Storage Cooler

Abstract

The present disclosure relates to a storage cooler. The storage cooler includes an integral heat transfer device and a pair of end covers. The integral heat transfer device includes a heat transfer tube, a first end block, a second end block, an inlet chamber, an outlet chamber, an inlet, and an outlet. The first end block and the second end block are structurally integrated with each heat transfer tube. The inlet chamber and the outlet chamber are partially defined by the first end block or the second end block. The inlet chamber and the outlet chamber are each fluidly coupled to a corresponding liquid flow channel of at least one heat transfer tube. The corresponding flow channels coupled to the inlet chamber and the outlet chamber can be the same or different to define a straight or serpentine flow path. Each end cover is attached to a corresponding one of the first end block and the second end block to define the inlet chamber and the outlet chamber together with the first end block and the second end block.

Description

Background Art

[0001] This section of the document introduces information about and / or from the art that may provide background for, or be related to, the subject matter described herein and / or claimed below. This information also provides background information to facilitate a better understanding of the various aspects of the present invention. Therefore, this is a discussion of "related" technology. The relevance of such technology in no way means that it is also "existing" technology. Related technology may or may not be prior art. The discussion in this section of the document should be read from this perspective, rather than as an admission of prior art.

[0002] The development of computing power with advances in computing technology has been accompanied in many ways by ever greater power consumption in larger, more powerful computing devices and systems. In particular, high performance computing ("HPC") systems experience this increased power consumption. Increased power consumption is often accompanied by more heat generation. Heat can be a problem that affects the operating efficiency and life of electronic components. Therefore, designers of computing devices research and implement techniques for handling the excess heat generated by electronic components to keep their operating conditions within specifications. As the heat generated in newer computing device designs increases, new techniques and devices for managing and handling excess heat are constantly sought. BRIEF DESCRIPTION OF THE DRAWINGS

[0003] The present invention may be understood by reference to the following description taken in conjunction with the accompanying drawings, in which like reference numerals represent like elements, and in which:

[0004] Figure 1 is a perspective assembly view of a memory cooler according to one or more examples disclosed herein.

[0005] Figure 2 yes Figure 1 A perspective exploded view of a memory cooler.

[0006] Figure 3 Shows that it can be used Figure 1 to Figure 2 A specific example of a heat transfer tube in an exemplary storage cooler of FIG.

[0007] Figure 4 Shows that it can be used Figure 1 to Figure 2 In an exemplary storage cooler of Figure 3 A second specific example of a heat transfer tube.

[0008] Figure 5 Depicted include Figure 1 to Figure 2 A memory cooler and at least one printed resistor assembly ("PCA").

[0009] Fig. 6A yes Figure 5 Exploded view of PCA.

[0010] Figure 6B Shows Figure 5 Aspects of PCA's leaf spring selection.

[0011] Figure 6C Shows Figure 5 The PCA(s) are mounted to a mainboard therein and the PCA(s) are mechanically joined and thermally coupled to the heat transfer tubes.

[0012] Figure 7 is a component according to one or more examples (such as Figure 5 Components) partially along Figure 5 A top view taken along line 7-7 of FIG. 7 and showing the coolant flow path.

[0013] Figure 8 It is shown that a memory cooler (such as Figure 1 to Figure 2 storage cooler) system.

[0014] Fig. 9 It shows that it can be implemented in Figure 8 A block diagram of a specific example of a design station in a system.

[0015] Fig.10 It shows that it can be implemented in Figure 8 A block diagram of a system for a specific example of a 3D printer.

[0016] Fig.11 A manufacturing method that can be implemented in one or more examples of the present disclosure is shown.

[0017] Fig.12 is a top view of another portion of an assembly illustrating a coolant flow path according to one or more examples.

[0018] Fig.13 is a top view of another portion of an assembly illustrating a coolant flow path according to one or more examples.

[0019] Fig.14 is a top view of another portion of an assembly illustrating a coolant flow path according to one or more examples.

[0020] Although the present invention is susceptible to various modifications and alternative forms, the accompanying drawings show specific examples described in detail by way of example herein. However, it should be understood that the description of specific examples herein is not intended to limit the present invention to the specific forms disclosed, but on the contrary, the present invention is intended to cover all modifications, equivalents and substitutes falling within the spirit and scope of the present invention defined by the appended claims. DETAILED DESCRIPTION

[0021] The following illustrative examples of the claimed subject matter will now be disclosed. For the sake of clarity, not all features of an actual implementation are described in this specification. It will be appreciated that in the development of any such actual implementation, many implementation-specific decisions must be made to achieve the developer's specific goals, such as compliance with system-related constraints and business-related constraints, which will vary from implementation to implementation. Furthermore, it will be appreciated that such development work, while complex and time-consuming, will be a routine task for those of ordinary skill in the art having the benefit of this disclosure.

[0022] Many computing systems employ double data rate ("DDR") and synchronous dynamic random access memory ("SDRAM"). A widely used, popular variant of DDR memory is DDR4, so named because it is considered the fourth generation of DDR memory. As part of the above computing system evolution, the industry is in the process of moving to DDR5, which is considered the fifth generation. Both DDR4 and DDR5 are defined by one or more industry standards.

[0023] Both DDR4 and DDR5 are often deployed in so-called dual in-line memory modules ("DIMMs"). It should be noted that in this context, the term "module" has a clear structural meaning to those skilled in the art, namely, multiple integrated circuit ("IC") chips and supporting electronic devices mounted to a printed circuit board ("PCB"). In the case of a memory module (such as a DIMM), the IC chip is a memory chip. In other cases, a module may be, for example, a processing module centered around processing resources rather than memory resources. Such a module may also be more generally referred to as a printed circuit assembly ("PCA"). Finally, the collection of PCAs and supporting components mounted to a "motherboard" may sometimes be generally referred to as a "compute module" or "compute node."

