Apparatus and system for holding printed circuit board between mounting plates

Abstract

Systems, apparatus, articles, and methods for holding a printed circuit board between mounting plates are disclosed. An example mount includes: a first segment adjacent a first surface of a metal plate; and a second segment extending from the first segment toward a second surface of the printed circuit board (PCB), the support for maintaining a separation between the metal plate and the PCB. Further, the example mount includes at least one of: a biasing assembly to suppress movement between the metal plate and the PCB; or a distal end on the second section, the first section having a first cross-sectional shape defining a first region, the distal end having a second cross-sectional shape defining a second region, the second region being smaller than the first region.

Description

Related applications

[0001] This patent originates from International Application No. PCT / CN2024 / 114455, filed on August 26, 2024. International Application No. PCT / CN2024 / 114455 is hereby incorporated herein by reference in its entirety. Priority to International Application No. PCT / CN2024 / 114455 is hereby claimed. Background Technology

[0002] A data center is a dedicated space (e.g., within one or more buildings) that houses computer systems and associated components (e.g., telecommunications systems, storage systems, etc.). Data centers can vary in size, power requirements, redundancy, and overall architecture. For example, data centers can be implemented as on-premises data centers (e.g., data centers at a business's operating premises), colocation data centers (e.g., data centers that provide computing resources for rent), hyperscale data centers (e.g., relatively large data centers that provide scalability for large workloads), and edge data centers (e.g., relatively small data centers located near the edge of a network). Attached Figure Description

[0003] Figure 1 The diagram illustrates one or more example environments in which the teachings of this disclosure can be implemented.

[0004] Figure 2 The figure illustrates at least one example of a data center used to perform workloads by utilizing decomposed resources.

[0005] Figure 3 The diagram illustrates what can be included. Figure 2 At least one example of a cabin in a data center.

[0006] Figure 4 It can be included Figure 3 A perspective view of at least one example of a rack in the cabin.

[0007] Figure 5 yes Figure 4 Side elevation view of the frame.

[0008] Figure 6 yes Figure 4 A perspective view of a frame with skids mounted therein.

[0009] Figure 7 yes Figure 6 A block diagram of at least one example of the top side of the skid.

[0010] Figure 8 yes Figure 7 A block diagram of at least one example of the underside of a skid.

[0011] Figure 9 It can be used Figure 2 A block diagram of at least one example of a computing skid for a data center.

[0012] Figure 10 yes Figure 9 A top-view perspective of at least one example of a calculation skid.

[0013] Figure 11 It can be used Figure 2 A block diagram of at least one example of an accelerator skid for a data center.

[0014] Figure 12 yes Figure 11 A top-down perspective view of at least one example of an accelerator skid.

[0015] Figure 13 It can be used Figure 2 A block diagram of at least one example of a storage skid for a data center.

[0016] Figure 14 yes Figure 13 A top perspective view of at least one example of a storage skid.

[0017] Figure 15 It can be used Figure 2 A block diagram of at least one example of a data center storage skid.

[0018] Figure 16 It is possible Figure 2 A block diagram of a system built within a data center to perform workloads using managed nodes that utilize decomposed resources.

[0019] Figure 17A The figure shows an isometric view of an example Open Compute Project (OCP) accelerator module (OAM).

[0020] Figure 17B yes Figure 17A An example of an OAM breakdown view.

[0021] Figure 18A The figure shows a cross-sectional view of the first example mounting plate, the example printed circuit board, and the second example mounting plate (collectively referred to as the example OAM card) mounted on the example system chassis.

[0022] Figure 18B The figure shows a cross-sectional view of a sample OAM card when the fasteners are installed onto the system chassis.

[0023] Figure 19 The diagram shows Figure 17A and Figure 17B A side view of an example OAM card, which includes a first support and a second support when assembled.

[0024] Figure 20 The diagram shows Figure 17A and Figure 17B The first example implementation when the first support is assembled in the example OAM card.

[0025] Figure 21A yes Figure 20 Isometric view of the cross section of the first support.

[0026] Figure 21B yes Figure 20 An isometric view of the first support.

[0027] Figure 21C yes Figure 20 The cross-sectional side view of the first support.

[0028] Figure 22 The diagram shows Figure 17A and Figure 17B A cross-sectional view of the second example implementation of the first support.

[0029] Figure 23 The diagram shows Figure 17A and Figure 17B The third example implementation when the first support is assembled in the example OAM card.

[0030] Figure 24A yes Figure 23 Isometric view of the first support.

[0031] Figure 24B yes Figure 23 Side view of the first support.

[0032] Figure 24C yes Figure 23 Another isometric view of the first support.

[0033] Figure 25A The diagram shows Figure 17A and Figure 17B A cross-sectional view of the fourth example implementation when the first support is assembled in the OAM card.

[0034] Figure 25B The diagram shows Figure 25A An isometric view of the first support.

[0035] Figure 26 The diagram shows Figure 17A and Figure 17B A cross-sectional view of the fifth example implementation of the first support.

[0036] Figure 27 The diagram shows Figure 17A and Figure 17B A cross-sectional view of the sixth example implementation of the first support.

[0037] Figure 28A The diagram shows Figure 17A and Figure 17B A cross-sectional view of the seventh example implementation of the first support.

[0038] Figure 28B The diagram shows Figure 28A An isometric view of the first support.

[0039] Figure 29 The diagram shows Figure 17A and Figure 17B A cross-sectional view of the eighth example implementation of the first support.

[0040] Generally, the same reference numerals will be used throughout the accompanying drawings and written description to refer to the same or similar parts. The drawings are not necessarily drawn to scale. Instead, the thickness of layers or regions may be enlarged in the drawings. While the drawings show layers and regions with clearly defined lines and boundaries, some or all of these lines and / or boundaries may be idealized. In reality, boundaries and / or lines may be unobservable, mixed, and / or irregular. Detailed Implementation

[0041] Large data centers (e.g., colocation data centers, hyperscale data centers, etc.) are industrial-scale operations that can use as much electricity as a small town. Many data centers operate across the globe to meet the processing demands of modern technology infrastructure. Therefore, operating data centers requires a significant amount of electricity. To improve the efficiency of operating data centers, industry leaders formed the Open Compute Project (OCP) to develop more energy-efficient data center computing technologies. For example, making servers taller allows space for more efficient heatsinks and allows fans to move more air with less energy. Furthermore, by utilizing specially designed servers, energy consumption can be reduced due to unnecessary expansion slots on motherboards and unnecessary components such as graphics cards.

[0042] Figure 1 The diagram illustrates one or more example environments in which the teachings of this disclosure can be implemented. Figure 1 One or more example environments may include one or more central data centers 102. These central data centers 102 may store, for example, a large number of servers used by one or more organizations for data processing, storage, etc. Figure 1As illustrated, one or more central data centers 102 include multiple immersion tanks 104 to facilitate cooling of servers and / or other electronic components stored at the central data centers 102. The immersion tanks 104 may provide single-phase or two-phase cooling.

[0043] Figure 1 An example environment could be part of an edge computing system. For example, Figure 1 Example environments may include edge data centers or micro data centers 106. One or more edge data centers 106 may include, for example, data centers located at the base of a cellular tower. In some examples, one or more edge data centers 106 are located on or near the top of a cellular tower and / or other utility pole. One or more edge data centers 106 include corresponding housings for storing one or more servers, wherein the servers can communicate with, for example, one or more servers, client devices, and / or other computing devices in the edge network stored at one or more central data centers 102. Example housings of one or more edge data centers 106 may include materials forming one or more outer surfaces that partially or completely protect the contents therein, wherein protection may include weather protection, hazardous environmental protection (e.g., electromagnetic interference (EMI), vibration, extreme temperatures), and / or enable immersion in water. Example housings may include power circuitry systems for providing power for stationary and / or portable implementations, such as alternating current (AC) power inputs, direct current (DC) power inputs, one or more AC / DC or DC / AC converters, power conditioners, transformers, charging circuitry systems, batteries, wired inputs, and / or wireless power inputs. Figure 1 As illustrated, one or more edge data centers 106 may include one or more immersion tanks 108 for storing one or more servers and / or one or more other electronic components located at one or more edge data centers 106.

[0044] Figure 1 One or more example environments may include building 110 for commercial and / or industrial purposes, which stores information technology (IT) equipment, for example, in one or more rooms of building 110. Figure 1As indicated, one or more servers 112 may be stored together with one or more server racks 114 supporting the servers 112 (e.g., in the openings of slots in rack 114). In some examples, the one or more servers 112 located at building 110 include one or more internal servers of an edge computing network, wherein the one or more internal servers communicate with one or more remote servers (e.g., one or more servers at edge data center 106) and / or one or more other computing devices within the edge network.

[0045] Figure 1 One or more example environments include one or more content delivery network (CDN) data centers 116. The one or more CDN data centers 116 in this example include one or more servers 118 that cache content such as images, web pages, videos, etc., accessed via user devices. The one or more servers 118 of the CDN data center 116 may be located in one or more immersion cooling tanks (such as immersion tanks 104, 108 shown in relation to data centers 102, 106).

[0046] In some instances, Figure 1 Example data centers 102, 106, 116, and / or one or more buildings 110 include servers and / or other electronic components cooled independently of immersion tanks (e.g., immersion tanks 104, 108) and / or associated immersion cooling systems. That is, in some examples, some or all of the servers and / or other electronic components in data centers 102, 106, 116, and / or one or more buildings 110 can be cooled by air and / or liquid coolant without immersing the servers and / or other electronic components therein. Thus, in some examples, Figure 1 The immersion tanks 104 and 108 can be omitted. Furthermore, Figure 1 Example data centers 102, 106, 116, and / or (one or more) buildings 110 can be combined with the following: Figures 2-16 The example data center 200 is described in further detail as corresponding to, implemented by, and / or adapted from the example data center 200.

[0047] Although the diagrams show a number of cooling tanks and other components, any number of such components may be present. Furthermore, the examples of cooling data centers and / or other structures or environments disclosed herein are not limited to... Figure 1The arrangement of the sizes depicted herein. For example, a structure including the example cooling system and / or components disclosed herein may have openings for accommodating service personnel (such as... Figure 1 The size of the example data center(s) 106(s) can be smaller (e.g., a "doghouse" enclosure). For example, the size of a structure containing the example cooling systems and / or components disclosed herein can be adjusted such that the access to the interior of the structure (e.g., the only access) is the port through which service personnel enter the structure. In some examples, the size of a structure containing the example cooling systems and / or components disclosed herein can be adjusted such that only tools can enter the enclosure, as the structure may be supported by, for example, a utility pole or radio tower, or a larger structure.

[0048] Figure 2 The illustration depicts an example data center 200 in which the decomposed resources can collaboratively perform one or more workloads (e.g., applications representing clients). The illustrated data center 200 includes multiple platforms 210, 220, 230, and 240 (referred to herein as cabins), each platform comprising one or more rack rows. Although the data center 200 is shown as having multiple cabins, in some examples, the data center 200 may be implemented as a single cabin. As described in more detail herein, a rack may house multiple skids. Skids may be primarily equipped with a specific type of resource (e.g., memory devices, data storage devices, accelerator devices, general-purpose programmable circuit systems), which may be logically coupled to form composite nodes. Some of these nodes may act as, for example, servers. In the illustrative example, skids in cabins 210, 220, 230, and 240 are connected to multiple cabin switches (e.g., switches that route data communications to and from skids within cabins). The cabin switches then connect to backbone switch 250, which switches communication between cabins (e.g., cabins 210, 220, 230, 240) within data center 200. In some examples, the skids may use Intel Omni-Path. TMTechnology and fabric connectivity. In other examples, skids can be connected to other fabric, such as InfiniBand or Ethernet. As described in more detail herein, resources within a skid in data center 200 can be assigned to a group (referred to herein as a “managed node”) containing resources from one or more skids that are shared for use in the execution of a workload. A workload can be executed as if the resources belonging to a managed node were located on the same skid. Resources in a managed node can belong to skids belonging to different racks, and even to different compartments 210, 220, 230, 240. Thus, some resources of a single skid can be assigned to a managed node, while other resources of the same skid are assigned to different managed nodes (e.g., a first programmable circuit system is assigned to a managed node, while a second programmable circuit system of the same skid is assigned to a different managed node).

