style heat-sinks 232 c positioned between the heat spreader plate 240 and the bottom plate 234. This forms a rigid truss plate structure while allowing some measure of heat dissipation profiling based on the placement and density of the pins.
In general, heat spreader plate 240, a lower portion of side walls 226, and bottom plate 234 cooperate to “sandwich” the passive radiator 232 into a solid rigid truss plate structure. The truss plate structure achieves a high strength to weight ratio by converting bending mode forces, applied to opposing plates, into compression and extension mode forces. This is similar to plates formed from conventional honey-comb or sandwich construction. However, unlike conventional “honey-comb” or sandwich construction, the present invention provides ducts or passageways through which cooling air (or other cooling fluid) is allowed to flow to aid in the efficient removal of heat from the top compartment 220.
The cooling assembly 230 may be assembled in a number of ways, with one goal being to keep the assembly process simple, while preserving structural rigidity and allowing the effective transfer of heat from the base-plate to the passive radiator 232. One way of doing this with a metallic passive radiator 232 is through welding. If a non-metallic passive radiator 232 is used, a thermally conductive adhesive may be used.
As illustrated, electronic circuit 210 is a compute element and includes a PCB 212, a plurality of processors 214 coupled to a front of the PCB 212, and a plurality of memory components 216 electrically coupled to a back of PCB 212. A thermal interposer 224 a is positioned to contact the back of PCB 212 and the memory components 216 to provide a heat exchange between PCB 212 and memory components 216. Typically, the interposer 224 is made up of a resilient plastic material, doped with a thermally conductive and insulating compound such as aluminum oxide, boron nitride or other materials. Alternatively, the interposer 224 may be formed from a gel or a foam. Alternatively, the top compartment 220 may be filled with thermally conductive foam. While this alternative provides structure and heat removal, it is not preferred due to the permanent nature of the installation. A removable interposer 224 is preferred to aid in the keeping the electronics inside the top compartment 220 serviceable.
As will be discussed in greater detail below, processors 214 and PCB 212 are positioned within top compartment 220 such that processors 214 are placed in physical contact with heat spreader plate 240, allowing for heat to be conducted away from processors 214. Alternatively, a heat conducting material, such as a thermal interposer similar to interposer 224, may be position between the processors 214 and the heat spreader plate 240. A second thermal interposer 224 is positioned between the memory components 216 and the top cover 222. Top compartment 220 is preferably sized to provide just enough vertical and horizontal room to fit electronic circuit 210 within its confines. In a preferred embodiment, thermal interposers 224 are created from a resilient material which is slightly compressed to ensure a “snug” fit for the electronic circuit 210 within top compartment 220. By ensuring that the thermal interposers 224 make tight contact with the top cover 222, additional thermal and structural benefits are realized.
FIG. 3 illustrates a cut-away structural detail of the assembled ruggedized electronics enclosure 200. As introduced in FIG. 2, in one embodiment, the electronic circuit 210 housed in the top compartment 220 is again a compute element. One of the objectives for the ruggedized electronics enclosure 200 is to provide protection to the electronic circuit 210 housed in the top compartment 220 from harsh operating environments. As noted above, the top compartment 220 may be completely sealed by appropriately sizing the side walls 226, front wall 227 (not shown), back wall (not shown), top cover 222 and heat spreader plate 240 to ensure that no open spaces exist in the top compartment 220 surface.
In addition to being able to make the top compartment 220 airtight, additional steps may be made to “ruggedize” the enclosure 200 to help reduce the effects of destructive shock events and destructive vibration events on the electronic circuit 210 housed within. A destructive shock event is any shock event that may render the electronic circuit 210 or enclosure 200 inoperative due to a large change in force and momentum being applied to the circuit 210 and enclosure 200. The circuit 210 or enclosure 200 may be rendered inoperative after a single destructive shock event or after a series of destructive shock events occurring between maintenance intervals. Examples of destructive shock events include impacts and explosions from bombs, missiles, other military ordinance, water craft hitting depth charges, aircraft hitting air pockets, wheeled vehicles hitting potholes as well as other impacts typically encountered by military or commercial vessels. One skilled in the art will recognize that other destructive shock events exist and that the above list provides only a general context for the nature of a destructive shock event.
Similarly, a destructive vibration event is any vibration event that may cause the electronic circuit 210 or enclosure 200 to fail due to a weakened structural integrity. Destructive vibration events may be isolated and short-lived in duration or may always be present in the operating environment. Examples of destructive vibration events include engine vibrations, turbine vibrations, screw vibrations, prolonged shock events, travel along uneven surfaces etc. One skilled in the art will recognize that other destructive vibration events exist and that the above list provides only a basic context for the nature of a destructive vibration event.
In typical military applications, the electronic circuit 210 must be able to survive and continue to operate efficiently after being subjected to an 60 G shock or constant vibration from engines and other movement. Military specifications MIL810, MIL901, MIL167 and ISO 10055 provide specific requirements for shock and vibration resistance depending on the desired application and are incorporated in their entireties herein. Typically, the individual chip-level components used in a standard commercial environment will withstand up to a 60 G shock load. This is due in part to the fact that the interconnects and silicon are packaged such that there is high structural rigidity in the component. However, one concern is with the printed circuit board (PCB) and its assembly. To minimize the shock impact to the PCB and the solder connections, it is beneficial to have structural ties between the board and its components and cooling assembly 230.