[0024] The cooling challenges faced from DDR4 DIMMs to DDR5 DIMMs increase dramatically. For example, the power consumption of DDR5 may increase by at least 100%, from 6-9W in DDR4 to 12-20W in DDR5. The spacing of DDR5 DIMMs may also decrease by 20%, from 9.4mm in DDR4 to 7.4mm in DDR5. In addition, the number of DIMMs per central processing unit ("CPU") in DDR5 may increase, from 8 DIMMs per CPU in DDR4 to up to 12 DIMMs per CPU in DDR5. In addition, CPU socket sizes continue to increase, thereby reducing the space for DIMMs, and PCA routing limitations are moving memory closer to the outer edge of the compute node. In addition, the system fluid temperature target (the entrance to liquid-cooled information technology ("IT") hardware) will rise from 27°C to 32°C, and the ability to increase fluid flow per compute node will be more limited.

[0025] These increases in power consumption and component count and density are accompanied by increases in heat generation. Managing heat as described above then becomes more difficult and challenges current technologies used for this purpose. The industry has sought to liquid cool the DIMMs of HPC systems by using various thermal management devices to circulate and deliver liquid coolants. The prevailing manufacturing technique is to use machined copper parts that are welded or brazed together at many points, creating a large number of potential leak points. Machined copper solutions are also heavy, expensive, and labor intensive to assemble.

[0026] The present disclosure provides a memory cooler that solves some or all of the problems discussed above. In various examples, a memory cooler is provided that includes only three components, each of which can be 3D printed from stainless steel. Other examples can use other additive manufacturing processes and other materials, as described more fully below. Most of the memory cooler is an integral heat transfer device. Terms such as "integral" and "structurally integrated" used herein mean that the heat transfer device is composed of a single piece.

[0027] More specifically, a storage cooler includes an integral heat transfer device and a pair of end caps. The integral heat transfer device includes a plurality of heat transfer tubes, a first end block, a second end block, an inlet chamber, an outlet chamber, an inlet leading to the inlet chamber, and an outlet leaving the outlet chamber. Each heat transfer tube has a first tube end and a second tube end and defines a corresponding liquid flow channel.

[0028] The first and second end blocks are structurally integrated with each of the plurality of heat transfer tubes at the respective first and second tube ends of the at least one heat transfer tube. The inlet chamber is partially defined by one of the first and second end blocks, and the inlet chamber is fluidly coupled to at least one respective liquid flow channel of at least one of the heat transfer tubes. The outlet chamber is partially defined by one of the first and second end blocks, and the outlet chamber is fluidly coupled to at least one respective liquid flow channel of at least one of the heat transfer tubes. Each end cap is attached to a respective one of the first and second end blocks to define the inlet chamber and the outlet chamber together with the first and second end blocks.

[0029] In a second example, an assembly includes a main printed circuit board ("PCB"), a memory cooler, and a plurality of memory boards. In various examples, the memory boards may be dual in-line memory modules ("DIMMs"). The memory boards are inserted into a main board, and each memory board is disposed in a corresponding slot defined by a heat transfer pipe of an integral heat transfer device. The memory cooler includes an integral heat transfer device and a pair of end caps, as just described above.

[0030] In another example, a method for manufacturing a memory cooler includes: developing a digital three-dimensional ("3D") model of a memory cooler, the memory cooler including an integral heat transfer device, a first end cap, and a second end cap; using an additive manufacturing process to manufacture the integral heat transfer device, the first end cap, and the second end cap; and attaching the first end cap and the second end cap to the integral heat transfer device. The memory cooler includes an integral heat transfer device and a pair of end caps, as described just above.

[0031] The subject matter claimed below will now be described with reference to the accompanying drawings. Various structures, systems and devices are schematically depicted in the accompanying drawings for the purpose of explanation only and in order not to obscure the present invention with details well known to those skilled in the art. However, the included drawings describe and explain illustrative examples of the present invention in order to further understand the content claimed below. However, the subject matter claimed is not limited to the disclosed examples.

[0032] Figure 1 is a perspective assembly view of a memory cooler 100 according to one or more examples disclosed herein, and Figure 2 yes Figure 1 1 is a perspective exploded view of a storage cooler 100. The storage cooler 100 includes an integral heat transfer device 102, a first end cap 104, and a second end cap 106. Figure 2As best shown, the integral heat transfer device 102 is manufactured in a single piece, as discussed further below. Thus, in this example, the storage cooler 100 includes only three pieces (ie, the integral heat transfer device 102 and the two end caps 104, 106).

[0033] The integral heat transfer device 102 includes a plurality of heat transfer tubes 108, a first end block 110, a second end block 112, a pair of side rails 114, and a middle rail 116. The integral heat transfer device 102 is manufactured as a single piece. Accordingly, the heat transfer tubes 108, the side rails 114, and the middle rail 116 are structurally integrated with the first end block 110 and the second end block 112. This manufacturing is performed using an additive manufacturing process, such as a three-dimensional ("3D") printing process further described below. The heat transfer tubes 108, the side rails 114, and the middle rail 116 define a plurality of slots 117 (only one is indicated), into which printed circuit assemblies ("PCAs") (such as DIMMs) can be arranged, as described below with respect to FIG. 6A to FIG. 6C Further discussion.

[0034] like Figure 3 As best shown, in this particular example, the heat transfer tubes 108 are heat transfer tubes having a rounded rectangular cross-section 306. However, alternative examples may have alternative cross-section geometries. Each heat transfer tube 108 has a respective first tube end 300 and a respective second tube end 302 defining a respective liquid flow channel 304 therebetween. The liquid flow channels 304 are internal to the body of the heat transfer tubes 108, but are shown here in solid lines for clarity. As described below with respect to Figure 7 As further discussed, during operation, liquid coolant flows through the liquid flow channel 304 to remove heat generated by the electronic components. Additionally, the heat transfer tube 108 includes optional cutouts 310 to accommodate the height of the electronic components of the DIMM.