[0049] Data centers that include decomposed resources (such as Data Center 200) can be used in a variety of scenarios, such as enterprises, governments, cloud service providers and communications service providers (e.g., telecom providers), and in a variety of sizes, from giant data centers of cloud service providers consuming more than 200,000 square feet to single-rack or multi-rack installations for use in base stations.

[0050] In some examples, resource decomposition is achieved by using skids that primarily comprise a single type of resource (e.g., a compute skid primarily comprising compute resources, a memory skid primarily comprising memory resources). This decomposition of resources, and the selective allocation and deallocation of resources used to form managed nodes assigned to perform workloads, improves the operation and resource utilization of a data center 200 relative to a typical data center. Such a typical data center includes hyperconverged servers that contain compute, memory, storage, and possibly additional resources in a single chassis. For example, since a given skid will primarily contain the same specific type of resource, that type of resource can be upgraded independently of other resources. Additionally, since different resource types (programmable circuitry systems, storage devices, accelerators, etc.) typically have different refresh rates, greater resource utilization and a lower total cost of ownership can be achieved. For example, a data center operator can upgrade programmable circuitry systems throughout the facility simply by swapping out a compute skid. In such cases, accelerators and storage resources may not be upgraded simultaneously but can be allowed to continue operating until those resources are scheduled for their own refresh. Resource utilization can also be increased. For example, if managed nodes are configured based on the requirements of the workloads that will run on them, the resources within a node are more likely to be fully utilized. Such utilization can allow more managed nodes to run in a data center with a given set of resources, or allow a data center expected to run a given set of workloads to be built with fewer resources.

[0051] Now for reference Figure 3In the illustrative example, bay 210 includes a set of rows 300, 310, 320, and 330 of racks 340. Each rack in rack 340 can accommodate multiple skids (e.g., sixteen skids) and provide power and data connectivity to the accommodated skids, as described in more detail herein. In the illustrative example, the racks are connected to multiple bay switches 350 and 360. Bay switch 350 includes a set 352 of ports to which the skids of bay 210's racks are connected, and another set 354 of ports connecting bay 210 to backbone switch 250 to provide connectivity to other bays in data center 200. Similarly, bay switch 360 includes a set 362 of ports to which the skids of bay 210's racks are connected, and a set 364 of ports connecting bay 210 to backbone switch 250. Thus, a certain amount of redundancy is provided to bay 210 using a pair of switches 350 and 360. For example, if any of switches 350 and 360 fails, the skid in compartment 210 can still maintain data communication with the rest of data center 200 (e.g., skids in other compartments) via another switch 350 or 360. Furthermore, in the illustrative example, switches 250, 350, and 360 can be implemented as dual-mode optical switches capable of routing both Ethernet protocol communication carrying Internet Protocol (IP) packets and communication according to a second high-performance link layer protocol (e.g., Peripheral Component Interconnect (PCI) Express (PCIe)) via optical signaling media of an optical structure.

[0052] It should be understood that any of the other cabins 220, 230, 240 (and any additional cabins in data center 200) can be similarly constructed as follows: Figure 3 Shown and about Figure 3 The disclosed compartment 210, and compartment 210 having similar components (e.g., a given compartment may have rows of racks accommodating multiple skids as described above). Additionally, although two compartment switches 350, 360 are shown, it should be understood that in other examples, different numbers of compartment switches may be presented, thus providing even more failover capabilities. In other examples, it may be with... Figure 2 and Figure 3 The rack row configurations shown in the diagram arrange the compartments differently. For example, a compartment may include multiple radially arranged rack sets (e.g., racks equidistant from a central switch).

[0053] Figures 4-6The illustration shows an example rack 340 of a data center 200. As shown in the illustrated example, rack 340 includes two vertically arranged elongated support columns 402, 404. For example, when deployed, the elongated support columns 402, 404 can extend upwards from the floor of the data center 200. Rack 340 also includes one or more horizontal pairs of elongated support arms 412 (in... Figure 4 (Identified by a dashed ellipse), which is configured to support the skid of the data center 200, as discussed below. One of a pair of elongated support arms 412 extends outward from the elongated support column 402, and the other elongated support arm 412 extends outward from the elongated support column 404.

[0054] In the illustrative example, at least some of the skids in data center 200 are chassis-less skids. That is, such skids have a chassis-less circuit board substrate on which physical resources (e.g., programmable circuit systems, memory, accelerators, storage devices, etc.) are mounted, as discussed in more detail below. Thus, rack 340 is configured to receive chassis-less skids. For example, a given pair 410 of elongated support arms 412 defines skid slots 420 in rack 340, which are configured to receive corresponding chassis-less skids. For this purpose, elongated support arms 412 include corresponding circuit board guides 430, which are configured to receive the chassis-less circuit board substrate of the skid. Circuit board guides 430 are secured to, or otherwise mounted to, the top side 432 of the corresponding elongated support arm 412. For example, in the illustrative example, circuit board guides 430 are mounted relative to the corresponding elongated support posts 402, 404 at the distal end of the corresponding elongated support arm 412. Figures 4-6 For clarity, not every circuit board guide 430 is referenced in every drawing. In some examples, at least some of the skids include a chassis, and the frame 340 is adapted to receive the chassis.

[0055] The circuit board guide 430 includes an inner wall defining a circuit board slot 480, which is configured to receive the chassisless circuit board substrate of the skid 500 when the skid 500 is received in a corresponding skid slot 420 of the frame 340. For this purpose, as... Figure 5 As shown, the user (or robot) aligns the chassisless circuit board substrate of the illustrative chassisless skid 500 with the skid slot 420. The user or robot can then slide the chassisless circuit board substrate forward into the skid slot 420, such that each side edge 514 of the chassisless circuit board substrate is received in a corresponding circuit board slot 480 of a pair 410 of elongated support arms 412 defining a corresponding skid slot 420, as... Figure 5As shown in the diagram, different types of resources can be upgraded independently of each other and at their own optimized refresh rates via skids that include decomposed resources and are robot-accessible and manipulable. Furthermore, the skids are configured for power and data communication cables in the blind-fit rack 340, thereby enhancing their ability to be quickly removed, upgraded, reinstalled, and / or replaced. Thus, in some examples, data center 200 can operate on data center floors without human intervention (e.g., performing workloads, conducting maintenance, and / or upgrades). In other examples, humans can facilitate one or more maintenance or upgrade operations within data center 200.

[0056] It should be understood that the circuit board guide 430 is double-sided. That is, the circuit board guide 430 includes an inner wall that defines a circuit board slot 480 on each side of the circuit board guide 430. In this way, the circuit board guide 430 can support a chassisless circuit board substrate on either side. Thus, a single additional elongated support post can be added to the rack 340 to transform the rack 340 into a two-rack solution that can accommodate twice the size of... Figure 4The number of skid slots 420 shown is illustrated. The illustrative rack 340 includes seven pairs 410 of elongated support arms 412 defining seven corresponding skid slots 420. The skid slots 420 are configured to receive and support corresponding skids 500 as discussed above. In other examples, the rack 340 may include additional or fewer pairs of elongated support arms 412 (e.g., additional or fewer skid slots 420). It should be understood that because the skid 500 is chassis-less, it can have a different overall height than a typical server. Thus, in some examples, the height of a given skid slot 420 may be shorter than the height of a typical server (e.g., shorter than a single rank unit, referred to as "1U"). That is, the vertical distance between pairs 410 of the elongated support arms 412 may be less than a standard rack unit "1U". Additionally, due to the relatively reduced height of the skid slots 420, in some examples, the overall height of rack 340 may be shorter than that of a conventional rack enclosure. For example, in some examples, the elongated support columns 402, 404 may have a length of six feet or less. Similarly, in other examples, the support 340 may have different dimensions. For example, in some examples, the vertical distance between pairs 410 of elongated support arms 412 may be greater than that of a standard rack unit "1U". In such examples, the increased vertical distance between the skids allows for the attachment of larger heat sinks to the physical resource and allows for the use of larger fans (e.g., in the fan array 470 described below) to cool the skids, which in turn allows the physical resource to operate at an increased power level. Furthermore, it should be understood that rack 340 does not include any walls, enclosures, etc. Instead, rack 340 is an enclosure-less rack open to the local environment. In some cases, where rack 340 forms a row-end rack in data center 200, end plates may be attached to one of the elongated support columns 402, 404.

[0057] In some examples, various interconnects may be routed upwards or downwards through elongated support posts 402, 404. To facilitate such routing, elongated support posts 402, 404 include inner walls defining internal chambers in which interconnects may be located. Interconnects routed through elongated support posts 402, 404 may be implemented as any type of interconnect, including but not limited to data or communication interconnects for providing a communication connection to skid slot 420, power interconnects for providing power to skid slot 420, and / or other types of interconnects.

[0058] In the illustrative example, rack 340 includes a support platform on which corresponding optical data connectors (not shown) are mounted. These optical data connectors are associated with corresponding skid slots 420 and are configured to mate with the optical data connectors of the corresponding skid 500 when the skid 500 is received in the corresponding skid slot 420. In some examples, optical connections between components in data center 200 (e.g., skids, racks, and switches) are achieved via blind-fit optical connections. For example, a door on a given cable can prevent dust from contaminating the optical fibers inside the cable. During connection to a blind-fit optical connector mechanism, the door is pushed open as the end of the cable approaches or enters the connector mechanism. The optical fibers inside the cable can then enter the gel within the connector mechanism, and the optical fibers of one cable can contact the optical fibers of another cable within the gel inside the connector mechanism.

[0059] The illustrative rack 340 also includes a fan array 470 coupled to cross support arms of the rack 340. The fan array 470 includes one or more rows of cooling fans 472 aligned on a horizontal line between elongated support columns 402, 404. In the illustrative example, the fan array 470 includes a row of cooling fans 472 for different skid slots 420 of the rack 340. As discussed above, in the illustrative example, the skid 500 does not include any onboard cooling system, and thus, the fan array 470 provides cooling for such skids 500 received in the rack 340. In other examples, some or all of the skids 500 may include onboard cooling systems. Further, in some examples, the skids 500 and / or the rack 340 may include and / or incorporate liquid and / or immersion cooling systems to facilitate cooling of one or more electronic components on the skids 500. In the illustrative example, the rack 340 also includes different power supplies associated with the different skid slots in the skid slots 420. A given power source is secured to one of a pair of elongated support arms 412 defining a corresponding skid slot 420. For example, rack 340 may include a power source coupled or secured to each of the elongated support arms 412 extending from the elongated support post 402. The given power source includes a power connector configured to mate with the power connector of skid 500 when skid 500 is received in the corresponding skid slot 420. In the illustrative example, skid 500 does not include any onboard power supply, and thus, when mounted to rack 340, power supplied in rack 340 supplies power to the corresponding skid 500. The given power source is configured to meet the power requirements of its associated skid, which may differ between skids. Additionally, the power sources disposed in rack 340 may operate independently of each other. In other words, within a single rack, a first power source supplying power to the computing skid can provide a different power level than a second power source supplying power to the accelerator skid. The power source can be controllable at the skid level or rack level, and can be controlled locally by components on the associated skid or remotely, such as by another skid or orchestrator.