One design goal is to make the entire enclosure assembly one rigid structural element in order to protect against destructive shock and vibration events. In one embodiment, the enclosure is rigidified by the truss plate structure in order to protect the electronic circuit against an anticipated destructive shock event. In one embodiment, the enclosure and circuit can withstand and survive a 60 G shock event. In alternate embodiments the enclosure is designed based upon various criteria (e.g., materials, mass, truss plate, dimensions, assembly methods, etc.) such that a particular enclosure and enclosed device (e.g., circuit) is designed to withstand and survive shock events in the range of 20 G to at least 60 G depending upon these design criteria.
One aspect of forming the enclosure 200 as a rigid structural element includes raising the enclosure's 200 resonant frequency to a frequency higher than the destructive vibration events to which the enclosure 200 will be subject. Two major factors that affect the resonant frequency of a given structure are the mass, and the material's inherent stiffness. Typically, the lower the mass, the higher the resonant frequency. Thus, the overall mass of the enclosure 200 helps determine the resonant frequency of the enclosure 200 as well as its susceptibility to vibrational damage. Also, the higher the material stiffness, the higher the resonant frequency. As noted above, from a vibration standpoint, it is desirable to have the resonant frequency above the frequencies of any anticipated destructive vibration events to keep the mechanical structure from adding to the vibration energy.
Thus, the enclosure 200 is formed from a material that balances stiffness and mass to provide an overall high resonant frequency which is higher than the anticipated destructive vibration event frequencies. In the preferred embodiment, the ruggedized enclosure 200 is composed primarily of aluminum. The use of aluminum offers a good compromise between strength needed to protect the electronic circuit 210, while providing a lower total mass for the enclosure. As will be discussed below, the use of aluminum also provides an efficient way of removing heat generated by the electronic circuit 210. In one embodiment, the enclosure 200 is designed to have a resonant frequency that is at least approximately twice the 12-25 Hz frequency of naval shock events. In an alternate embodiment, the enclosure 200 has a resonant frequency in the range of hundreds of Hz, to protect the enclosure against an aircraft's prop or turbine vibrations. The specific resonant frequency chosen will be dictated by the specific vibrational frequency of the prop or turbine engine used, e.g., between 200 Hz and 1 kHz. These frequencies are merely examples of the resonant frequencies supported by the present invention. Alternate embodiments will have a resonant frequency selected to be greater than the vibrational frequency of an anticipated shock event that is to be dissipated by the enclosure 200.
Another aspect of the ruggedized enclosure 200 is its overall profile. In a preferred embodiment, the overall vertical height of the enclosure 200 is 1 rack unit (“U”) or 1.75 inches. Additionally, in one embodiment, the top compartment 220 is configured to house the electronic circuit 210 snugly, without allowing for significant horizontal or vertical movement within the compartment 220. Further cushioning and insulation from vibration is garnered by the use of the thermal interposers 224 which may be compressed slightly to ensure a snug fit while providing an efficient heat conduit to remove heat from the electronic circuit 210.
Passive radiator 232 provides additional resistance to destructive shock and vibration events. By using a passive radiator and fluid channel structure such as the corrugated fin 232 a, the triangular fin 232 b, or the pin-style heatsink 232 c, a light-weight rigid truss plate structure may be formed from the cooling assembly 230. This structure is stiffened by cross coupling (via the passive radiator 232) between the top compartment 220 and bottom plate 234. By forming the truss plate structure, the passive radiator 232 provides the cooling assembly 230 with structural properties similar to a solid thick plate from a rigidity standpoint for resisting destructive shock and vibration events. While a solid thick plate generally provides additional structural integrity to the enclosure 200, there is a tradeoff between plate thickness and overall mass. As noted above, the resonant frequency of the enclosure 200 would be decreased by the increased mass of a solid plate. By instead using a truss plate structure for the cooling assembly 230, the enclosure 200 retains the benefit of a thick plate while avoiding the lower resonant frequency associated with a thick, heavy plate.
In addition to the passive radiator 232, the interposers act to absorb high frequency vibrations by acting as lossy dissipative elements. The combination of top cover 222, thermal interposers 224, electronic circuit 210, and cooling assembly 230 in a small vertical space helps makes the total enclosure 200 very stiff. Furthermore, the interposers reduce the transfer of energy between the bottom plate 234 and the top cover 222, essentially dissipating the conducted vibrational energy. Additionally, materials used in bottom plate 234, heat spreader plate 240 and top cover 222 may be selected to dissipate mechanical (vibrational) energy. In particular, composite materials can offer a combination of high strength (stiffness) and damping (mechanical energy dissipation).