[0035] In some examples, the interior surface of the liquid flow channel may include surface extensions to facilitate heat transfer. Figure 4 One such example is shown: the heat transfer tube 108a defines a liquid flow channel 304a. The inner surface 400 of the liquid flow channel 304a defines a plurality of surface extensions 402. In the example shown, the surface extensions are "fins" and may be collectively referred to as a "fin array". Other examples may use other types of surface extensions. For example, some examples may use ridges or buttons instead of fins. The number and spatial distribution of the surface extensions are implementation specific depending on the desired heat transfer and flow specifications and may be determined by modeling or other means.

[0036] Return to see Figure 1 to Figure 2, the first end block 110 is structurally integrated with each heat transfer tube in the plurality of heat transfer tubes 108. More specifically, the first end block 110 is structurally integrated at a corresponding first tube end 300 of the heat transfer tube 108. The second end block 112 is structurally integrated with each heat transfer tube in the plurality of heat transfer tubes 108 at a corresponding second tube end 302, as shown in FIG. Figure 3 Thus, integral heat transfer device 102 is "integral" in the sense that first end block 110 and second end block 112 are each structurally integrated with each heat transfer tube 108 such that they are fabricated as a single piece.

[0037] The first end block 110 partially defines a circulation chamber (eg Figure 7 ), the circulation chamber is fluidly coupled to each corresponding liquid flow channel 304 defined by each structurally integrated heat transfer tube 108. The second end block 112 partially defines an inlet chamber 200 fluidly coupled to a first portion of the liquid flow channel 304, and an outlet chamber 202 fluidly coupled to a second portion of the liquid flow channel 304. The second end block 112 also includes an inlet hose barb 206 and an outlet hose barb 208, as shown. Figure 2 The inlet hose barb 206 and the outlet hose barb 208 are also structurally integrated with the second end block 112. It should be noted that other examples may use some other type of 3D-printable connection.

[0038] In this particular example, the characterizations "inlet" and "outlet" are for convenience only and relate to the direction of fluid flow through storage cooler 100. In other examples, the direction of fluid flow may be reversed, and chamber 200 may become the "outlet" chamber, while chamber 202 is the "inlet" chamber. Similarly, in such an example, hose barb 206 may be the "outlet" hose barb, while barb 208 may be the "inlet hose barb."

[0039] As discussed further below, the first and second end blocks 110, 112 may define, together with the first and second end caps 104, 106, the inlet chamber 220 and the outlet chamber 202 and, if used, one or more of the circulation chamber(s). The inlet chamber, the outlet chamber, and the circulation chamber(s) are fluidly coupled to the flow passages of the heat transfer tubes 108 to define a liquid flow path through the storage cooler 100. In various examples, the liquid flow path may be straight or serpentine, which in turn determines which chambers are fluidly coupled to corresponding flow paths of which heat transfer tubes 108.

[0040] Additionally, depending on the particular example being implemented, the inlet and outlet to the flow path may be in the first end block 100 or the second end block 112. Still further, in some examples, the inlet chamber and the outlet chamber may be at least partially defined by the same end block or different end blocks. Figure 7 and Figures 12 to 14 Examples of these variations are shown and discussed.

[0041] The first end cap 104 is attached to the first end block 110 so as to define the circulation chamber 202 together with the first end block 110. The second end cap 106 is attached to the second end block 112 so as to define the inlet chamber 200 together with the second end block 112. The attachment may be achieved by, for example, but not limited to, brazing, bonding, welding, etc. The attachment technique is selected not only to achieve attachment, but also to be able to achieve a liquid-tight seal to prevent leakage of liquid coolant.

[0042] The memory cooler 100 may be used to cool one or more PCAs in an assembly such as a computer module. Figure 5 Depicted include Figure 1 to Figure 2 1 and a plurality of PCAs 502. For purposes of illustration, one jacketed PCA 502 is shown removed from the assembly 500. In this particular example, the assembly 500 is a computing module that may include other electronics (not shown), such as one or more central processing units ("CPUs"), etc., to facilitate the function.

[0043] In this example, assembly 500 includes a "main" board 504. Main board 504 is so called because it "hosts" other boards such as printed circuit boards of jacketed PCA 502, as discussed below. Main board 504 is a printed circuit board ("PCB") that also includes components for electrically connecting jacketed PCA 502 to other components of main board 504. These components may include, for example, but not limited to, a plurality of sockets and electrical traces, none of which are shown for clarity and due to their prevalence in the art. In the example shown, this component includes a female connector (not shown) to mate with edge connector 506 of jacketed PCA 502.

[0044] The memory cooler 100 is mounted to the motherboard 504 in a manner not shown. The mounting can be achieved, for example, by fastening the memory cooler 100 to the motherboard using fasteners (such as screws, bolts, etc.) (not shown). The first end block 110 and the second end block 112 can be manufactured with mounting holes (not shown) to achieve this particular purpose. However, any suitable mounting technique known in the art can be used.

[0045] Fig. 6A yes Figure 56 is an exploded view of a jacketed PCA 502. For purposes of the present disclosure, the jacketed PCA 502 includes a PCA 600, which in the example shown is a memory board. The PCA 600 includes a PCB 602 on which a plurality of electronic components 604 (not all shown) of various types are mounted. The electronic components 604 include a plurality of memory modules 606. The electronic components 604 and the edge connector 506 are electrically interconnected by a plurality of electrical traces (not shown), and are electrically interconnected with other components (not shown) of the motherboard 504 through the edge connector 506.