[0060] Now for reference Figure 7 In the illustrative example, as discussed above, skid 500 is configured for installation in a corresponding rack 340 of data center 200. In some examples, a given skid 500 may be optimized or otherwise configured to perform specific tasks (such as compute tasks, acceleration tasks, data storage tasks, etc.). For example, skid 500 may be implemented as described below. Figure 9 and Figure 10 The calculation skid 900 discussed below, regarding Figure 11 and Figure 12 The accelerator skid 1100 discussed below, and the following about Figure 13 and Figure 14 The storage skid 1300 discussed may also be implemented as a skid optimized or otherwise configured to perform other specialized tasks (such as those described below). Figure 15 The memory skid 1500 discussed.

[0061] As discussed above, the illustrative skid 500 includes a chassis-less circuit board substrate 702 that supports various physical resources (e.g., electrical components) mounted thereon. It should be understood that the circuit board substrate 702 is "chassis-less" because the skid 500 does not include a housing or enclosure. Instead, the chassis-less circuit board substrate 702 is open to the local environment. The chassis-less circuit board substrate 702 can be formed of any material capable of supporting the various electrical components mounted thereon. For example, in the illustrative example, the chassis-less circuit board substrate 702 is formed of FR-4 glass-reinforced epoxy laminate. In other examples, other materials may be used to form the chassis-less circuit board substrate 702.

[0062] As discussed in more detail below, the chassisless circuit board substrate 702 includes several features that improve the thermal cooling characteristics of various electrical components mounted on it. As discussed, the chassisless circuit board substrate 702 does not include a housing or enclosure, which can improve airflow over the electrical components of the skid 500 by reducing those structures that might obstruct airflow. For example, since the chassisless circuit board substrate 702 is not positioned in a separate housing or enclosure, there is no vertically arranged backplate (e.g., a chassis backplate) attached to it that could obstruct airflow across the electrical components. Additionally, the chassisless circuit board substrate 702 has a geometry configured to reduce the length of the airflow path across the electrical components mounted to it. For example, the illustrative chassisless circuit board substrate 702 has a width 704 greater than the depth 706 of the chassisless circuit board substrate 702. In a specific example, the chassisless circuit board substrate 702 has a width of approximately 21 inches and a depth of approximately 9 inches, compared to a typical server with a width of approximately 17 inches and a depth of approximately 39 inches. Consequently, the airflow path 708 extending from the front edge 710 to the rear edge 712 of the chassisless circuit board substrate 702 has a shorter distance compared to a typical server, which can improve the thermal cooling characteristics of the skid 500. Furthermore, as discussed in more detail below, although not in... Figure 7As illustrated in the diagram, but in this example, the various physical resources mounted to the chassisless circuit board substrate 702 are positioned such that no two substantially heat-generating electrical components are shaded from each other. That is, two electrical components that do not generate significant heat during operation (e.g., more than the nominal heat sufficient to adversely affect the cooling of another electrical component) are mounted on the chassisless circuit board substrate 702, which is linearly aligned with each other along the direction of the airflow path 708 (e.g., along the direction extending from the front edge 710 towards the rear edge 712 of the chassisless circuit board substrate 702). When cooling (one or more) of the electrical components via liquid (e.g., one-phase or two-phase cooling), the placement and / or structure of the features can be appropriately adjusted.

[0063] As discussed above, the illustrative skid 500 includes one or more physical resources 720 mounted to the top side 750 of the chassisless circuit board substrate 702. Although in Figure 7 Two physical resources 720 are shown, but it should be understood that in other examples, skid 500 may include one, two, or more physical resources 720. Physical resources 720 can be implemented as any type of programmable circuit system, controller, or other computing circuit capable of performing various tasks (such as computational functions) and / or controlling the functionality of skid 500 depending on, for example, the type or intended function of skid 500. For example, as discussed in more detail below, in an example where skid 500 is implemented as a computing skid, physical resource 720 may be implemented as a high-performance processor circuit system; in an example where skid 500 is implemented as an accelerator skid, physical resource 720 may be implemented as an accelerator coprocessor circuit system or circuit; in an example where skid 500 is implemented as a storage skid, physical resource 720 may be implemented as a storage controller; or in an example where skid 500 is implemented as a memory skid, physical resource 720 may be implemented as a set of memory devices.

[0064] Skid 500 also includes one or more additional physical resources 730 mounted to the top side 750 of the chassisless circuit board substrate 702. In the illustrative example, as discussed in more detail below, the additional physical resource includes a network interface controller (NIC). Depending on the type and function of skid 500, in other examples, physical resource 730 may include additional or other electrical components, circuitry, and / or devices.

[0065] Physical resource 720 is communicatively coupled to physical resource 730 via input / output (I / O) subsystem 722. I / O subsystem 722 may be implemented as a circuit system and / or component for facilitating input / output operations of physical resource 720, physical resource 730, and / or other components of skid 500. For example, I / O subsystem 722 may be implemented as or otherwise include a memory controller hub, input / output control hub, integrated sensor hub, firmware device, communication link (e.g., point-to-point link, bus link, line, cable, waveguide, optical guide, printed circuit board trace, etc.) and / or other components and subsystems for facilitating input / output operations. In the illustrative example, I / O subsystem 722 is implemented as or otherwise includes a double data rate (DDR) data bus (such as a DDR4 or DDR5 data bus).

[0066] In some examples, skid 500 may also include a resource-to-resource interconnect 724. The resource-to-resource interconnect 724 can be implemented as any type of communication interconnect capable of facilitating resource-to-resource communication. In illustrative examples, the resource-to-resource interconnect 724 is implemented as a high-speed point-to-point interconnect (e.g., faster than I / O subsystem 722). For example, the resource-to-resource interconnect 724 can be implemented as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to resource-to-resource communication.

[0067] Skid 500 also includes a power connector 740 configured to mate with a corresponding power connector of rack 340 when skid 500 is mounted in corresponding rack 340. Skid 500 receives power from the power supply of rack 340 via power connector 740 to power various electrical components of skid 500. That is, skid 500 does not include any local power supply (e.g., onboard power) for supplying power to the electrical components of skid 500. Excluding local or onboard power helps reduce the overall footprint of chassis-less circuit board substrate 702, which can increase the thermal cooling characteristics of the various electrical components mounted on chassis-less circuit board substrate 702, as discussed above. In some examples, voltage regulators are placed with programmable circuit system 920 (see...). Figure 9 The bottom side 850 of the directly opposite chassis-less circuit board substrate 702 (see...) Figure 8On the circuit board substrate 702, power is routed from the voltage regulator to the programmable circuit system 920 via vias extending through the circuit board substrate 702. Compared to a typical printed circuit board where processor power is delivered from the voltage regulator partially through printed circuit traces, this configuration provides an increased thermal budget, additional current and / or voltage, and better voltage control.

[0068] In some examples, skid 500 may also include mounting features 742 configured to mate with a robot's mounting arm or other structure to facilitate the robot's placement of skid 500 into frame 340. Mounting features 742 can be implemented as any type of physical structure that allows the robot to grip skid 500 without damaging the chassis-less circuit board substrate 702 or the electrical components mounted thereon. For example, in some examples, mounting features 742 may be implemented as non-conductive pads attached to the chassis-less circuit board substrate 702. In other examples, mounting features may be implemented as brackets, straps, or other similar structures attached to the chassis-less circuit board substrate 702. The specific number, shape, size, and / or configuration of mounting features 742 may depend on the design of the robot configured to manage skid 500.

[0069] Now for reference Figure 8 In addition to physical resources 730 mounted on the top side 750 of the chassisless circuit board substrate 702, skid 500 also includes one or more memory devices 820 mounted to the bottom side 850 of the chassisless circuit board substrate 702. That is, the chassisless circuit board substrate 702 is implemented as a dual-sided circuit board. Physical resources 720 are communicatively coupled to memory devices 820 via I / O subsystem 722. For example, physical resources 720 and memory devices 820 may be communicatively coupled by one or more vias extending through the chassisless circuit board substrate 702. In some examples, different physical resources in physical resources 720 may be communicatively coupled to different sets of one or more memory devices 820. Alternatively, in other examples, different physical resources in physical resources 720 may be communicatively coupled to the same memory device in memory devices 820.

[0070] The memory device 820 can be implemented as any type of memory device capable of storing data for physical resource 720 during operation of skid 500, such as any type of volatile memory (e.g., dynamic random access memory, DRAM, etc.) or non-volatile memory. Volatile memory can be a storage medium that requires power to maintain the state of the data stored by the medium. Non-limiting examples of volatile memory can include various types of random access memory (RAM), such as dynamic random access memory (DRAM) or static random access memory (SRAM). One particular type of DRAM that can be used in a memory module is synchronous dynamic random access memory (SDRAM). In specific examples, the DRAM of a memory component may conform to standards known as JESD, issued by the Joint Electron Device Engineering Council (JEDEC), such as JESD79F for DDR SDRAM, JESD79-2F for DDR2 SDRAM, JESD79-3F for DDR3 SDRAM, JESD79-4A for DDR4 SDRAM, JESD209 for Low Power DDR (LPDDR), JESD209-2 for LPDDR2, JESD209-3 for LPDDR3, and JESD209-4 for LPDDR4. Such standards (and similar standards) may be referred to as DDR-based standards, and the communication interfaces of memory devices implementing such standards may be referred to as DDR-based interfaces.

[0071] In one example, the memory device is a block-addressable memory device, such as a memory device based on NOT AND (NAND) or NOT OR (NOR) technology. The memory device may also include next-generation non-volatile devices, such as Intel 3DXPoint. TMMemory or other byte-addressable write-in-place non-volatile memory devices. In one example, the memory device may be or may include a memory device using chalcogenide glass, multi-threshold level NAND flash memory, NOR flash memory, single-level or multi-level phase change memory (PCM), resistive memory, nanowire memory, ferroelectric transistor random access memory (FeTRAM), antiferroelectric memory, magnetoresistive random access memory (MRAM) incorporating memristor technology, resistive memory including metal oxide substrate, oxygen vacancy substrate and conductive bridge random access memory (CB-RAM), or spintransfer torque (STT)-MRAM, a device based on spintronic magnetic junction memory, a device based on magnetic tunneling junction (MTJ), a device based on domain wall (DW) and spin-orbit transfer (SOT), a thyristor-based memory device, or any combination of the above or other memories. A memory device may refer to the die itself and / or a packaged memory product. In some examples, a memory device may include a transistorless, stackable cross-point architecture in which memory cells are located at the intersection of word lines and bit lines and are individually addressable, and where bit storage is based on variations in body resistance.

[0072] Now for reference Figure 9 In some examples, skid 500 may be implemented as computation skid 900. Computation skid 900 is optimized or otherwise configured to perform computational tasks. As discussed above, computation skid 900 may rely on other skids (such as acceleration skids and / or storage skids) to perform such computational tasks. Computation skid 900 includes components similar to those already implemented... Figure 9 Various physical resources (e.g., electrical components) of the skid 500, identified using the same reference numerals. (The above refers to...) Figure 7 and Figure 8 The description of such components provided applies to the corresponding components of the computation skid 900, and will not be repeated in this document for the sake of clarity in the description of the computation skid 900.