As noted above, the truss plate structure helps rigidify the enclosure 200 by cross coupling the top compartment 220 and the bottom plate 234. For example, the use of the triangular fin structure 232 b or corrugated fin 232 a as the passive radiator 232 may also help reduce the effects of destructive shear events and destructive vibration events in the horizontal direction indicated by arrow 310 and in a vertical direction indicated by arrow 320. Using a corrugated fin 232 a for the passive radiator 232 provides a good structure to transfer energy in both horizontal and vertical direction. The corrugation directs forces along the axes of the structure. The corrugations may also act to reduce the vibrational energy by acting as a dissipative spring. Tying the corrugations to the top and bottom plate 240, 234 at the peaks stiffens the structure in the “vertical” direction, effectively raising the structure's vertical (or bending mode) resonant frequency.
FIG. 4 illustrates a cut-away diagram of the ruggedized electronics enclosure showing heat and airflow related to the ruggedized electronics enclosure 200. In FIG. 4, to more clearly illustrate the heatflow and airflow, the top compartment 220 is not fully shown, but it is understood that the cooling assembly 230 is coupled to a top compartment 220 which houses and protects electronic circuit 210 as illustrated in FIG. 2.
FIG. 4 illustrates two directions for heat flow from electronic circuit 210, here illustrated as PCB 212 and processor 214. A primary direction for heat flow is illustrated by an arrow 410. This heat flow is accomplished by putting the processor 214 in thermally conductive contact with heat spreader plate 240. In one embodiment contact may be made by placing the processor 214 in direct contact with the heat spreader plate 240. Alternatively contact may be made by placing a heat conductive medium between the processor 214 and the heat spreader plate 240. Preferably, heat spreader plate 240 has a high thermal conductivity. In a preferred embodiment, processor 214 is oriented to be upside down so that its “top” is pressed against heat spreader plate 240. This arrangement allows for direct heat conduction between processor 214 and heat spreader plate 240. In conventional microchips, the main direction for heat to escape the chip is through its “top”. By positioning the top of the processor 214 against the heat spreader plate 240, heat is efficiently conducted from the processor 214 to the heat spreader plate 240. Alternatively, the microchips may face with their “tops” away from the heat spreader plate 240 and a thermal interposer 224 or other thermally conductive medium may be placed between the microchip and the heat spreader plate 240.
Heat spreader plate 240 conducts heat away from the electronic circuit 210 in the direction indicated by arrow 410, and into the passive radiator 232. Passive radiator 232 is designed to radiate the heat conducted from the electronic circuit 210 into the environment. Preferably, passive radiator 232 is exposed to an air flow across its surface area. This air flow is indicated by arrow 430 in FIG. 4. By inducing an air flow 430 through the spaces formed from passive radiator 232 and top and bottom plates 240, 234, heat may be efficiently removed from the electronic circuit 210 and from the ruggedized electronic enclosure 200 in general. Alternatively, the cooling assembly 230 can be mounted vertically to allow the heated air to rise, cooling the assembly through thermally induced convection currents. The specific proportions of passive radiator 232 directly affect its efficiency in removing heat from the enclosure 200. For instance, the overall height and width of a single “segment” directly affects the amount of surface area present for radiating heat, as well as changing the profile of the air channels. The profile of the air channels affects the channel's impedance to airflow and thus, the rate of airflow (for a given pressure differential) through the air channels of the passive radiator 232 and consequently the enclosure 200.
Additionally, for low airflow situations, the cooling assembly 230 is designed to radiate the maximum amount of heat to the ambient air. Increasing the surface area increases the heat transfer between the processor and the air. This may result in a “tighter” corrugation or more transitions between the heat spreader plate 240 and the bottom plate 234. If, however, an externally generated pressure differential is used to induce air movement past the passive radiator 232, then the design may optimize the passageways through the passive radiator 232 for optimum heat transfer at a given pressure differential. The size of the passageways directly affects the impedance of air that may flow across the passive radiator 232. As the passageways decrease in size, the air flow for a given pressure differential, and therefore, the heat transfer efficiency of the cooling assembly 230, will also decrease. Thus, one design goal is to balance the surface area of the passive radiator 232 against the size of the passageways and resultant air flow and heat transfer efficiency. In this way, different operating conditions may be met by adjusting the proportions of the passive radiator 232 to the requirements of the specific application and environment.
As noted above with respect to FIG. 3, the passive radiator 232 also provides shock and vibration protection. These shock and vibration aspects of the passive radiator 232 are also dependent on the proportions of each “segment”. It may be necessary to balance the application's need for shock and vibration protection against the operating temperature requirements. Typically, it is required that systems operate at ambient temperature extremes above 50 degrees Celsius. Maximum chip case temperatures measured at the package are commonly specified not to exceed 75 degrees C. For low power devices, this is easily achieved. For higher power devices, the thermal resistance from the electronics to air becomes a significant factor. In the case of higher power devices, a different material may be used for the passive radiator 232 in order to improve the heat transfer to the cooling assembly 230, such as copper or carbon composite materials.
As noted above, heat spreader plate 240 is preferably formed from a material with a high thermal conductivity, such as aluminum. Alternatively, the heat spreader plate 240 may be formed from copper or a carbon composite in order to provide a higher thermal conductivity and improved cooling efficiency at higher rates of airflow. Any type of material may be used for the passive radiator 232 in this alternate embodiment.