[0046] The jacketed PCA 502 includes a PCA 600 comprised of a PCB 602 on which a plurality of electronic components 604 are mounted. In the example shown, the PCA 600 is a memory board, and the electronic components 604 are a memory module 606 and a capacitor 607, respectively. More specifically, the memory module 606 is a DRAM, and the PCA 600 is a DIMM. It should be noted that in other examples, the memory module 606 may be some kind of memory module other than a DIMM. The PCA 600 may include other kinds of electronic components in addition to or in place of the memory module 606 and the capacitor 607.

[0047] For example, PCA 600 may also include an inductor (not shown), a voltage regulator (also not shown), a controller (also not shown), etc. In some examples, PCA 600 may be a processor board. In these examples, the electronic components may be processors or other types of electronic components for implementing the functions of the processor board. Some examples of using multiple jacketed PCAs 502 may include jacketed PCAs 502 in which the PCAs 600 are configured to perform different functions.

[0048] In addition, the jacketed PCA 502 includes a heat transfer jacket, which in the example shown includes a first cover 610 and a second cover 612. The first cover 610 and the second cover 612 each include a plurality of leaf springs 510, the form and function of which will be described below with respect to FIG. 6B to FIG. 6C More fully described. In the example shown, the first cover 610 and the second cover 612 are each formed of a single layer of material having a thermal conductivity of at least 350 W / mK and a thickness of at most 0.25 mm. However, other examples may use more than one layer of material (provided that the spacing specifications are met), materials with different thermal conductivities, and materials with different thicknesses.

[0049] The first cover 610 and the second cover 612 are opposed to each other and mechanically engaged in a manner to enclose the PCA 600. When enclosing the PCA 600, the heat transfer jacket is mechanically engaged and thermally coupled to the PCA 600 in a manner more fully described below. As used herein, "thermally coupling" two objects means providing a thermally conductive path between the objects that allows heat to be conducted between the objects. Two objects can be considered to be thermally coupled if any of the following is true: (1) the two objects are in "dry" or direct contact, where the thermal resistance of the contact interface is 0.1°C / W or greater, or are in contact with each other via a thermal interface material ("TIM"), where the thermal conductivity of the thermal interface material is 3W / mK or greater, and (2) both objects are thermally coupled to a thermal transfer device (or to a chain of thermal transfer devices that are thermally coupled to each other). A TIM may be a relatively thin, thermally conductive, and compliant (easily deformable) material that is placed between two devices at the thermal interface of the two devices to improve the heat transfer rate by filling air gaps that would otherwise appear between the devices due to surface roughness and / or misalignment. Common examples of TIMs include thermal gap pads, thermal grease, and thermal paste.

[0050] See again Figure 5 , these heat transfer tubes 108 are a heat transfer device, such as the overall heat transfer device 102. As described below Figure 6C As described, heat transfer tube 108 is thermally coupled and mechanically engaged with jacketed PCA 502, but not mechanically fastened or otherwise attached to jacketed PCA 502. A "heat transfer device" may be any thermally conductive device configured to receive heat from one solid body via conduction (contact) and transfer heat to a second solid body via conduction (contact). Examples of heat transfer devices include, but are not limited to, heat pipes, vapor chambers, heat spreaders (such as solid metal rods or strips), cooling tubes, and the like.

[0051] The jacketed PCA 502 also includes an optional plurality of pull tabs 508 attached to the second cover 612. The pull tabs 508 can be used to remove the jacketed PCA 502 from the assembly 500. The pull tabs 508 can be made of a flexible material so that they can bend. For example, they can be serpentine when the computing system is placed in a tray (not shown) and inserted into a rack (also not shown) or other enclosure. Some examples can omit the pull tabs 508.

[0052] The jacketed PCA 502 also includes a plurality of leaf springs 510, Figure 5 Only one leaf spring is indicated. More specifically, the first cover 610 and the second cover 612 each include a plurality of leaf springs 510. This Figure 6BIt can be seen that each leaf spring 510 is formed integrally with the first cover 610 or the second cover 612, in the sense that the leaf spring is formed from a single piece with the first cover or the second cover.

[0053] Now turning to leaf spring 510, Figure 6B Selected details of the construction of leaf spring 510 are shown. Figure 6B 6 is a perspective enlarged view of the bottom edge 614 of the first cover 610 of the heat jacket, which also includes a second cover 612. The leaf spring 510 is constrained at both ends. The leaf spring 510 is created by removing material from the first cover 610 on each side of what will become the leaf spring 510. The flexure 626 and opening 627 are then created. The design and construction of the leaf spring 510 in the second cover 612 can be similar.

[0054] The integral heat transfer device 102 (including the heat transfer tube 108 and each of the first cover 610 and the second cover 612) is made of one or more thermally conductive materials. For the purposes of this disclosure, an object, device, or assembly (which may include multiple different bodies that are thermally coupled and may include multiple different materials) is "thermally conductive" if any of the following is true: (1) the object is a continuous piece of material that has a thermal conductivity (commonly expressed as k) of 10 W / mK or greater at any temperature between 0°C and 100°C, or (2) the object is a heat pipe, a heat sink, a continuous copper body, or a continuous aluminum body, or a continuous stainless steel body. Those skilled in the art who have the benefit of this disclosure should understand that the thermal conductivity across multiple bodies (sometimes also referred to as the effective thermal conductivity) is different from the thermal conductivity across a single body. Objects or devices composed of multiple bodies are "thermally conductive" if they exhibit an effective thermal conductivity of 5 W / mK or greater across the multiple bodies. Examples of thermally conductive materials include nearly all metals and their alloys (e.g., copper, aluminum, gold, stainless steel, etc.), some plastics (e.g., TC compounds, D series thermally conductive plastics), and many other materials.