[0073] In the illustrative computing skid 900, physical resources 720 include a programmable circuit system 920. Although in Figure 9Only two boxes of the programmable circuit system 920 are shown, but it should be understood that in other examples, the computing skid 900 may include an additional programmable circuit system 920. Illustratively, the programmable circuit system 920 corresponds to the high-performance processor circuit system 920 and can be configured to operate at relatively high power ratings. While the high-performance programmable circuit system 920 generates additional heat when operating at higher power ratings than a typical processor circuit system (which operates at approximately 155-230 W), the enhanced thermal cooling characteristics of the chassisless circuit board substrate 702 discussed above facilitate higher power operation. For example, in the illustrative example, the programmable circuit system 920 is configured to operate at a power rating of at least 250 W. In some examples, the programmable circuit system 920 may be configured to operate at a power rating of at least 350 W.

[0074] In some examples, the computing skid 900 may also include a programmable circuit system-to-programmable circuit system interconnect 942. Similar to the resource-to-resource interconnect 724 of the skid 500 discussed above, the programmable circuit system-to-programmable circuit system interconnect 942 can be implemented as any type of communication interconnect capable of facilitating communication between the programmable circuit system-to-programmable circuit system interconnect 942. In illustrative examples, the programmable circuit system-to-programmable circuit system interconnect 942 is implemented as a high-speed point-to-point interconnect (e.g., faster than I / O subsystem 722). For example, the programmable circuit system-to-programmable circuit system interconnect 942 can be implemented as a fast path interconnect (QPI), a super path interconnect (UPI), or other high-speed point-to-point interconnect dedicated to programmable circuit system-to-programmable circuit system communication.

[0075] The computing skid 900 also includes communication circuitry 930. The illustrative communication circuitry 930 includes a network interface controller (NIC) 932, which may also be referred to as a host fabric interface (HFI). The NIC 932 may be implemented as or otherwise include any type of integrated circuit, discrete circuitry, controller chip, chipset, plug-in board, daughter card, network interface card, or other device that the computing skid 900 may use to connect to another computing device (e.g., other skids 500). In some examples, the NIC 932 may be implemented as part of a system-on-a-chip (SoC) including one or more processor circuits, or the NIC 932 may be included in a multi-chip package that also includes one or more processor circuits. In some examples, the NIC 932 may include local processor circuitry (not shown) and / or local memory (not shown), both of which are local to the NIC 932. In such examples, the local processor circuitry of the NIC 932 may be able to perform one or more functions of the programmable circuit system 920. Additionally or alternatively, in such examples, the local memory of the NIC 932 may be integrated into one or more components of the computing skid at the board level, socket level, chip level, and / or other levels.

[0076] Communication circuitry 930 is communicatively coupled to optical data connector 934. Optical data connector 934 is configured to mate with a corresponding optical data connector of rack 340 when computing skid 900 is mounted in rack 340. Illustratively, optical data connector 934 includes a plurality of optical fibers extending from mating surfaces of optical data connector 934 to optical transceiver 936. Optical transceiver 936 is configured to convert incoming optical signals from rack-side optical data connectors into electrical signals, and to convert electrical signals into outgoing optical signals destined for rack-side optical data connectors. Although shown in the illustrative example as forming part of optical data connector 934, in other examples, optical transceiver 936 may form part of communication circuitry 930.

[0077] In some examples, the computing skid 900 may also include an expansion connector 940. In such examples, the expansion connector 940 is configured to mate with a corresponding connector on an extended chassisless circuit board substrate to provide additional physical resources to the computing skid 900. These additional physical resources may be used, for example, by a programmable circuit system 920 during operation of the computing skid 900. The extended chassisless circuit board substrate may be substantially similar to the chassisless circuit board substrate 702 discussed above and may include various electrical components mounted thereon. The specific electrical components mounted to the extended chassisless circuit board substrate may depend on the intended function of the extended chassisless circuit board substrate. For example, the extended chassisless circuit board substrate may provide additional computing resources, memory resources, and / or storage resources. Therefore, the additional physical resources for extending the chassis-less circuit board substrate may include, but are not limited to: processor circuitry, memory devices, storage devices and / or accelerator circuitry, including, for example, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), security coprocessor circuitry, graphics processing units (GPUs), machine learning circuitry, or other dedicated processor circuitry, controllers, devices and / or circuitry.

[0078] Now for reference Figure 10 An illustrative example of a computing skid 900 is shown. As illustrated, a programmable circuit system 920, communication circuitry 930, and an optical data connector 934 are mounted to the top side 750 of a chassis-less circuit board substrate 702. Any suitable attachment or mounting technique can be used to mount the physical resources of the computing skid 900 to the chassis-less circuit board substrate 702. For example, various physical resources can be mounted in corresponding sockets (e.g., processor circuit sockets), holders, or brackets. In some cases, some electrical components can be directly mounted to the chassis-less circuit board substrate 702 via soldering or similar techniques.

[0079] As discussed above, the separate programmable circuit system 920 and communication circuit 930 are mounted to the top side 750 of the chassisless circuit board substrate 702 such that no two heat-generating electrical components are shaded from each other. In the illustrative example, the programmable circuit system 920 and communication circuit 930 are mounted in corresponding positions on the top side 750 of the chassisless circuit board substrate 702 such that no two physical resources are linearly aligned with other physical resources along the direction of the airflow path 708. It should be understood that although the optical data connector 934 is aligned with the communication circuit 930, the optical data connector 934 does not generate heat or generate nominal heat during operation.

[0080] As discussed above regarding skid 500, the memory device 820 of computing skid 900 is mounted to the bottom side 850 of the chassisless circuit board substrate 702. Although mounted to the bottom side 850, the memory device 820 is communicatively coupled to the programmable circuit system 920 located on the top side 750 via I / O subsystem 722. Because the chassisless circuit board substrate 702 is implemented as a dual-sided board, the memory device 820 and the programmable circuit system 920 can be communicatively coupled via one or more vias, connectors, or other mechanisms extending through the chassisless circuit board substrate 702. In some examples, different programmable circuit systems 920 (e.g., different processor circuit systems) may be communicatively coupled to one or more memory devices 820 from different sets. Alternatively, in other examples, different programmable circuit systems 920 (e.g., different processor circuit systems) may be communicatively coupled to the same memory device within the memory device 820. In some examples, the memory device 820 may be mounted to one or more memory interlayers on the bottom side of the chassisless circuit board substrate 702 and may be interconnected with the corresponding programmable circuit system 920 via a ball grid array.

[0081] Different programmable circuit systems 920 (e.g., different processor circuit systems) include corresponding heat sinks 950 fixed thereon, and / or are associated with corresponding heat sinks 950 fixed thereon. Due to the mounting of the memory device 820 to the bottom side 850 of the chassisless circuit board substrate 702 (and the vertical spacing of the skids 500 in the corresponding rack 340), the top side 750 of the chassisless circuit board substrate 702 includes additional “free” areas or spaces, which facilitate the use of heat sinks 950 that are larger in size than conventional heat sinks used in typical servers. Additionally, due to the improved thermal cooling characteristics of the chassisless circuit board substrate 702, no programmable circuit system heat sink 950 includes a cooling fan attached thereto. That is, the heat sink 950 can be a fanless heat sink. In some examples, the heat sink 950 mounted on top of the programmable circuit system 920 may overlap with the heat sink attached to the communication circuitry 930 in the direction of the airflow path 708 due to its increased size, as by Figure 10 To illustrate.

[0082] Now for reference Figure 11 In some examples, skid 500 may be implemented as accelerator skid 1100. Accelerator skid 1100 is configured to perform specialized computational tasks, such as machine learning, encryption, hashing, or other computationally intensive tasks. In some examples, for instance, computational skid 900 may migrate tasks to accelerator skid 1100 during operation. Accelerator skid 1100 includes features similar to those already implemented in [the previous example]. Figure 11 Various components of the skid 500 and / or the calculated skid 900, identified using the same reference numerals. (The above refers to...) Figure 7 , Figure 8 and Figure 9 The description of such components provided applies to the corresponding components of the accelerator skid 1100, and will not be repeated herein for the sake of clarity in the description of the accelerator skid 1100.

[0083] In the illustrative accelerator skid 1100, physical resource 720 includes accelerator circuitry 1120. Although in Figure 11 Only two accelerator circuits 1120 are shown, but it should be understood that in other examples, the accelerator skid 1100 may include additional accelerator circuits 1120. For example, as Figure 12 As shown, the accelerator skid 1100 may include four accelerator circuits 1120. The accelerator circuits 1120 may be implemented as any type of processor circuit system, coprocessor circuit system, computing circuit, or other device capable of performing computation or processing operations. For example, the accelerator circuits 1120 may be implemented as, for example, an FPGA, ASIC, secure coprocessor circuit system, neuromorphic processor unit, quantum computer, machine learning circuit, or other dedicated processor circuit system, controller, device, and / or circuit.

[0084] In some examples, accelerator skid 1100 may also include an accelerator-to-accelerator interconnect 1142. Similar to the resource-to-resource interconnect 724 of skid 500 discussed above, the accelerator-to-accelerator interconnect 1142 can be implemented as any type of communication interconnect capable of facilitating accelerator-to-accelerator communication. In illustrative examples, the accelerator-to-accelerator interconnect 1142 is implemented as a high-speed point-to-point interconnect (e.g., faster than I / O subsystem 722). For example, the accelerator-to-accelerator interconnect 1142 can be implemented as a Fast Path Interconnect (QPI), a Hyper Path Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to programmable circuit system-to-programmable circuit system communication. In some examples, accelerator circuitry 1120 may be daisy-chained, with primary accelerator circuitry 1120 connected to NIC 932 and memory 820 via I / O subsystem 722, and secondary accelerator circuitry 1120 connected to NIC 932 and memory 820 via primary accelerator circuitry 1120.

[0085] Now for reference Figure 12An illustrative example of an accelerator skid 1100 is shown. As discussed above, accelerator circuitry 1120, communication circuitry 930, and optical data connector 934 are mounted to the top side 750 of the chassisless circuit board substrate 702. Similarly, as discussed above, the respective accelerator circuitry 1120 and communication circuitry 930 are mounted to the top side 750 of the chassisless circuit board substrate 702 such that no two heat-generating electrical components are shaded from each other. A memory device 820 of the accelerator skid 1100 is mounted to the bottom side 850 of the chassisless circuit board substrate 702, as discussed above with respect to skid 500. Although mounted to the bottom side 850, the memory device 820 is communicatively coupled to the accelerator circuitry 1120 located on the top side 750 via an I / O subsystem 722 (e.g., through vias). Furthermore, the accelerator circuit 1120 may include, and / or be associated with, a heatsink 1150 that is larger than a conventional heatsink used in servers. (See above reference.) Figure 9 The heat sink 950 discussed here, heat sink 1150 can be larger than a conventional heat sink because the “free” area provided by memory resources 820 is located on the bottom side 850 of the chassisless circuit board substrate 702 rather than on the top side 750.

[0086] Now for reference Figure 13 In some examples, skid 500 may be implemented as storage skid 1300. Storage skid 1300 is configured to store data in a local data storage device 1350. For example, during operation, computing skid 900 or accelerator skid 1100 may store data to and retrieve data from the data storage device 1350 of storage skid 1300. Storage skid 1300 includes features similar to those already present in... Figure 13 Various components of the skid 500 and / or the calculated skid 900, identified using the same reference numerals. (The above refers to...) Figure 7 , Figure 8 and Figure 9 The description of such components provided applies to the corresponding components of storage skid 1300, and will not be repeated here for the sake of clarity in the description of storage skid 1300.