In one embodiment, heat spreader plate 240 or the passive radiator 232 may be configured to conduct heat from a “hotter” exhaust side 715 of the air channels to a “cooler” inlet side 710, to allow the energy flux into the air channel to stay constant, along an axis of the heat spreader plate 240. This can be accomplished by making the heat spreader plate relatively thicker at the inlet side 710 and thinner at the exhaust side 715. In another embodiment, a turbulence gradient may be achieved by varying the cooling assembly 230 channel capacity, or by varying the pin density of the passive radiator 232, (if a pin-style heat sink similar to pin-style heat sink 232 c is used,) by changing the profile of pins, or by any other means. FIG. 7 illustrates a cooling assembly 230 with a turbulence gradient. The cooling assembly 230 has an intake 710 represented by the air-flow arrow 710 a and an exhaust 715, represented by arrow 715 a. Near the intake 710 of the cooling assembly 230 the passive radiator 232 is comprised of elliptical pin fins 232 d. As air moves along the passive radiator 232 from intake 710 to exhaust 715, along a direction indicated by arrow 720, the pressure drop along the direction 720 of airflow is increased. At the exhaust 715 end of the cooling assembly 230, the pin fins 232 e are shaped to be more cylindrical, which may be similar to the pin style heat sink 232 c. These cylindrical pin-fins 232 e induce more turbulence and thus create a higher pressure drop. The varying turbulence caused by changing the pin profile along arrow 720, tends to keep the rate of energy transfer constant, even though the temperature of the air increases from the intake 710 to the exhaust 715 of the cooling assembly 230. This turbulence profiling makes it easier for the heat spreader to maintain an isotherm. The thermal conductivity of the heat spreader can be increased, usually meaning the mass can be reduced, thus allowing the structure's resonant frequency (for flexure modes) to be increased, with no reduction in heat transfer efficiency.
The turbulence profiling described above helps maintain several chips in contact with the heat spreader plate 240 at a similar temperature. This may be especially helpful in the situation where high rates of airflow 430 are induced by an externally generated pressure differential from inlet to exhaust. Referring back to FIG. 4, as the air flows in the direction of arrow 430, it will be heated by passive radiator 232, thereby reducing its effectiveness in cooling the remainder of the passive radiator 232. By designing the turbulence profile to match the changes in airflow temperature, the temperature of the electronic circuit 210 may be maintained. By maintaining a substantially uniform temperature across all components in electronic circuit 210, timing variances due to temperature variations between components may be reduced. This may be especially important if several processors are operating in parallel.
While the above discussion focused primarily on an embodiment of the enclosure 200 which utilizes an air cooled corrugated fin passive radiator 232 a, one skilled in the art will recognize that liquids such as sea water or a commercial refrigerant, other gasses such as gaseous nitrogen, may be used to conduct heat away from the passive radiator 232. Alternatively, there may be no liquid or gas present in the system and thermal transfer is achieved by radiation or convection from the external surfaces of the enclosure. One embodiment utilizes a liquid heat exchanger, substituting fluid channels for the passive radiator 232. All the mechanical benefits of the truss plate structure would be retained, and the modest increase in mass would be more than compensated for in heat transfer efficiency. Another embodiment puts the passive radiator in physical contact with a cold wall in an aircraft. Additionally, heat pipes may be embedded in the heat spreader plate 240 to help remove heat to an external heat exchanger. Additionally, while a corrugated fin 232 a and a triangular fin truss 232 b have proven to be advantageous from a production and structure standpoint, one skilled in the art will recognize that other passive radiators are also contemplated by this disclosure. Examples of other possible passive radiators include punched corrugated fins, conventional fin-style heat sinks that may be coupled to the top and bottom plates 240, 234, honey-comb truss structures oriented to allow air to pass through them, or a solid metal plate with longitudinal channels or holes placed therein.
FIG. 8 illustrates a liquid cooling assembly 800 adapted for cooling electronic components. The liquid cooling assembly 800 utilizes structural foam to withstand a destructive shock event in accordance with an embodiment of the present invention. The cooling assembly 800 includes at least the following: a heat spreader plate 240; a plurality of fluid channels 810; a plurality of fluid channel grooves 840; a layer of structural foam 820; and a bottom plate 234.
In an embodiment of cooling assembly 800, the plurality of fluid channels 810 are positioned within a plurality of fluid channel grooves 840, substantially between the structural foam 820 and the heat spreader plate 240. The structural foam 820 is positioned in between the plurality of fluid channels 810 and the bottom plate 234. In an embodiment, the plurality of fluid channel grooves 840 are molded into an outer portion of the layer of structural foam 820 such that the plurality of fluid channels 810 rest within a recess formed by the plurality of fluid channel grooves 840. In one embodiment, the plurality of fluid channels 810 are thermally adhered to the plurality of fluid channel grooves 840 with a thermally conductive epoxy. In an alternative embodiment the fluid channels 852 can be formed as part of the heat spreader plate 240 and can be positioned above the cover plate 854 in one embodiment. The fluid channels 852 can be used in a manner similar to that described with reference to fluid channels 810.