[0055] The integral heat transfer device 102, the first cover 610, and the second cover 612 may all be made of the same material or different materials. For example, the integral heat transfer device 102 may be made by printing a material (such as stainless steel) via 3D printing, while the first cover 610 and the second cover 612 may be stamped from a sheet of a metal or metal alloy (such as copper). In addition to the thermal properties discussed, factors such as cost and weight may also be considerations for material selection. For the integral heat transfer device 102, the material should be a material that can be used in a 3D printing process.

[0056] Typically, it will be desirable that at least some of these components (such as the integral heat transfer device 102, the first cover 610, and the second cover 612) be made of one or more materials that are not only thermally conductive but also highly thermally conductive. An object, device, or assembly (which may include multiple different bodies that are thermally coupled and may include multiple different materials) is "highly thermally conductive" as that term is used herein if any of the following is true: (1) the object is a continuous piece of material that has a thermal conductivity (commonly expressed as k) of 50 W / mK or greater at any temperature between 0°C and 100°C, or (2) the object is a heat pipe, a vapor chamber, a continuous copper body, or a continuous aluminum body, or a continuous stainless steel body. Examples of materials that have a thermal conductivity of 50 W / mK or greater between 0°C and 100°C include certain types of copper, aluminum, silver, gold, and stainless steel.

[0057] Now turn to Figure 6C , shows a pair of jacketed PCAs 502 mounted to a motherboard 504. More specifically, the jacketed PCAs 502 are mounted in adjacent electronic sockets 630 of a first printed circuit board 504. The sockets 630 are female connectors and are connected to the motherboard 504. Figure 1 The slots 108 shown together form a PCA slot into which the jacketed PCA 502 can be inserted and installed. Figure 6C A heat transfer tube 108 is shown between two jacketed PCAs 502. The jacketed PCAs 502 are oriented vertically relative to gravity, and the main plate 504 is oriented horizontally. This is their nominal orientation in operation, although other examples may use different orientations.

[0058] When the jacketed PCA 502 is inserted into the electronic receptacle 630, the leaf spring 510 is compressed inwardly by the surrounding components relative to the centerline 632 of each jacketed PCA 502. Figure 6C In the example of FIG. 5 , compression of leaf spring 510 is caused by the mechanical engagement of heat transfer jacket 634 with heat transfer tube 108 and the distance between the two jacketed PCAs 502. It should be noted that such compression can similarly be caused by mechanical engagement with the side rails or center rails. Compression permits jacketed PCAs 502 to be "squeezed" into the heat transfer tubes 108, side rails 114, or center rails 116 (all in FIG. 5 ). Figure 1 ) in the available space limited by the .

[0059] The leaf spring 510 compresses as the jacketed PCA 502 is inserted into the electronic socket 630 and helps establish much of the aforementioned thermal coupling. For example, the compression ensures physical engagement, and therefore thermal coupling, between the electronic component 604 and the heat transfer jacket 634, and in turn the heat transfer jacket 634 physically engages the aforementioned adjacent structures. Again, this physical engagement also establishes thermal coupling between all of these parts of the heat transfer assembly and the overall computing system.

[0060] Figure 7 The partial cross-sectional top view of the storage cooler 100 shows the Figure 5 The line 7-7 passes through Figure 5 The liquid coolant flow path of the assembly 500. It should be noted that Figure 7 The inlet hose barb 206 and the outlet hose barb 208 (both of which are firstly related to Figure 2 100). Liquid coolant is delivered to and from the storage cooler 100 by a separate subsystem (not shown). In the following discussion, the terms "top" and "bottom" are relative to Figure 7 In terms of the rendered component.

[0061] The received coolant then flows into the storage cooler 100 through the inlet 700 of the second end block 112 as indicated by the arrow graphical element 704. Next, the liquid coolant flows into the inlet chamber 200 and passes therethrough into the "top" three heat transfer tubes 108 to flow from the second end block 112 to the first end block 110 as indicated by the arrow graphical element 706. The liquid coolant flows into the circulation chamber 708 and flows downward as indicated by the arrow graphical element 710 to flow into the "bottom" three heat transfer tubes 108. Next, the liquid coolant flows through the bottom three heat transfer tubes 108 from the first block 110 to the second block 112 as indicated by the arrow graphical element 712. Next, the liquid coolant flows into the outlet chamber 202 and exits the storage cooler 100 through the outlet 702 as indicated by the arrow graphical element 714.

[0062] When the liquid coolant flows along this path, the electronic component 604 ( Fig. 6A The heat generated by the electronic component 604 is transferred to the liquid coolant. The electronic component 604 is mechanically coupled to the heat transfer jacket 634 ( Figure 6C 108), and the mechanical engagement of the heat transfer jacket 634 with the heat transfer tube 108. Each of these mechanical engagements also establishes a thermal coupling through which heat from the electronic components 604 is collectively transferred away by the liquid coolant.

[0063] Storage cooler 100 (first as Figure 11) is 3D printed as discussed above. This is one reason why the illustrated example of storage cooler 100 is made of stainless steel, a 3D printable material. (In this application, stainless steel is also thermally conductive, relatively strong, and relatively lightweight.) An overview of the 3D printing process will now be presented.

[0064] 3D printing is a form of what is generally and generally referred to as additive manufacturing. In this context, additive manufacturing processes are desirable because the storage cooler 100 is designed to minimize the number of parts to be assembled. One aspect of this constraint is that the integral heat transfer device 102 is manufactured as a single piece, as discussed above. The design of the integral heat transfer device 102 is not suitable for manufacturing in a single piece using traditional manufacturing methods (such as stamping, casting, extrusion, etc., which are commonly used in conventional processes). Additive manufacturing processes can overcome the shortcomings that affect conventional processes.