[0087] In the illustrative storage skid 1300, physical resource 720 includes storage controller 1320. Although in Figure 13Only two storage controllers 1320 are shown, but it should be understood that in other examples, the storage skid 1300 may include additional storage controllers 1320. The storage controller 1320 can be implemented as any type of programmable circuit system, controller, or control circuit capable of controlling the storage and retrieval of data in the data storage device 1350 based on requests received via communication circuitry 930. In the illustrative example, the storage controller 1320 is implemented as a relatively low-power programmable circuit system or controller. For example, in some examples, the storage controller 1320 may be configured to operate at a rated power of approximately 75 watts.

[0088] In some examples, the storage skid 1300 may also include a controller-to-controller interconnect 1342. Similar to the resource-to-resource interconnect 724 of the skid 500 discussed above, the controller-to-controller interconnect 1342 can be implemented as any type of communication interconnect capable of facilitating controller-to-controller communication. In illustrative examples, the controller-to-controller interconnect 1342 is implemented as a high-speed point-to-point interconnect (e.g., faster than the I / O subsystem 722). For example, the controller-to-controller interconnect 1342 can be implemented as a Fast Path Interconnect (QPI), a Hyper Path Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to programmable circuit system-to-programmable circuit system communication.

[0089] Now for reference Figure 14 An illustrative example of a storage skid 1300 is shown. In this illustrative example, the data storage device 1350 is implemented as, or otherwise includes, a storage cage 1352 configured to accommodate one or more solid-state drives (SSDs) 1354. For this purpose, the storage cage 1352 includes a plurality of mounting slots 1356 configured to receive corresponding SSDs 1354. The mounting slots 1356 include a plurality of drive guides 1358 that cooperate to define an access opening for the corresponding mounting slot 1356. The storage cage 1352 is secured to a chassis-less circuit board substrate 702 such that the access opening faces away from the chassis-less circuit board substrate 702 (e.g., towards the front of the chassis-less circuit board substrate 702). Thus, the solid-state drives 1354 are accessible when the storage skid 1300 is mounted in a corresponding rack 340. For example, the solid-state drive 1354 can be swapped out of the rack 340 (e.g., via a robot), while the storage skid 1300 remains mounted in the corresponding rack 340.

[0090] Storage cage 1352 illustratively includes sixteen mounting slots 1356 and is capable of mounting and storing sixteen solid-state drives 1354. In other examples, storage cage 1352 may be configured to store additional or fewer solid-state drives 1354. Additionally, in the illustrative example, the solid-state drives are mounted vertically in storage cage 1352, but in other examples, the solid-state drives may be mounted in storage cage 1352 with different orientations. A given solid-state drive 1354 can be implemented as any type of data storage device capable of storing long-term data. For this purpose, solid-state drive 1354 may include the volatile memory devices and non-volatile memory devices discussed above.

[0091] like Figure 14 As shown, the storage controller 1320, communication circuitry 930, and optical data connector 934 are illustratively mounted to the top side 750 of the chassisless circuit board substrate 702. Similarly, as discussed above, any suitable attachment or mounting techniques can be used to mount the electrical components of the storage skid 1300 to the chassisless circuit board substrate 702, including, for example, sockets (e.g., processor circuit sockets), retainers, brackets, solder joints, and / or other mounting or securing techniques.

[0092] As discussed above, the various memory controllers 1320 and the communication circuitry 930 are mounted on the top side 750 of the chassisless circuit board substrate 702 such that no two heat-generating electrical components are shaded from each other. For example, the memory controllers 1320 and the communication circuitry 930 are mounted in corresponding positions on the top side 750 of the chassisless circuit board substrate 702 such that no two electrical components are linearly aligned with each other along the direction of the airflow path 708.

[0093] As discussed above regarding skid 500, the memory device 820 of storage skid 1300 (not in...) Figure 14 (As shown in the image) is mounted to the bottom side 850 of the chassisless circuit board substrate 702 (not in the image). Figure 14(As shown in the diagram). Although mounted to the bottom side 850, the memory device 820 is communicatively coupled to the memory controller 1320 located on the top side 750 via the I / O subsystem 722. Similarly, because the chassisless circuit board substrate 702 is implemented as a dual-sided circuit board, the memory device 820 and the memory controller 1320 can be communicatively coupled via one or more vias, connectors, or other mechanisms extending through the chassisless circuit board substrate 702. The memory controller 1320 includes a heatsink 1370 attached thereto, and / or is associated with the heatsink 1370 attached thereto. As discussed above, due to the improved thermal cooling characteristics of the chassisless circuit board substrate 702 of the memory skid 1300, the heatsink 1370 does not include a cooling fan attached thereto. That is, the heatsink 1370 can be a fanless heatsink.

[0094] Now for reference Figure 15 In some examples, skid 500 may be implemented as memory skid 1500. Memory skid 1500 is optimized or otherwise configured to provide other skids 500 (e.g., compute skid 900, accelerator skid 1100, etc.) with access to a memory pool local to memory skid 1300 (e.g., in two or more sets 1530, 1532 of memory device 820). For example, during operation, compute skid 900 or accelerator skid 1100 may use a logical address space mapped to physical addresses in memory sets 1530, 1532 to remotely write to one or more of memory sets 1530, 1532 of memory skid 1300 and / or read from one or more of memory sets 1530, 1532 of memory skid 1300. Memory skid 1500 includes features similar to those already implemented in memory devices 820. Figure 15 Various components of the skid 500 and / or the calculated skid 900, identified using the same reference numerals. (The above refers to...) Figure 7 , Figure 8 and Figure 9 The description of such components provided applies to the corresponding components of memory skid 1500, and will not be repeated herein for the sake of clarity in the description of memory skid 1500.

[0095] In the illustrative memory skid 1500, physical resources 720 include a memory controller 1520. Although in Figure 15Only two memory controllers 1520 are shown, but it should be understood that in other examples, memory skid 1500 may include additional memory controllers 1520. Memory controllers 1520 may be implemented as any type of programmable circuitry, controller, or control circuit capable of controlling the writing and reading of data to and from memory sets 1530, 1532 based on requests received via communication circuitry 930. In the illustrative example, memory controllers 1520 are connected to the respective memory sets 1530, 1532 to write to and read from memory devices 820 (not shown) within the respective memory sets 1530, 1532, and to implement any permissions (e.g., read, write, etc.) associated with skid 500 that has sent a request to memory skid 1500 to perform memory access operations (e.g., read or write).

[0096] In some examples, memory skid 1500 may also include a controller-to-controller interconnect 1542. Similar to the resource-to-resource interconnect 724 of skid 500 discussed above, the controller-to-controller interconnect 1542 may be implemented as any type of communication interconnect capable of facilitating controller-to-controller communication. In illustrative examples, the controller-to-controller interconnect 1542 is implemented as a high-speed point-to-point interconnect (e.g., faster than I / O subsystem 722). For example, the controller-to-controller interconnect 1542 may be implemented as a Fast Path Interconnect (QPI), a Hyper Path Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to programmable circuit system-to-programmable circuit system communication. Thus, in some examples, memory controller 1520 can access memory within a memory set 1532 associated with another memory controller 1520 via the controller-to-controller interconnect 1542. In some examples, the scalable memory controller consists of multiple smaller memory controllers (referred to herein as “chiplets”) on a memory skid (e.g., memory skid 1500). Chipsets can be interconnected (e.g., using Embedded Multi-Die Interconnect Bridge (EMIB) technology). Combined chiplet memory controllers can scale to a relatively large number of memory controllers and I / O ports (e.g., up to 16 memory channels). In some examples, memory controller 1520 can implement memory interleaving (e.g., one memory address is mapped to memory set 1530, the next memory address is mapped to memory set 1532, and a third address is mapped to memory set 1530, etc.). Interleaving can be managed within memory controller 1520, or across network links from (e.g., from compute skid 900) CPU sockets to memory sets 1530, 1532, and can improve latency associated with performing memory access operations compared to accessing consecutive memory addresses from the same memory device.

[0097] Furthermore, in some examples, the memory skid 1500 can be connected via a waveguide connector 1580 to one or more other skids 500 (e.g., in the same rack 340 or adjacent racks 340). In an illustrative example, the waveguide is a 74 mm waveguide providing 16 Rx (e.g., receive) channels and 16 Tx (e.g., transmit) channels. In an illustrative example, the different channels are 16 gigahertz (GHz) or 32 GHz. In other examples, the frequencies may be different. Using a waveguide provides high-throughput access to a memory pool (e.g., memory sets 1530, 1532) to another skid (e.g., a skid 500 in the same rack 340 or adjacent rack 340 as the memory skid 1500) without adding load to the optical data connector 934.

[0098] Now for reference Figure 16A system for performing one or more workloads (e.g., applications) may be implemented according to data center 200. In an illustrative example, system 1610 includes orchestrator server 1620, which may be implemented as a managed node including computing devices (e.g., programmable circuit system 920 on compute skid 900) that perform management software (e.g., cloud operating environments such as OpenStack), the orchestrator server 1620 being communicatively coupled to a plurality of skids 500, including a large number of compute skids 1630 (e.g., similar to compute skid 900), memory skids 1640 (e.g., similar to memory skid 1500), accelerator skids 1650 (e.g., similar to accelerator skid 1100), and storage skids 1660 (e.g., similar to storage skid 1300). One or more of skids 1630, 1640, 1650, and 1660 may be grouped by orchestrator server 1620 into managed nodes 1670 to collectively perform workloads (e.g., application 1632 running in a virtual machine or container). Managed node 1670 may be implemented as a component of physical resources 720 (such as programmable circuitry system 920, memory resources 820, accelerator circuitry 1120, or data storage device 1350) from the same or different skids 500. Furthermore, managed nodes may be created, defined, or "spin-up" by orchestrator server 1620 when a workload is to be assigned to that managed node or at any other time, and managed nodes may exist regardless of whether any workload is currently assigned to them. In an illustrative example, orchestrator server 1620 may selectively allocate and / or deallocate physical resources 720 from skid 500 and / or add one or more skids 500 to or remove one or more skids 500 from managed node 1670 based on Quality of Service (QoS) objectives (e.g., target throughput, target latency, target instructions per second, etc.) associated with a Service Level Protocol (SLP) for a workload (e.g., application 1632). In doing so, orchestrator server 1620 may receive telemetry data indicating the performance status (e.g., throughput, latency, instructions per second, etc.) of different skids among the skids 500 of managed node 1670 and compare this telemetry data with the QoS objectives to determine whether the QoS objectives are met. Orchestrator server 1620 may additionally determine whether one or more physical resources can be deallocated from managed node 1670 while still satisfying the QoS objectives, thereby freeing up those physical resources for use on another managed node (e.g., to perform a different workload). Alternatively, if the QoS objective is not currently met, the orchestrator server 1620 may determine, as the workload is executed, to dynamically allocate additional physical resources to assist the execution of the workload (e.g., application 1632).Similarly, if the orchestrator server 1620 determines that releasing physical resources would result in the QoS objective still being met, the orchestrator server 1620 may determine to dynamically release physical resources from the managed nodes.