In one embodiment, the plurality of fluid channels 810 provide a conduit for channeling a cooling fluid through the cooling assembly 800 in order to draw heat away from the heat spreader plate 240. The plurality of fluid channels 810 can be formed from any thermally conductive material like copper, aluminum, or a carbon fiber composite. In an embodiment, the plurality of fluid channels 810 can be arranged in a single, serpentine arrangement such that minimal heat buildup occurs within the plurality of fluid channels 810. In an alternate embodiment, the fluid channels include any number of individual channels that are fed by a common liquid-producing source. One of ordinary skill in the art will appreciate a variety of geometrical arrangements that the fluid channels 810 can take on depending on the particular physical constraints inherent in a given housing environment. The exact number of fluid channels is an application-specific parameter that can vary depending on the precise pressure of fluid flow required to efficiently cool a particular electronics apparatus.
The cooling fluid (not shown) that passes through the fluid channels 810 acts to increase thermal dissipation away from the heat spreader plate 240. The cooling fluid can be selected from a group of liquids including de-ionized water or some mixture of de-ionized water and ethylene glycol, ammonia, or alcohol. In an embodiment, the cooling fluid passing through the plurality of fluid channels 810 is Fluorinert™. One of ordinary skill in the art will appreciate a variety of liquids that are resistant to a variety of environmental conditions, like intense heat, can be used as a cooling fluid.
In one embodiment, the structural foam 820 surrounds the fluid channels 810 such that the structural foam 820 provides mechanical rigidity against a destructive shock event and thermal conduction of heat away from the heat spreader plate 240. The structural foam 820 can be replaced with light-weight but stiff, thermally conductive material, thus provides added heat dissipation to cooling assembly 800 by supplementing the heat dissipation provided by the fluid cooling channels and heat spreader plate 240. The structural foam 820 also prevents deformation against compressive forces. Typically, the structural foam 820 is formed from a thermally conductive, closed-cell foam or a porous, open-cell material. Typical porous materials used to form structural foam 820 include styrofoam, styrene-based foam, or urethane-based foam. One of ordinary skill in the art will appreciate a variety of open or closed-cell materials can be used to form structural foam 820 such that adequate structural rigidity is achieved to withstand a destructive shock event.
Forming the structural foam 820 with an open or closed-cell foam material to support fluid channels 810 of the cooling assembly 800 proves beneficial for a variety of reasons. Structural, closed or open-cell foam is mechanically rigid and thermally conductive. Over a large area, closed or open-cell foam is relatively non-compressible, thus providing an excellent mechanical buffer for sensitive electronic components during catastrophic shock events. Also, closed or open-cell foam is lightweight and cost efficient for large-scale use.
A cross-sectional view of cooling assembly 230, as shown in FIG. 9, includes reinforcing fiber 910 embedded within the bottom plate 234 of truss plate structure. The fiber 910 provides structural reinforcement, stiffening the plate for compression and extension forces in the plane of the plate, that improves the truss section mechanical performance by providing stiffness for loads that are normal to the plane of the plate. The reinforcing fiber 910 may be oriented in the plane of the surface and may contain mixed orientations to improve isotropic stiffness, and may be formed from carbon or any other high tensile strength material depending on the cost and rigidity constraints of a particular cooling apparatus. Structural foam stiffness can also be enhanced with variously oriented additives, such as nanotubes or other strength enhancing add mixtures.
FIG. 10A illustrates a liquid cooling assembly 1000, adapted for cooling electronic components, utilizing a maze of walls structure 1010 adapted to withstand a destructive shock event in accordance with an embodiment of the present invention. The cooling assembly 1000: a heat spreader plate 240; a plurality of fluid channels 810; a plurality of fluid channel grooves 840; a maze structure 1010; and a bottom plate 234. In an alternate embodiment, the fluid channels 810 may be formed into the heat spreader plate 240, as in the alternate embodiment shown in FIG. 8, in which case the fabrication of maze structure 1010 would be simplified because the groves 840 would not be required.
According one embodiment, the plurality of fluid channel grooves 840 can be formed by mechanically boring grooves through an upper portion of the maze structure 1010. Typically, the plurality of fluid channels 810 are thermally adhered to the plurality of fluid channel grooves 840 with a thermally conductive epoxy. In an alternative embodiment the fluid channels 852 can be formed as part of the heat spreader plate 240 as described above. In another embodiment, the fluid channels 810 can be glued or mechanically fastened with fasteners to the upper portion of the maze structure 1010. In another embodiment, the fluid channels 810 can be coupled to the maze structure 1010 in a similar fashion as the coupling between the fluid channels 810 and the structural foam 820 described, above, in relation to FIG. 8.
The maze structure 1010 of the present invention provides both mechanical support for cooling assembly 1000 and thermal heat dissipation away from the heat spreader plate 240. In an embodiment, the maze structure can be formed from aluminum, graphite, or a carbon fiber. One skilled in the art will appreciate a variety of metallic or composite materials that can form maze structure 1010 to accomplish similar structural rigidity and thermal conduction for any particular application.