[0065] The 3D printing process manufactures the final product through one of several types of processes, such as direct metal laser sintering. The process uses a digital 3D model of the final product to "build" the final product. Metal powder is deposited one layer at a time in a selected pattern governed by the digital 3D model. Once the powder is deposited, the 3D printer changes the physical state of the powder to solidify it. For example, in direct metal laser sintering, the 3D printer uses a laser to melt the deposited metal powder, and the melted metal powder becomes solid. This continues one layer at a time until the product is built.

[0066] Figure 8 A specific example of a system 800 is shown by which the memory cooler 100 can be 3D printed. The system 800 includes at least a design facility 802 and a manufacturing facility 804 that communicate via a network 806. Those skilled in the art who benefit from this disclosure should understand that this level of discussion omits some details of the implementation of the system 800. Such omitted details are omitted for clarity so as not to obscure the subject matter claimed below. In some examples, design and manufacturing can be performed in the same facility, and in all examples, communication is not required over a network. In this example, the network 806 is a public network with a private or secure connection, such as the Internet.

[0067] Designer 808 at design station 810 develops a digital 3D model of storage cooler 100 . Fig. 9 One specific example of a design station 810 is shown. Figure 8As with the system 800 in FIG. 1 , selected routine details of the design station 810 that are not germane to the process being discussed are omitted for clarity so as not to obscure what is claimed below. The design station 810 may be, for example, a computing device that includes a programmed processing resource 900, such as a processor. The processor may be a central processing unit (“CPU”) or a processor chipset. In other examples, the processing resource 900 may alternatively be, for example, a programmed application specific integrated circuit (“ASIC”), or an electrically erasable programmable read-only memory (“EEPROM”).

[0068] In the example shown, the processing resource 900 is a CPU programmed with instructions 902 residing in a memory 904, with which the processing resource communicates via a bus 906. The memory 904 may be a random access memory ("RAM"), a read only memory ("ROM"), or more likely some combination of the two. The memory 904 also has resident thereon 3D modeling software 908 and once developed digital 3D model 910.

[0069] The designer 808 invokes the 3D modeling software 908 through a user interface 912, which includes, for example, a display 914, a keyboard 916, and a pointing device 918. The 3D modeling software 908 may be any suitable 3D modeling software known in the art. One popular 3D modeling software is SOLIDWORKS available from Dassault Systemes. TM . However, other computer-aided design ("CAD") and / or computer-aided engineering ("CAE") software packages may be used. Designer 808 then uses 3D modeling software 908 to design digital 3D model 910. It should be noted that in this example, since the memory cooler 100 includes three components, the digital 3D model 910 not only models the memory cooler 100 as a whole, but also models the three components of the memory cooler 100 separately.

[0070] Now see together Figure 8 and Fig. 9 Once the digital 3D model 910 is developed, the designer 808 outputs the digital 3D model 910 to one or more 3D printers 812 in the manufacturing facility 804. It should be appreciated by those of ordinary skill in the art that examples may include multiple 3D printers in order to achieve the desired throughput and product volume. In addition, while the discussion herein includes sequential production of the integral heat transfer device 102, the first end cap 104, and the second end cap 106, other examples may include simultaneous production of these components. For clarity, the current discussion chooses to focus on a single 3D printer used in sequential production so as not to obscure the content of the protection claimed below.

[0071] Fig.10 Shows Figure 8 A specific example of a 3D printer 812 is shown with which the system 800 may be implemented. Figure 8 System 800 and Fig. 9 As with the design station 810 in FIG. 1 , selected routine details of the 3D printer 812 that are not germane to the process being discussed are omitted for clarity so as not to obscure what is claimed below. The 3D printer includes a platform 1000 in a chamber 1002 on which a product 1004 is manufactured. In some examples, the platform 1000 can rotate on an axis 1006 driven by a motor 1010. In this example, depending on the stage of manufacturing, the product 1004 can be an integral heat transfer device 102, a first end cap 104, and a second end cap 106 (all in Figure 1 Any one of the above is shown.

[0072] The 3D printer 812 also includes a deposition subsystem and a printing subsystem. The deposition subsystem includes a deposition head 1012 mounted to a cantilever 1014, which can be driven by a motor 1010 to move in two or three degrees. Deposition material (e.g., powdered metal) is fed from a hopper 1016 to the deposition head 1012 through a feeder 1018. The printing subsystem includes a print head 1020 mounted to a cantilever 1022, which can also be driven by a motor 1010 to move in two or three degrees.

[0073] The 3D printer 812 also includes a processor-based controller 1030. It should be noted that in the computing context, the term "controller" has a clear structural meaning to those skilled in the art, namely, a processor-based arrangement that operates under program control to perform program functions. In this example, the controller 1030 includes a processing resource 1032, which can be, for example, a microcontroller. The processing resource 1032 can alternatively be a CPU, a processor chipset, an ASIC, or an EEPROM. The processing resource 1032 executes the output digital 3D model 910 (e.g., Fig. 9 The instructions may be stored in a memory (not shown) of the controller 1030 (such as Fig. 9 904), or can be obtained from a manufacturing facility 804 (such as Figure 8 Received by other components of a computing system (not shown) of the present invention.

[0074] So, in one example, Figure 1 The memory cooler 100 can be used Fig.11 The method 1100 is shown in FIG. The method 1100 begins with developing (at 1102) a memory cooler (such as Figure 1 to Figure 2 A digital three-dimensional ("3D") model of the memory cooler 100 (such as Fig. 9 The method 1100 begins with a digital 3D model 910 shown in FIG. 1100 . The method 1100 then 3D prints (at 1104 ) an integral heat transfer device, a first end cap, and a second end cap of the memory cooler. The method 1100 ends with attaching (at 1106 ) the first end cap and the second end cap to the integral heat transfer device. As described above, Fig.11 The steps presented in the and the steps recited in the following claims do not necessarily imply any particular order in which to manufacture the storage cooler. The steps may be performed sequentially in a different order or simultaneously.