[0099] Additionally, in some examples, orchestrator server 1620 may identify trends in the resource utilization of a workload (e.g., application 1632) by identifying the phases in which different operations with different resource utilization characteristics are executed (e.g., time periods in which different operations with different resource utilization characteristics are executed) and preemptively identifying available resources in data center 200 and allocating them to managed nodes 1670 (e.g., within a predefined time period at the start of the associated phase). In some examples, orchestrator server 1620 may model performance based on various latency and allocation schemes to place workloads between compute skids and other resources (e.g., accelerator skids, memory skids, storage skids) in data center 200. For example, orchestrator server 1620 may utilize a model that takes into account the performance of resources on skid 500 (e.g., FPGA performance, memory access latency, etc.) and the performance of paths over the network to resources (e.g., FPGA) (e.g., congestion, latency, bandwidth). Thus, the orchestrator server 1620 can determine which resource(s) should be used with which workloads based on the total latency associated with one or more different potential resources available in the data center 200 (e.g., latency associated with the performance of the resource itself, in addition to latency associated with the path of the network between the computing skid performing the workload and the skid 500 on which the resource is located).

[0100] In some examples, orchestrator server 1620 may use telemetry data (e.g., temperature, fan speed, etc.) reported from skid 500 to generate a heat generation map of data center 200 and allocate resources to managed nodes based on this heat generation map and predicted heat generation associated with different workloads to maintain target temperature and heat distribution in data center 200. Additionally or alternatively, in some examples, orchestrator server 1620 may organize the received telemetry data into a hierarchical model indicating relationships between managed nodes (e.g., spatial relationships such as the physical location of resources of managed nodes within data center 200, and / or functional relationships such as grouping of managed nodes by clients serving them, the types of functions typically performed by managed nodes, managed nodes typically sharing or exchanging workloads with each other, etc.). Based on differences in physical location and resources among managed nodes, a given workload may exhibit different resource utilization across the resources of different managed nodes (e.g., causing different internal temperatures, using different percentages of programmable circuitry or memory capacity). The orchestrator server 1620 can determine these discrepancies based on telemetry data stored in the hierarchical model and incorporate these discrepancies into predictions of future resource utilization for a workload when it is reassigned from one managed node to another, in order to accurately balance resource utilization across the data center 200. In some examples, the orchestrator server 1620 can identify patterns in the resource utilization phases of a workload and use these patterns to predict future resource utilization for the workload.

[0101] To reduce the computational load on orchestrator server 1620 and the data transmission load on the network, in some examples, orchestrator server 1620 may send self-test information to skid 500 so that a given skid 500 can locally (e.g., on the skid 500) determine whether the telemetry data generated by the skid 500 meets one or more conditions (e.g., available capacity meeting a predefined threshold, temperature meeting a predefined threshold, etc.). The given skid 500 can then report a simplified result (e.g., yes or no) back to orchestrator server 1620, which can utilize this simplified result when determining resource allocation to managed nodes.

[0102] As described above, OCP was established to develop more energy-efficient data center computing technologies, thereby improving the efficiency of data center operations. Artificial intelligence (AI) is an evolving field driven by increasing demands on data centers. To meet these demands, many developers have created accelerators for machine learning, deep learning, and high-performance computing. Example accelerators include GPUs, FPGAs, ASICs, infrastructure processing units (IPUs), neural processing units (NPUs), XPUs, and so on. For example, an XPU can be implemented using a heterogeneous computing system that includes various types of programmable circuit systems (e.g., one or more FPGAs, one or more central processing units (CPUs), one or more GPUs, one or more NPUs, one or more digital signal processors (DSPs), etc., and / or any combination thereof) and orchestration technologies (e.g., one or more application programming interfaces (APIs)) that can assign computational tasks to any of the various types of programmable circuit systems suitable for and usable for performing the computational tasks.

[0103] To integrate accelerators into servers, they are typically implemented as expansion cards. For example, an expansion card is a printed circuit board (PCB) that can be inserted into an electrical connector on the motherboard of a computing system or into an expansion slot. However, expansion cards are implemented in various shapes, with a range of thermal characteristics, wiring diagrams, and unique sockets. While some standardized form factors exist for expansion cards, these are not suitable for AI workloads. For instance, expansion cards designed according to the PCIe card electromechanical (CEM) specification do not have sufficient bandwidth and interconnect flexibility for the data requirements of AI workloads. Therefore, members of the OCP developed a universal form factor according to which different types of accelerators (e.g., GPUs, FPGAs, ASICs, IPUs, NPUs, XPUs, etc.) can be implemented as expansion cards. Accelerators implemented according to this universal OCP form factor are called OCP Accelerator Modules (OAMs).

[0104] Figure 17A The figure shows an isometric view of the example OCP Accelerator Module (OAM) 1700. Figure 17B yes Figure 17A An example of an exploded view of the OAM1700. Figure 17A and Figure 17B In this document, they are collectively referred to as Figure 17. In the example of Figure 17, the OAM 1700 includes an example heat sink 1702, a first example mounting plate 1704, an example printed circuit board (PCB) 1706, and a second example mounting plate 1708. In the example of Figure 17, the PCB 1706 is housed between the first mounting plate 1704 and the second mounting plate 1708. For example, the first mounting plate 1704, the PCB 1706, and the second mounting plate 1708 are coupled together by a first example fastener 1710. Furthermore, for example, the heat sink 1702 is coupled to the first mounting plate 1704 by a second example fastener 1712.

[0105] In the illustrated example of Figure 17, the first fastener 1710 and the second fastener 1712 are implemented by screws (e.g., threaded fasteners). In the example of Figure 17, the washer can be implemented by one or more of the following: at least one of the first fasteners 1710 or at least one of the second fasteners 1712. In additional or alternative examples, at least one of the first fasteners 1710 is implemented by bolts or any other type of fastener. Additionally or alternatively, at least one of the second fasteners 1712 is implemented by bolts or any other type of fastener. In some examples, one or more of the first mounting plate 1704, PCB 1706, or second mounting plate 1708 are coupled together by crimping, welding, brazing, soldering, gluing, bonding, adhesive, or by another adhesive.

[0106] In the example illustrated in Figure 17, when the OAM 1700 is assembled, the first mounting plate 1704, PCB 1706, and second mounting plate 1708 are coupled together and referred to as example OAM card 1714. For example, the first mounting plate 1704, PCB 1706, and second mounting plate 1708 are coupled together by a first fastener 1710, which holds the PCB 1706 between the first mounting plate 1704 and the second mounting plate 1708. In the example of Figure 17, the first fastener 1710 extends through one or more of the first mounting plate 1704, PCB 1706, or second mounting plate 1708.

[0107] In the example illustrated in Figure 17, some extensions of the first fastener 1710 pass through mounting holes in the first mounting plate 1704, PCB 1706, and second mounting plate 1708, and are received by corresponding threaded mounting holes in the chassis or substrate. Furthermore, in the example of Figure 17, some extensions of the first fastener 1710 pass through mounting holes in the first mounting plate 1704 and PCB 1706, and are received by corresponding threaded mounting holes in the second mounting plate 1708.

[0108] In the example illustrated in Figure 17, the first mounting plate 1704 and the second mounting plate 1708 are load-bearing components. For example, the compressive load applied by the first fastener 1710 is transferred to the first mounting plate 1704 and the second mounting plate 1708, as well as around the PCB 1706. Furthermore, when the OAM 1700 is assembled, the first mounting plate 1704 and the heatsink 1702 are coupled together via the second fastener 1712. In the example of Figure 17, the second fastener 1712 holds the heatsink 1702 to the first mounting plate 1704. In the example of Figure 17, the second fastener 1712 extends through a mounting hole in the heatsink 1702 and is received by a corresponding threaded mounting hole in the first mounting plate 1704.

[0109] In hardware design, particularly in the context of OAM cards, the utilization of physical space (e.g., board space) is a critical consideration. This consideration of physical space is emphasized by the spatial constraints of OAM cards. For example, in an OAM card, some electrical components are located near the proximal end of PCB mounting holes (e.g., openings, vias, etc.). This close proximity of electrical components to PCB mounting holes presents challenges, potentially leading to (e.g., generating) detrimental consequences such as mechanical stress concentration, which can cause failures associated with surface mount technology (SMT) components.

[0110] Figure 18A The figure shows a cross-sectional view of a first example mounting plate 1802, an example printed circuit board (PCB) 1804, and a second example mounting plate 1806 (collectively referred to as example OAM card 1808) mounted to an example system chassis 1810. Figure 18A In the example, PCB 1804 includes example mounting holes 1812. (As...) Figure 18A As illustrated, mounting hole 1812 can receive example fastener 1814, which can be used to attach OAM card 1808 to system chassis 1810 (e.g., substrate, reinforcement, etc.) via example motherboard 1816 (e.g., Unified Base Board, UBB)). Figure 18A The figure shows fastener 1814 when it is not installed on system chassis 1810. Figure 18B The figure shows a cross-sectional view of an example OAM card 1808 when fastener 1814 is mounted to system chassis 1810. The involvement of system chassis 1810 increases the risk of failure of PCB 1804 and / or electrical components mounted to PCB 1804 (e.g., SMT components) in the area near mounting holes 1812, especially under shock and vibration loads.

[0111] As described above, OAM cards are inherently space-constrained, which often results in components being placed very close to PCB mounting holes (e.g., mounting hole 1812). When the mount is positioned too close to the SMT component (e.g., within a threshold distance), the risk of short circuits or interference increases. Additionally, the rigid attachment of the PCB to the mounting holes can introduce localized stress concentrations around the holes (e.g., during shock, vibration, and thermal cycling). In some examples, even microvias in the top layer of the PCB can cause mechanical failure. Furthermore, SMT components are particularly susceptible to damage due to localized stress because of the relatively weak solder joints (e.g., compared to other mounting techniques).

[0112] Referring back to Figure 17, as described above, the OAM card 1714 includes a first mounting plate 1704 (e.g., a top plate) and a second mounting plate 1708 (e.g., a back plate) that clamps a PCB 1706 (e.g., positioned above and below the PCB 1706, respectively). In this way, the first mounting plate 1704 and / or the second mounting plate 1708 significantly reduce the overall bending strain on the PCB 1706. Additionally, the OAM card 1714 advantageously includes a first example support 1716 and a second example support 1718 between the first mounting plate 1704 and the PCB 1706. For example, Figure 19 The figure shows a side view of an example OAM card 1714 of Figure 17, which includes a first support 1716 and a second support 1718 when assembled.

[0113] exist Figure 19 In the illustrated example, the first support 1716 is positioned over an opening (e.g., a mounting hole) in the PCB 1706, which poses a high risk of localized stress. For example, the PCB 1706 includes an opening and at least a threshold number of components (e.g., a first example component 1902) surrounding that opening. Additionally or alternatively, the PCB 1706 includes an opening and one or more components (e.g., the first component 1902) within an example threshold distance 1904 of that opening. Figure 19In the example, the threshold distance 1904 corresponds to a distance from the support that would cause the component (and / or the solder associated with the component) to be damaged by stresses imposed on the component (and / or the solder) due to loading on the support. Additionally or alternatively, the threshold distance 1904 corresponds to the difference between a first radius of the outermost periphery of the support and a second radius of the periphery of the opening on which the support is positioned. Thus, if the opening is within the threshold distance 1904 of the component (and / or the solder associated with the component), the component (and / or the solder) may be damaged by stresses imposed on the component (and / or the solder) due to loading on the support positioned above the opening. Advantageously, in the example disclosed herein, the first support 1716 is designed to have at least one of a compliant structure (e.g., a spring, a resiliently compressible material, etc.) or a different cross-sectional shape at opposite ends (e.g., a first cross-sectional shape adjacent to the first mounting plate 1704 and a second, different cross-sectional shape adjacent to the PCB 1706). As a result, the first support 1716 reduces the failure of at least one of the PCB 1706 or the components (e.g., the first component 1902) within the threshold distance 1904 of the first support 1716 due to localized stress concentration.