In an embodiment shown in FIG. 10B, the maze structure 1010 includes a matrix of cells 1050 fabricated from a high tensile strength, highly thermally conductive material. For example, the matrix of cells 1050 can be fabricated from aluminum, graphite, or any particular reinforced carbon fiber. In an embodiment, where the dimensions of cooling assembly 1000 are approximately 4″×4″, the maze structure 1010 includes at least nine cells 1050 that form the matrix of cells 1050. The exact number of cells 1050 contained within the maze structure 1010 depends on the size of cooling assembly 1000 and the precise stability requirements needed to withstand a particular catastrophic shock environment. In an embodiment, as shown in FIG. 10B, each cell 1050 within the matrix of cells can be rectangular in shape. In an alternate embodiment, each cell 1050 within the matrix of cells can be square, circular, elliptical, triangular, hexagonal, pentagonal, or other shape. In one embodiment, each cell 1050 within the matrix of cells can be the same regular geometric shape or each cell 1050 can be a different regular geometric shape. In another embodiment, each cell 1050 can be the same or a different irregular geometric shape. One of ordinary skill in the art will appreciate the cell 1050 formed in any variety of regular or irregular geometric shapes depending on the particular design constraints of the destructive shock environment. Non-regular shapes may be preferred in applications where space is at a premium, such as in aircraft, or military ordinance.
FIG. 11 illustrates a ruggedized electronics enclosure 1100 including an arrangement of thermal shunts 1160, in accordance with an embodiment of the present invention. In this embodiment, ruggedized enclosure 1100 includes a top compartment 220, for housing electronic components and a cooling assembly 230. The cooling assembly is mechanically and thermally coupled to the bottom of the top compartment 220 in order to provide structural support for top compartment 220 along with a means for efficient dissipation of heat generated by electronic components housed within top compartment 220. More specific details regarding the coupling of top compartment 220 has been discussed previously with regards to FIG. 2. Some of the components of the ruggedized electronics enclosure 1100 have similar function and form as has been described above with reference to FIGS. 2-7, so like reference numerals and terminology have been used to indicate similar functionality.
In an embodiment, the top compartment 220 includes: a top cover 222; a pair of side walls 226; a front wall 227; a rear wall (not shown); a heat spreader plate 240; a thermally conductive elastomer 1150 adjacent to the top cover 222; a first thermal interposer 224 a adjacent to the thermally conductive elastomer 1150; a first electronics layer 1140 adjacent to the first thermal interposer 224 a; a second thermal interposer 224 b adjacent to the first electronics layer 1140; a second electronics layer 1130 adjacent to the second thermal interposer 224 b; one or more electrical connectors 1165 providing electrical coupling between the first and second electronics layers 1140, 1130; one or more air gaps 1145 adjacent to the pair of side walls 226 and the heat spreader plate 240; one or more structural supports 1170 providing mechanical coupling between the second electronics layer 1130 and the heat spreader plate 240; and one or more thermal shunts 1160 positioned adjacent to the first electronics layer 1140, second electronics layer 1130, and the heat spreader plate 240.
In one embodiment, the thermally conductive elastomer 1150 is rigidly coupled in between the top cover 222 of top compartment 220 and one of the first thermal interposers 224 a. The thermally conductive elastomer 1150 advantageously provides a thermal conduction path from the thermal interposer 224 to the top cover 222 in order to draw heat away from the electronic layers. Also, thermally conductive elastomer 1150 provides mechanical dampening from destructive shock vibrations passing between the first thermal interposer 224 a and the top cover 222. In one embodiment, thermally conductive elastomer 1150 may be formed from any semi-metallic material that is thermally conductive and mechanically resilient. For example, the thermally conductive elastomer 1150 may be formed from aluminum nitride, silicon carbide, or boron nitride. Also, one skilled in the art will recognize that thermally conductive elastomer 1150 may be formed from a rubberized or plastic material doped with a thermally conductive compound like carbon or the like. In a preferred embodiment, thermally conductive elastomer 1150 is created from a resilient material which is slightly compressed to ensure a “snug” fit between first thermal interposer 224 a and top cover 222. By ensuring that the thermally conductive elastomer 1150 makes tight contact with the top cover 222, additional thermal and structural benefits are realized.
In one embodiment, the first thermal interposer 224 a is positioned between the first electronics layer 1140 and the thermally conductive elastomer 1150. By ensuring that the first thermal interposer 224 a makes tight contact with the thermally conductive elastomer 1150, additional thermal and structural benefits are realized. Typically, the first thermal interposer 224 a is oriented such that any heat generated by the first electronics layer 1140 is dissipated away from the first electronics layer 1140 in a direction substantially parallel to the first electronics layer 1140.
In one embodiment, the first electronics layer 1140 is positioned in between the first thermal interposer 224 a and the second thermal interposer 224 b. The first electronics layer 1140 can be directly connected with the second electronics layer by way of electrical connectors 1165. The resulting combination of the first 1140 and second electronics layer 1130, along with the integrated circuits 1175, advantageously provides increased electronic functionality for the particular system that the ruggedized enclosure 1100 is servicing. The first electronics layer 1140 can include one or more memory components 216 as described above.
In an embodiment, the second thermal interposer 224 b is positioned in between the first electronics layer 1140 and the second electronics layer 1130. Typically, the first thermal interposer 224 b is oriented such that any heat generated by the first electronics layer 1140 or second electronics layer 1130 is dissipated away from the first electronics layer 1140 or second electronics layer 1130 in a direction substantially parallel to the first electronics layer 1140 or second electronics layer 1130. In a preferred embodiment, thermal interposers 224 a and 224 b are created from a resilient material which is slightly compressed to ensure a “snug” fit for the first and second electronics layers 1140, 1130 within top compartment 220.