[0075] Therefore, return to see Figure 1 to Figure 2 , the examples disclosed above use 3D printed stainless steel for most of the storage cooler, which has only three parts. The reduced number of parts per storage cooler significantly reduces the number of potential leak points, which is a solution that is easier to manufacture and assemble, lighter in weight, and ultimately low in cost. These examples also include a greater number of coolant channels that reduce pressure. Further still, because the overall heat transfer device is made as a unitary object, the use of TIMs and thereby improved thermal conductivity can be eliminated without the need for some kind of joining and the use of TIMs.

[0076] The hereinafter claimed subject matter also allows for variations in the flow path of the liquid coolant through the storage cooler, as discussed above. Figure 12 to Figure 14 Several examples of how the flow paths may be altered are shown. These changes in the flow paths imply changes in the structure of the overall heat transfer device, as will now be discussed.

[0077] Fig.12 1 is a top view of a partial assembly 1200 according to one or more examples, showing a coolant flow path. In this particular example, the storage cooler 100a includes an integral heat transfer device 102a, a first end block 110a, a second end block 112a, and a plurality of heat transfer tubes 108. The first end block 110a and the second end block 112a define an outlet chamber 1205 and an inlet chamber 1210 together with the first end cap 104 and the second end cap 106, respectively. The inlet chamber 1210 is in fluid communication with the inlet 1215, and the outlet chamber 1205 is in fluid communication with the outlet 1220.

[0078] The inlet chamber 1210 and the outlet chamber 1205 are each in fluid communication with each corresponding liquid flow channel of each heat transfer tube 108. The received liquid coolant enters the inlet 1215 as indicated by the arrow graphic element 1225, enters the inlet chamber 1210, flows through the heat transfer tube 108 from the inlet chamber 1210 to the outlet chamber 1205 (as indicated by the arrow graphic element 1230), and then exits through the outlet 1220 (as indicated by the arrow graphic element 1235). Therefore, the fluid flow path between the inlet 1215 and the outlet 1220 is straight rather than Figure 7 The serpentine shown. Therefore, in this example, there is no circulation chamber.

[0079] Fig.13 1 is a top view of another partial assembly 1300 according to one or more examples, showing a coolant flow path. In this particular example, the storage cooler 100b includes an integral heat transfer device 102b, a first end block 110b, a second end block 112b, and a plurality of heat transfer tubes 108. The first end block 110b, together with the first end cap 104, defines a plurality of circulation chambers 1305, each circulation chamber 1305 being in fluid communication with at least one heat transfer tube 108. The second end block 112b, together with the second end cap 106, defines an inlet chamber 1310, a plurality of circulation chambers 1315, and an outlet chamber 1320. The inlet chamber 1310, the circulation chamber 1315, and the outlet chamber 1320 are each in fluid communication with at least one liquid flow channel of the heat transfer tube 108. The inlet chamber 1310 is in fluid communication with the inlet 1325, and the outlet chamber 1320 is in fluid communication with the outlet 1330.

[0080] Therefore, in Fig.13 In the example of FIG. 1 , the received liquid coolant enters the inlet chamber 1310 through the inlet 1325 (as indicated by the arrow graphical element 1335) and flows through the heat transfer tube 108 to the first circulation chamber 1305. The circulation chambers 1305, 1315 and the heat transfer tube 108 establish a serpentine flow pattern as indicated by the arrow graphical elements 1340, 1345, 1350 and 1355. The serpentine flow path terminates at the outlet chamber 1320 where the liquid coolant exits the overall heat transfer device 100b through the outlet 1330 as indicated by the arrow graphical element 1360.

[0081] Fig.1414 is a top view of another partial assembly 1400 showing a coolant flow path according to one or more examples. In this particular example, the storage cooler 100c includes an integral heat transfer device 102c, a first end block 110c, and a second end block 112c, and a plurality of heat transfer tubes 108. The first end block 110c defines a circulation chamber 1405 and an outlet chamber 1410. The second end block 112c defines an inlet chamber 1415 and a plurality of circulation chambers 1420.

[0082] Liquid coolant enters inlet chamber 1415 through inlet 1425. Heat transfer tube 108 and circulation chambers 1420, 1405 define a serpentine flow path, as indicated by arrow graphical elements 1425, 1430, 1435. After flowing through the serpentine flow path, liquid coolant enters outlet chamber 1410 and exits integral heat transfer device 100c through outlet 1417.

[0083] Those skilled in the art who benefit from this disclosure will appreciate that in other examples, other flow paths may be implemented. The flow path may be designed by the location and arrangement of the circulation chamber (if any) and the inlet and outlet and the accompanying inlet and outlet chambers. Some examples may also employ multiple inlets and / or outlets for straight or serpentine flow paths.

[0084] Thus, the inlet chamber and the outlet chamber are each fluidly coupled to at least one corresponding liquid flow channel of at least one heat transfer tube. Fig.12 In some other embodiments, for example, the inlet chamber and the outlet chamber are fluidly coupled to the same flow channel (s) of the same heat transfer tube (s). Figure 7 and Figure 13 to Figure 14 In the serpentine flow path embodiments, the inlet chamber and the outlet chamber can be fluidly coupled to the corresponding liquid flow channels of the different heat transfer tubes. These embodiments will employ circulation chambers that are also coupled to the liquid flow channels of the various heat transfer tubes. The circulation chambers help define the serpentine flow path and permit the liquid coolant to circulate through the overall thermal device from the inlet to the outlet.

[0085] The following patent applications are incorporated herein by reference to the extent not inconsistent with the disclosure herein, with reference to the disclosures therein relating to jacketed PCAs (such as Figure 5 Instructions for the design and manufacture of jacketed PCA 502) in:

[0086] U.S. application serial number 17 / 237,732, filed on April 21, 2021, in the name of inventors John Franz and Ernesto Ferrer and entitled “Temperature Control of Closely Packed Electronic Assemblies,” which application is commonly assigned herein.