[0114] In the examples disclosed herein, the first support 1716 includes an adjustable support having a compliant structure (e.g., a spring, a resilient and compliant material). Figure 20 The figure shows a cross-sectional view of a first example implementation of the first support 1716 in Figure 17. Figures 21A-21C The diagram shows Figure 20 Additional details of the first support 1716. Figure 22 The figure shows a cross-sectional view of a second example implementation of the first support 1716 of Figure 17. In some examples, the first support 1716 includes a gap cut, such that the first support 1716 includes different cross-sectional shapes at opposite ends of the first support 1716. For example, the gap cut ensures that a component mounted to the surface of PCB 1706 is at least a threshold distance from the inner periphery of the first support 1716, even if the component is within a threshold distance from the outer periphery of the first support 1716. Figure 23 The figure shows a third example implementation of the first support 1716 in Figure 17. Figures 24A-24C The diagram shows Figure 23 Additional details of the first support 1716. Figure 25- Figure 29 The figure illustrates an additional or alternative implementation of the first support 1716 in Figure 17.

[0115] exist Figure 19In the illustrated example, the second support 1718 is positioned above an opening (e.g., a mounting hole) in PCB 1706, which introduces a high risk of localized stress. For example, PCB 1706 includes an opening and components surrounding that opening at a less than a threshold amount. Alternatively, PCB 1706 includes an opening and components (e.g., a second example component 1906) at a distance greater than a threshold distance 1904 from that opening. In some examples, the second support 1718 is positioned above an area of ​​PCB 1706 that does not include mounting holes. In such examples, the second support 1718 may not include a channel passing through it (e.g., the second support 1718 does not serve as a channel for fasteners to mount the OAM card 1714 to a reinforcement or substrate).

[0116] In the examples disclosed herein, the second support 1718 is a fixed-length support (e.g., substantially rigid, without compliant structure). Additionally or alternatively, the second support 1718 has a uniform cross-sectional shape along its length. For example, the second support 1718 has a uniform cylindrical, rectangular, or hexagonal shape. Furthermore, for example, the second support 1718 is a solid material with a uniform shape and is used to separate the first mounting plate 1704 from the PCB 1706. In the examples disclosed herein, the second support 1718 may be made of stainless steel, aluminum, brass, and / or nylon.

[0117] In some examples, the second support 1718 is coated with an electrically insulating material. In some examples, the second support 1718 includes threads at one end to secure the second support 1718 to the PCB 1706. Additionally or alternatively, in some examples, the second support 1718 is threadedly coupled to the first mounting plate 1704. In other examples, the second support 1718 is an integral extension of the first mounting plate 1704. As described above, in some examples, the first support 1716 satisfies at least one of the following: having a different shape than the second support 1718; or comprising a different material than the second support 1718. In the examples disclosed herein, by utilizing supports such as the first support 1716 over areas of the PCB 1706 at high risk of causing localized stress, and (2) utilizing supports such as the second support 1718 over areas of the PCB 1706 at low risk of causing localized stress, the examples disclosed herein reduce stress concentration in the OAM card 1714.

[0118] Figure 20 The diagram illustrates a first example implementation of the first support 1716 of Figure 17 when it is assembled into the example OAM card 1714. Figure 20In the example, the first example fastener 1710A of the first fastener 1710 extends through the first example mounting hole 2002 of the first mounting plate 1704, the first support 1716, the second example mounting hole 2004 of the PCB 1706, and the third example mounting hole 2006 of the second mounting plate 1708. Thus, the first support 1716 surrounds (e.g., to surround) the first fastener 1710A. Furthermore, in Figure 20 In the example, the first support 1716 includes a first example segment 2008, a second example segment 2010, an example compliant structure 2012, and one or more example pins 2014.

[0119] exist Figure 20 In the illustrated example, the first segment 2008 is adjacent to a first example surface 2016 of the first mounting plate 1704 and extends from the first example surface 2016 of the first mounting plate 1704 toward a second example surface 2018 of the second mounting plate 1708. For example, the first segment 2008 (e.g., a first portion) is an integral extension of the first mounting plate 1704 (e.g., a metal sheet). In some examples, the first segment 2008 is different from and coupled to (e.g., threaded to, adhered to, welded to, etc.) the first mounting plate 1704. Figure 20 In some examples, the first segment 2008 is made of metals such as stainless steel, aluminum, and / or brass. In some examples, the first segment 2008 is coated with an electrically insulating material. Figure 20 In the example, the first segment 2008 is cylindrical in shape (e.g., has a shape defined by a cylinder) and has a first example cavity 2020 that defines the inner surface of the first segment 2008.

[0120] exist Figure 20 In the illustrated example, the compliant structure 2012 is housed within the first cavity 2020 of the first segment 2008. Figure 20 In one example, the compliant structure 2012 is compressively held between the first surface 2016 and the first example end 2022 of the second segment 2010. For example, the compliant structure 2012 is implemented by a spring, which is compressively held between the first surface 2016 and the first end 2022. In additional or alternative examples, the compliant structure 2012 is implemented by at least one of the following: rubber, silicone, foam, or shape-memory alloy (SMA).

[0121] exist Figure 20In the illustrated example, the compliant structure 2012 is a biasing component that biases the second segment 2010 relative to the first mounting plate 1704. For example, the compliant structure 2012 applies a force to at least one of the second segment 2010 or the first mounting plate 1704 to hold the second segment 2010 in a first position relative to the first mounting plate 1704. Additionally or alternatively, the compliant structure 2012 applies a force to at least one of the second segment 2010 or the first mounting plate 1704 to move the second segment 2010 relative to the first mounting plate 1704 (e.g., induced movement of the second segment 2010 relative to the first mounting plate 1704).

[0122] exist Figure 20 In the illustrated example, the second segment 2010 extends from the first segment 2008 toward the second surface 2018 of the PCB 1706. For example, the second segment 2010 (e.g., the second portion) extends from the first segment 2008 and rests on the second surface 2018 of the PCB 1706. Figure 20 In some examples, the second segment 2010 is made of metals such as stainless steel, aluminum, and / or brass. In some examples, the second segment 2010 is coated with an electrically insulating material. Figure 20 In one example, the second segment 2010 is cylindrical in shape (e.g., has a shape defined by a cylinder) and has a second example cavity 2024 defining the inner surface of the second segment 2010. In other examples, one or both of the first segment 2008 and the second segment 2010 have different cross-sectional shapes.

[0123] exist Figure 20 In the illustrated example, the outer diameter of the second segment 2010 is smaller than the inner diameter of the first segment 2008. Therefore, the second segment 2010 can be inserted into the first cavity 2020 of the first segment 2008. Furthermore, the inner diameter of the second segment is large enough to allow the first fastener 1710A to pass through the second segment 2010. Figure 20 In the example, the second segment 2010 also includes an example opening 2026 (e.g., a slot) to receive a corresponding pin in one or more pins 2014.

[0124] exist Figure 20In the illustrated example, each of one or more pins 2014 extends from the inner surface of the first segment 2008 into the first cavity 2020. For example, each of one or more pins 2014 is embedded in the first segment 2008, extends from the outer surface of the first segment 2008 through the first segment 2008, and enters the first cavity 2020. Thus, the second segment 2010 can be moved relative to the first mounting plate 1704 (e.g., movable relative to the first mounting plate 1704). For example, the movement of the second segment 2010 is guided by the opening 2026 to slide along one or more pins 2014. Thus, the length of the first support 1716 is adjustable (e.g., has an adjustable length), and the spring load (e.g., provided by the compliant structure 2012) allows the first support 1716 to absorb stress.

[0125] For example, when the OAM card 1714 is subjected to loading (e.g., shock loading, vibration loading, thermal cycling, etc.), the second segment 2010 can move as the compliant structure 2012 compresses under load. In this way, even if the first component 1902 is within the threshold distance 1904 of the first support 1716, the first support 1716 prevents the generation of localized stress in the first component 1902, its solder joints, or the PCB 1706. Thus, the first support 1716 reduces the failure of at least one of the PCB 1706 or the first component 1902 due to localized stress concentration in the area surrounding the first support 1716.

[0126] Figure 21A yes Figure 20 Isometric view of the cross-section of the first support 1716. Figure 21B yes Figure 20 An isometric view of the first support 1716, and Figure 21C yes Figure 20 A cross-sectional side view of the first support 1716. As described above, the second segment 2010 includes an opening 2026 to receive a corresponding pin in one or more pins 2014. For example, as... Figures 21A-21C As illustrated, the second segment 2010 includes a first example opening 2026A for receiving a first example pin 2014A. The second segment 2010 also includes a second example opening 2026B for receiving a second example pin 2014B. Figures 21A-21C As shown in the figure, the second pin 2014B is positioned relative to the first pin 2014A, and the second opening 2026B is positioned relative to the first opening 2026A.

[0127] Thus, one or more pins 2014 and openings 2026 operate to restrict movement of the second segment 2010 relative to the first segment 2008 along the axis of the first support 1716. For example, when the OAM card 1714 is under compression under load, the compliant structure 2012 allows the second segment 2010 to move relative to the first segment 2008, assuming the first segment 2008 is fixed (e.g., fixed to the first mounting plate 1704). Thus, the second segment 2010 can move axially within the first cavity 2020, reducing the stress applied to the bottom of the second segment 2010. Accordingly, the first support 1716 applies a first stress at a first location on the PCB 1706, which is less than the second stress applied by the second support 1718 at a second location on the PCB 1706.

[0128] Figure 22 The diagram illustrates a second example implementation when the first support 1716 of Figure 17 is assembled into the example OAM card 1714. Figure 22 In the example, the compliant structure 2012 is implemented by a rubber cylinder having an opening to allow the first fastener 1710A to pass through the compliant structure 2012. As described above, in additional or alternative examples, the compliant structure 2012 may be implemented by at least one of the following: silicone, foam, or other elastic compressible material, and / or SMA.

[0129] Figure 23 The diagram illustrates a third example implementation when the first support 1716 of Figure 17 is assembled into the example OAM card 1714. Figure 23 In the example, the first support 1716 includes a first example segment 2302 and a second example segment 2304. Furthermore, in... Figure 23 In one example, the first segment 2302 is adjacent to the first surface 2016 of the first mounting plate 1704 and extends from the first surface 2016 of the first mounting plate 1704 toward the second surface 2018 of the PCB 1706. For example, the first segment 2302 is an integral extension of the first mounting plate 1704 (e.g., a metal plate). In some examples, the first segment 2302 is different from and coupled to (e.g., threaded to, adhered to, soldered to, etc.) the first mounting plate 1704.

[0130] exist Figure 23 In the illustrated example, the second segment 2304 extends from the first segment 2302 toward the second surface 2018 of the PCB 1706. For example, the second segment 2304 is an integral extension of the first segment 2302. In some examples, the second segment 2304 is different from and coupled to (e.g., threaded to, adhered to, soldered to, etc.) the first segment 2302. Figure 23In some examples, the second segment 2304 extends from the first segment 2302 and rests on the second surface 2018 of the PCB 1706. In some examples, the second segment 2304 is coupled to the PCB 1706 (e.g., via threads, adhesive, solder, etc.). In some examples, the first support 1716 (e.g., the first segment 2302 and the second segment 2304) is independent of the first mounting plate 1704 and is secured between the first mounting plate 1704 and the PCB 1706 by compressive force.