The electrical connectors 1165, providing electrical connection between the first and second electronics layers 1140, 1130, may be formed from a variety of electrically conductive materials. In an embodiment, electrical connectors 1165 are formed by embedding a metallic conductor, like copper, within any particular type of plastic material. In another embodiment, electrical connectors 1165 are formed by plating a plastic material on its periphery with any particular metallic material. Typically, electrical connectors 1165 are shaped in a regular geometric shape like a cylinder or cube. One of ordinary skill in the art will appreciate a variety of shapes and a variety of materials for forming electrical connectors 1665 in order to provide sufficient electrical contact between the first and second electronics layers 1140, 1130.
In an embodiment, the second electronics layer 1130 is mechanically coupled to the heat spreader plate 240 by way of structural supports 1170. The second electronics layer 1130 preferably includes an etched circuit board assembly 212 containing electrically conductive circuit paths. The second electronics layer 1130 can be constructed from any insulating material, like fiberglass or the like, and the conductive strips are generally laid down onto a surface of the second electronics layer 1130 through conventional etching techniques.
The structural supports 1170, providing additional support against destructive shock events for the second electronics layer 1130, can be formed from a variety of metallic materials like Al, Cu, stainless steel, or brass. The structural supports 1170 can function as a “stand-off” such that they physically separate, or hold off, the second electronics layer 1130 from the heat spreader plate 240. In an embodiment, the structural supports 1170 provide some thermal heat dissipation away from the second electronics layer 1130. Typically, the structural supports can be circular or hexagonal in shape depending on packaging constraints. One skilled in the art will appreciate a variety of support materials in a variety of shapes that can be used to form the structural supports 1170 to accomplish similar functionality.
In an embodiment, one or more integrated circuits 1175 are electrically coupled to a back surface of the second electronics layer 1130 and placed in direct contact with heat spreader plate 240. Providing direct contact between the integrated circuits 1175 and the heat spreader plate 240 allows for heat to be conducted away from the integrated circuits 11175.
In one embodiment, one or more thermal shunts 1160 are positioned adjacent to and/or through the first electronics layer 1140, second electronics layer 1130, and the heat spreader plate 240 to provide a thermal conduction path from the first electronics layer 1140, through the second thermal interposer 224 b and the second electronics layer 1130, to the heat spreader plate 240. The presence of the thermal shunts 1160 provides a direct conduit for thermal energy generated by the first electronics layer 1140 to be dissipated by the heat spreader plate 240. The thermal shunts 1160 also provide an avenue for dissipation of mechanical shock between the heat spreader plate 240 and the layers of electronics 1140, 1130. In an embodiment, thermal or mechanical energy passing through the thermal shunts 1160 is in a direction substantially perpendicular to the first and second thermal interposers 224 a, 224 b.
In one embodiment the thermal shunts 1160 are coupled by mechanically boring a plurality of aligned via holes within the structure of first electronics layer 1140, second thermal interposer 224 b, and second electronics layer 1130 and positioning each thermal shunt 1160 to fit within the plurality of the via holes. The thermal shunts 1160 can be circular or hexagonal or any particular shape depending on the constraints of the given geometry of the electronics layers through which the thermal shunts 1160 are bored. In another embodiment, the thermal shunts 1160 can be coupled with bent sheet metal and mounted to a bracket on the first and second electronics layers 1140, 1130 with a screw or fastener. Also, thermal shunts 1160 can be coupled to electronics layers 1140, 1130 and heat spreader plate 240 by way of thermally conductive grease.
In one embodiment, two thermal shunts per 4″×4″ electronics layer can produced adequate heat transfer. One of skill in the art, however, will appreciate any number of thermal shunts per layer of electronics can be used to suit the needs of a variety of applications. The thermal shunts 1160 may be arranged in a serial arrangement, where a first thermal shunt 1160 a (not shown) provides thermal connection between the first electronics layer 1140 and the second electronics layer 1130 and a second thermal shunt 1160 b (not shown), serially offset from the first thermal shunt 1160 a, provides thermal connection between the second electronics layer 1130 and the heat spreader plate 240. One skilled in the art can envision an electronics arrangement including a plurality of different electronics layers stacked in any particular arrangement where the advent of any number of variously arranged thermal shunts 1160 can offer a direct thermal contact between the each layer within the plurality of different electronics layers and the heat spreader plate 240.
In an embodiment, the thermal shunts 1160 are fabricated from a material with high thermal conductivity and high structural integrity like graphite or copper. The thermal shunts 1160 can also be formed from a variety of doped aluminum materials, including AlSiC. In another embodiment, the thermal shunts 1160 are formed from sputtered diamond. In another embodiment, the thermal shunts 1160 can be any particular heat pipe arrangement that provides both mechanical structure and conduction of thermal energy. More details regarding a possible heat pipe arrangement for thermal shunts 1160 are found, above, with reference to the descriptions for FIG. 2. One skilled in the art will appreciate a variety of thermally conductive, structurally sound materials that can form thermal shunts 1160.
While the invention has been particularly shown and described with reference to a preferred embodiment and various alternate embodiments, it will be understood by persons skilled in the relevant art that various changes in form and details can be made therein without departing from the spirit and scope of the invention.