[0087] In the event of a conflict between the incorporated application and the present disclosure, the present specification, including definitions, will control.

[0088] This concludes the detailed description. The specific examples disclosed above are illustrative only, as the invention may be modified and implemented in different but equivalent manners apparent to those skilled in the art having the benefit of the teachings herein. Furthermore, no limitation is intended to the details of construction or design shown herein except as described in the appended claims. It is therefore apparent that the specific examples disclosed above may be changed or modified, and all such variations are considered to be within the scope and spirit of the invention. Therefore, the protection sought herein is as described in the claims below.

Claims

1. A storage cooler, comprising: An integral heat transfer device comprising: a plurality of heat transfer tubes, each heat transfer tube having a first tube end and a second tube end and defining a respective liquid flow passage; a first end block structurally integrated with each of the heat transfer tubes at a corresponding first tube end of the heat transfer tube; a second end block structurally integrated with each of the heat transfer tubes at a corresponding second tube end; an inlet chamber defined in part by the second end block, the inlet chamber fluidly coupled to at least one corresponding liquid flow channel of at least one of the heat transfer tubes; an inlet leading to the inlet chamber; an outlet chamber defined in part by the first end block, the outlet chamber fluidly coupled to at least one corresponding liquid flow passage of at least one of the heat transfer tubes; and an exit, leaving the exit chamber; a first end cap attached to the first end block to define, together with the first end block, the outlet chamber and at least a first circulation chamber; and A second end cap is attached to the second end block to define the inlet chamber and at least a second circulation chamber together with the second end block.

2. The storage cooler according to claim 1, wherein: The at least one respective liquid flow channel fluidly coupled to the inlet chamber is the at least one respective liquid flow channel fluidly coupled to the outlet chamber.

3. The storage cooler of claim 1, wherein: at least one respective liquid flow channel of the at least one heat transfer tube is defined by a respective inner surface of the at least one heat transfer tube; and The inner surface includes a plurality of surface extensions.

4. The storage cooler according to claim 1, wherein: The at least one respective liquid flow channel has a rounded rectangular cross-section.

5. The storage cooler of claim 1, wherein: Each of the heat transfer tubes is structurally integrated to the first end block at the respective first tube end and is integrated to the second end block at the respective second tube end.

6. A component comprising: Main printed circuit board; A memory cooler mounted to the main printed circuit board, the memory cooler comprising: An integral heat transfer device comprising: a plurality of heat transfer tubes, each heat transfer tube having a first tube end and a second tube end and defining a respective liquid flow passage; a first end block structurally integrated with each of the heat transfer tubes at a corresponding first tube end of the heat transfer tube; a second end block structurally integrated with each of the heat transfer tubes at a corresponding second tube end; an inlet chamber defined in part by the second end block, the inlet chamber fluidly coupled to at least one corresponding liquid flow channel of at least one of the heat transfer tubes; an inlet leading to the inlet chamber; an outlet chamber defined in part by the first end block, the outlet chamber fluidly coupled to at least one corresponding liquid flow passage of at least one of the heat transfer tubes; and an exit from the exit chamber; and a first end cap attached to the first end block to define, together with the first end block, the outlet chamber and at least a first circulation chamber; a second end cap attached to the second end block to define, together with the second end block, the inlet chamber and at least a second circulation chamber, and A plurality of storage boards are inserted into the main board, and each storage board is arranged in a corresponding slot defined by the heat transfer pipe of the integral heat transfer device.

7. The assembly of claim 6, wherein: The at least one respective liquid flow channel of the at least one heat transfer tube is defined by a respective inner surface of the at least one heat transfer tube; and The inner surface includes a plurality of surface extensions.

8. The assembly of claim 6, wherein: The at least one respective liquid flow channel has a rounded rectangular cross-section.

9. The assembly of claim 6, wherein: Each of the heat transfer tubes is structurally integrated to the first end block at the respective first tube end and is integrated to the second end block at the respective second tube end.

10. The assembly of claim 6, further comprising a central processing unit inserted into the main printed circuit board.

11. A storage cooler comprising: An integral heat transfer device comprising: a plurality of heat transfer tubes, each heat transfer tube having a first tube end and a second tube end and defining a respective liquid flow passage; a first end block structurally integrated with each of the heat transfer tubes at a corresponding first tube end of the heat transfer tube; a second end block structurally integrated with each of the heat transfer tubes at a corresponding second tube end; an inlet chamber defined in part by the second end block, the inlet chamber fluidly coupled to at least one corresponding liquid flow channel of at least one of the heat transfer tubes; an inlet leading to the inlet chamber; an outlet chamber defined in part by the second end block, the outlet chamber fluidly coupled to at least one corresponding liquid flow channel of at least one of the heat transfer tubes; and an exit, leaving the exit chamber; a first end cap attached to the first end block to define at least a first circulation chamber and a second circulation chamber together with the first end block; and A second end cap is attached to the second end block to define the outlet chamber, the inlet chamber and at least a third circulation chamber together with the second end block.

12. The storage cooler of claim 11, wherein: The at least one respective liquid flow channel fluidly coupled to the inlet chamber is the at least one respective liquid flow channel fluidly coupled to the outlet chamber.

13. The storage cooler of claim 11, wherein: at least one respective liquid flow channel of the at least one heat transfer tube is defined by a respective inner surface of the at least one heat transfer tube; and The inner surface includes a plurality of surface extensions.

14. The storage cooler of claim 11, wherein: The at least one respective liquid flow channel has a rounded rectangular cross-section.

15. The storage cooler of claim 11, wherein: Each of the heat transfer tubes is structurally integrated to the first end block at the respective first tube end and is integrated to the second end block at the respective second tube end.

Images