Increased power of integrated circuit chips, and the modules containing the chips, increases processor performance and heat generated in densely packed memory designs. Chip count and functionality on memory modules continue to increase, while the spacing between modules is minimized. This trend poses cooling challenges.
In the following detailed description, reference is made to the accompanying drawings which form a part hereof, and in which is shown by way of illustration specific examples in which the disclosure may be practiced. It is to be understood that other examples may be utilized and structural or logical changes may be made without departing from the scope of the present disclosure. The following detailed description, therefore, is not to be taken in a limiting sense, and the scope of the present disclosure is defined by the appended claims. It is to be understood that features of the various examples described herein may be combined, in part or whole, with each other, unless specifically noted otherwise.
In many cases, the electrical components in processors are cooled by air moving in parallel airflow paths, usually front-to-back, impelled by one or more air moving devices (e.g., fans or blowers). In some cases it may be possible to handle increased power dissipation by providing greater airflow, for example, through the use of a more powerful air moving device or by increasing the rotational speed (i.e., RPM) of an existing air moving device. Heat is typically carried from the electronic components by the air, or other fluid, passing through and exiting the processing subsystem or system. The fluid absorbs the heat dissipated by the components/modules to an outside environment, whether air or other liquid-coolant. The ability to cool integrated circuit chips, and the modules containing the chips, is a function of the volume of air, or coolant flow, and the surface area on each face of the module available to transfer heat to the passing coolant (e.g., air or liquid).
Electronic module 10 includes heat spreader 14 thermally coupled to electronic components 24 to be cooled. Heat spreader 14 includes a base 26 and fins 28 extending in a direction away from base 26. Fins 28 are coupled to, or part of, heat spreader 14 coupled to at least one electronic component 24 of the plurality of electronic components 24 on printed circuit board 12 of electronic module 10. Multiple fins 28 extend from, and are spaced along, base 26. Fins 28 and base 26 of heat spreader 14 provide surface area available to contact with the cooling medium, or cooling fluid, surrounding and passing by heat spreader 14. Fins 28 add surface area to the surface area provided by base 26 for heat conduction from the surface of heat spreader 14 to the fluid (e.g., air or liquid coolant). Fins 28 facilitate cooling of heat generated by electronic components 24 by providing additional surface area for heat conductance to the fluid. Thermal adhesive or thermal grease can be included between electrical components 24 on printed circuit board 12 and heat spreader 14 to fill gaps and provide thermal connection between electrical components 24 and heat spreader 14. Heat, or thermal energy, generated by electrical components 24 is transferred from heat spreader 14 to the fluid medium. Heat spreader 14 is made of highly thermally conductive material. In one example, base 26 and fins 28 on each surface 22, 24, respectively, of printed circuit board 12 are formed together of the same material. In one example, fins 28 and base 26 are formed from a sheet of conductive material bent or otherwise formed into the desired shape.
Electronic module 10 can be a memory module, an input-output module, a high density compute module, a switch module, for example. In one example, module 10 is a dual inline memory module (DIMM). In a DIMM, electrical components 24, such as semiconductor memory integrated circuits and capacitors are mounted on each face of printed circuit board 12. Heat generated in electrical components 24 is radiated from the surface of electrical components 24. Heat spreader 14 disposed on the respective surface 20, 22 is configured to thermally couple with the electronic components 24 on the respective surface 20, 22.
In an example multiple dual inline memory modules (DIMMs) 10, modules 10 can collectively maximize the heat dissipation of heat generating electronic components 24 housed on DIMMs 10 into a fluid flow. DIMMs 10, in particular, fins 28 on DIMMs 10 can collectively direct, or channel, the flow of fluid over and between DIMMs 10 in order to optimize the heat dissipation. For example, the fluid flow can be divided and directed to cool specific components 24.
Fins 28 can extend linearly from a first end 30 to an opposing second end 32 of electronic module 10. Alternatively, fins 28 can be contoured to direct cooling flow over components 24 that are desirably cooled such as high powered or heat sensitive components, rather than over relatively unpopulated portions of the DIMM that may be less desirable to cool. Fluid flow over components 24 can be segregated with contoured fins 28 to direct fluid flow over specific desired components 24, such as heat sensitive components, and around hot components, while a hot channel is routed past the heat sensitive component, over the hot components(s), and steered around rather than over other heat sensitive components. The channel may be configured around multiple dimensions (i.e., three dimensionally) to include deeper or shallower fins 28 to route around components 24 such that the channel formed between components 24 can be compressed on a vertical dimension to leave room for a segregated channel to expand and route over component 24. Adjacent DIMMs 10 share volumetric space with each other through interdigitated fins 28 within the processor. Fins 28 of adjacent electronic modules 10 cross extend into shared space between printed circuit boards 12 of electronic modules 10.
The flow path is initiated at a first end 30 of the at least two electronic modules 10. In one example, the flow is initiated by a fan either pushing or pulling the fluid over surfaces of the electronic modules 10. Cool air or coolant passes around and between fins 28, cooling electronic components 24 of the electronic modules 10 as heat is transferred from the surfaces of fins 28 and base 26 of heat spreader 14. Flow passes over the surfaces of interdigitated adjacent fins 28 of adjacent electronic modules 10, passing between the electronic modules 10 and exiting at second end 32 of electronic modules 10. The fluid passing from the first end 30 is heated by heat transferred from the electronic components 24 closest to the first end, and is accordingly pre-heated when passing over components closer to the second end 32.
An electronic device, such as an electronic subsystem 36, including a plurality of electronic components 24 to be cooled is illustrated in
The interdigitated electronic modules 10 can include a false, or dummy, memory module (i.e., without electronic components) to provide desired flows for cooling components 24 on adjacent electronic modules 10. For example, the last in a series of electronic modules 10 inserted into slots 34 can be included to interdigitate with fins 28 of the adjacent electronic module 10 to provide the desired flows at the edge of the group. The interdigitated electronic modules 10 can be coordinated with other components of the electrical system or subsystem, such a processor heat sink's configuration, to provide space to the electronic modules 10 and fluid flow channel at the edge of the group of electronic modules 10.
Although specific examples have been illustrated and described herein, a variety of alternate and/or equivalent implementations may be substituted for the specific examples shown and described without departing from the scope of the present disclosure. This application is intended to cover any adaptations or variations of the specific examples discussed herein. Therefore, it is intended that this disclosure be limited only by the claims and the equivalents thereof.
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/US2016/014884 | 1/26/2016 | WO | 00 |