Claims

1. A liquid cooling assembly for cooling electronic components, the cooling assembly comprising:

a heat spreader unit;

a structural foam layer, rigidly coupled to said heat spreader unit, providing mechanical support and thermal dissipation for the electronic components;

a fluid channel, rigidly coupled to said structural foam layer, for directing a cooling fluid in a first direction; and

a bottom plate rigidly coupled to said structural foam layer, wherein said heat spreader unit, said structural foam layer, and said bottom plate providing a rigid structure that does not substantially deform in response to one or more destructive shock events, to protect the electronic components against said one or more destructive shock events and to provide thermal dissipation of heat generated by the electronic components.

2. The liquid cooling assembly of claim 1, wherein said structural foam layer further comprises a fluid channel groove.

3. The liquid cooling assembly of claim 1, wherein said fluid channel is adhered to said fluid channel groove by way of a thermally conductive epoxy.

4. The liquid cooling assembly of claim 1, wherein said structural foam layer is formed with a closed-cell foam.

5. The liquid cooling assembly of claim 1, wherein said structural foam layer is formed with a substantially non-compressible material that has a substantially high thermal conductivity

6. The liquid cooling assembly of claim 1, wherein said bottom plate is embedded with a reinforcing fiber to provide structural strength and stiffness for compressive and extension forces, to improve the normal mode mechanical performance of the truss structure.

7. A liquid cooling assembly for cooling electronic components, the cooling assembly comprising:

a heat spreader unit;

a maze structure, rigidly coupled to said heat spreader unit, providing mechanical support and thermal dissipation for the electronic components;

a fluid channel, rigidly coupled to said maze structure, for directing a cooling fluid in a first direction; and

a bottom plate rigidly coupled to said maze structure, wherein said heat spreader unit, said maze structure, and said bottom plate providing a rigid structure that does not substantially deform in response to one or more destructive shock event, to protect the electronic components against said one or more destructive shock events and to provide thermal dissipation of heat generated by the electronic components.

8. The liquid cooling assembly of claim 7, wherein said maze structure further comprises a matrix of cells.

9. The liquid cooling assembly of claim 8, wherein said matrix of cells are fabricated from a high tensile strength, highly thermally conductive material.

10. The liquid cooling assembly of claim 8, wherein said matrix of cells is fabricated from aluminum, graphite, or any particular reinforced carbon fiber.