[0131] exist Figure 23 In the illustrated example, the first segment 2302 and the second segment 2304 are made of metals such as stainless steel, aluminum, and / or brass. In some examples, the first segment 2302 and the second segment 2304 are coated with an electrically insulating material. Figure 23 In one example, the first support 1716 is cylindrical in shape (e.g., has a shape defined by a cylinder), and the first segment 2302 defines an example outer periphery 2306 (e.g., the outermost periphery) of the first support 1716. In other examples, the outer periphery 2306 of the first support 1716 has a shape other than cylindrical. Furthermore, the first support 1716 has a first example cavity 2308A and a second example cavity 2308B, which define an example inner periphery 2310 of the first support 1716.

[0132] exist Figure 23 In the illustrated example, at least one of the first cavity 2308A or the second cavity 2308B is positioned toward (e.g., facing) the first component 1902 (and / or the associated solder therewith). For example, the second cavity 2308B is positioned toward the first component 1902 (and / or the associated solder therewith). Furthermore, the held portion of the first support 1716 (e.g., the second segment 2304) is positioned away from the first component 1902 (and / or the associated solder therewith).

[0133] exist Figure 23 In the illustrated example, the first segment 2302 has a first cross-sectional shape defining a first region, and the second segment 2304 has a second cross-sectional shape defining a second region smaller than the first region. Thus, even if the first example end 2314 of the first support 1716 is within a threshold distance 1904 of the first component 1902 (and / or the associated solder), the example distal end 2312 of the first support 1716 is also spaced apart from the first component 1902 (and / or the associated solder) by at least the threshold distance 1904. For example, the distal end 2312 is spaced apart from the first component 1902 in a direction normal to the outer surface of the first support 1716.

[0134] exist Figure 23In the illustrated example, the example distance 2316 between the inner periphery 2310 and the first component 1902 (and / or its associated solder) is greater than or equal to a threshold distance 1904. Thus, the first support 1716 includes a gap cut (e.g., at least one of a first cavity 2308A or a second cavity 2308B) that ensures the first component 1902 (and / or its associated solder) is sufficiently far away from the portion of the first support 1716 that abuts against the PCB 1706 to avoid problematic stress concentration on the PCB 1706. Additionally, the distal end 2312 of the first support 1716 has a rounded edge (e.g., at least one rounded edge) that distributes stress more evenly along the second surface 2018 of the PCB 1706. Therefore, the first support 1716 applies a first stress at a first location on the PCB 1706 that is less than the second stress applied by the second support 1718 at a second location on the PCB 1706.

[0135] Figure 24A yes Figure 23 Isometric view of the first support 1716 Figure 24B yes Figure 23 The side view of the first support 1716, and Figure 24C yes Figure 23 Another isometric view of the first support 1716. Figures 24A-24C Collectively referred to as FIG. 24. As illustrated in FIG. 24, the first support 1716 has an example through-hole 2402 (e.g., a mounting hole) extending along the axial length of the first support 1716. Furthermore, as illustrated in FIG. 24, the first cavity 2308A and the second cavity 2308B intersect with the through-hole 2402. In other examples, the first cavity 2308A and the second cavity 2308B do not intersect with the through-hole 2402.

[0136] In the illustrated example of Figure 24, each of the first cavity 2308A and the second cavity 2308B is defined by a first example planar surface 2404 and a second example planar surface 2406, the first example planar surface 2404 being approximately parallel to the axial length of the first support 1716, and the second example planar surface 2406 being approximately perpendicular to the axial length of the first support 1716. As used herein, approximately parallel is defined to mean exactly parallel or within five degrees of being exactly parallel, and approximately perpendicular is defined to mean exactly perpendicular or within five degrees of being exactly perpendicular. In other examples, the first planar surface 2404 and the second planar surface 2406 may be at any other suitable angle relative to the axial length of the first support 1716.

[0137] Figure 25AThe figure shows a cross-sectional view of a fourth example implementation when the first support 1716 of Figure 17 is assembled in the OAM card 1714. Figure 25B The diagram shows Figure 25A Isometric view of the first support 1716. Figure 25A and Figure 25B Collectively referred to as FIG. 25. In the example of FIG. 25, the first fastener 1710A extends through a first example mounting hole 2502 of the first mounting plate 1704, a first support 1716, a second example mounting hole 2504 of the PCB 1706, and a third example mounting hole 2506 of the second mounting plate 1708. Thus, the first support 1716 surrounds (e.g., is to surround) the first fastener 1710A. Furthermore, in the example of FIG. 25, the first support 1716 includes a first example segment 2508 and a second example segment 2510.

[0138] In the example illustrated in Figure 25, the first segment 2508 is adjacent to the first surface 2016 of the first mounting plate 1704 and extends from the first surface 2016 of the first mounting plate 1704 toward the second surface 2018 of the PCB 1706. For example, the first segment 2508 is an integral extension of the first mounting plate 1704 (e.g., a metal plate). In some examples, the first segment 2508 is different from and coupled to (e.g., threaded to, adhered to, soldered to, etc.) the first mounting plate 1704.

[0139] In the illustrated example of Figure 25, the second segment 2510 extends from the first segment 2508 toward the second surface 2018 of the PCB 1706. For example, the second segment 2510 is an integral extension of the first segment 2508. In some examples, the second segment 2510 is different from and coupled to (e.g., threaded to, adhered to, soldered to, etc.) the first segment 2508. In the example of Figure 25, the second segment 2510 extends from the first segment 2508 and rests on the second surface 2018 of the PCB 1706. In some examples, the second segment 2510 is coupled to the PCB 1706 (e.g., via threads, adhesive, solder, etc.). In some examples, the first support 1716 (e.g., the first segment 2508 and the second segment 2510) is independent of the first mounting plate 1704 and is secured between the first mounting plate 1704 and the PCB 1706 by compressive force.

[0140] In the illustrated example of Figure 25, the first segment 2508 and the second segment 2510 are made of metals such as stainless steel, aluminum, and / or brass. In some examples, the first segment 2508 and the second segment 2510 are coated with an electrically insulating material. In the example of Figure 25, the first support 1716 is cylindrical in shape (e.g., having a shape defined by a cylinder). In other examples, the first support 1716 has a shape other than a cylinder. Furthermore, in the example of Figure 25, the example distal end 2512 of the second segment 2510 has a chamfered edge. For example, the chamfered edge of the distal end 2512 defines an example outer perimeter 2514 and an example inner perimeter 2516 of the first support 1716.

[0141] In the illustrated example of FIG. 25, the chamfered edge of the distal end 2512 defines an example cavity 2518 in the first support 1716. In the example of FIG. 25, at least some of the cavity 2518 is positioned toward (e.g., facing) the first component 1902 (and / or the solder associated therewith). Furthermore, the held portion of the first support 1716 (e.g., the portion of the distal end 2512 defined by the inner periphery 2516) is positioned away from the first component 1902 (and / or the solder associated therewith).

[0142] In the example illustrated in FIG25, the first segment 2508 has a first cross-sectional shape defining a first region, and the distal end 2512 of the second segment 2510 has a second cross-sectional shape defining a second region smaller than the first region. In this example, the first and second regions have the same shape (e.g., circular) but different sizes. Thus, even though the first example end 2520 of the first support 1716 is within a threshold distance 1904 of the first component 1902 (and / or the solder associated therewith), the distal end 2512 of the first support 1716 is spaced apart from the first component 1902 (and / or the solder associated therewith) by at least a threshold distance 1904. For example, the distal end 2512 is spaced apart from the first component 1902 in a direction normal to the outer surface of the first support 1716.

[0143] In the illustrated example of Figure 25, the example distance 2522 between the inner periphery 2516 and the first component 1902 (and / or its associated solder) is greater than or equal to the threshold distance 1904. Thus, the first support 1716 includes a gap cut (e.g., cavity 2518) that ensures the first component 1902 (and / or its associated solder) is sufficiently far away from the portion of the first support 1716 that contacts the PCB 1706 to avoid problematic stress concentration on the PCB 1706. Additionally, the distal end 2512 of the first support 1716 is a chamfered edge that distributes stress more evenly along the second surface 2018 of the PCB 1706.

[0144] As described above, the first fastener 1710A extends through the first support 1716 illustrated in FIG. 25. For example, the first support 1716 has an example through-hole 2524 (e.g., a mounting hole) extending along the axial length of the first support 1716. Furthermore, as illustrated in FIG. 25, a cavity 2518 is spaced apart from the through-hole 2524. In the example of FIG. 25, the cavity 2518 is defined by a tapered surface that is curved about the axial length of the first support 1716 and inclined relative to the axial length of the first support 1716. For example, the tapered surface defining the cavity 2518 is inclined at 30 degrees relative to the axial length of the first support 1716.

[0145] Figure 26 The figure shows a cross-sectional view of a fifth example implementation of the first support 1716 in Figure 17. Figure 26 In the example, the first fastener 1710A extends through a first example mounting hole 2602 of the first mounting plate 1704, a first support 1716, a second example mounting hole 2604 of the PCB 1706, and a third example mounting hole 2606 of the second mounting plate 1708. Thus, the first support 1716 surrounds (e.g., is to surround) the first fastener 1710A. Furthermore, in Figure 26 In the example, the first support 1716 includes a first example end 2608 and a second example end 2610.

[0146] exist Figure 26 In the illustrated example, the first end 2608 is adjacent to the first surface 2016 of the first mounting plate 1704 and extends from the first surface 2016 of the first mounting plate 1704 toward the second surface 2018 of the PCB 1706. In some examples, the first support 1716 is an integral extension of the first mounting plate 1704 (e.g., a metal plate). In some examples, the first support 1716 is different from and coupled to (e.g., threaded to, adhered to, soldered to, etc.) the first mounting plate 1704. Figure 26 In the example, the second end 2610 of the first support 1716 extends toward the second surface 2018 of the PCB 1706, but does not contact the PCB 1706.

[0147] exist Figure 26 In the illustrated example, the first end 2608 and the second end 2610 are made of metals such as stainless steel, aluminum, and / or brass. In some examples, the first support 1716 is coated with an electrically insulating material. Figure 26 In one example, the first support 1716 is cylindrical in shape (e.g., having a shape defined by a cylinder). In other examples, the first support 1716 has a shape other than a cylinder. As described above, the first fastener 1710A extends through... Figure 26The first support 1716 is illustrated in the figure. For example, the first support 1716 has an example through-hole 2612 (e.g., a mounting hole) extending along the axial length of the first support 1716. Advantageously, at least because the first support 1716 does not contact the PCB 1706, the first component 1902 (and / or its associated solder) is not subjected to stress when the OAM card 1714 (e.g., by compressive force applied to the first fastener 1710A) is loaded. In some such examples, due to other supports between the first mounting plate 1704 and the PCB 1706 (e.g., those in Figure 17 and / or Figure 18), the first component 1902 is not subjected to stress. Figure 19 The second support 1718 is longer than the first support 1716, thus maintaining a gap between the second end 2610 of the first support 1716 and the PCB 1706. In some examples, during impact loading, vibration loading, and / or another loading event, the second end 2610 of the first support 1716 may come into contact with the second surface 2018 of the PCB 1706. However, the stress generated on the PCB 1706 by the forces of such contact will be less than if the first support 1716 is always in contact with the PCB 1706.

[0148] Figure 27 The figure shows a cross-sectional view of a sixth example implementation of the first support 1716 in Figure 17. Figure 27 In the example, the first fastener 1710A extends through a first example mounting hole 2702 of the first mounting plate 1704, a first support 1716, a second example mounting hole 2704 of the PCB 1706, and a third example mounting hole 2706 of the second mounting plate 1708. Thus, the first support 1716 surrounds (e.g., is to surround) the first fastener 1710A. Furthermore, in Figure 27 In the example, the first support 1716 includes a first example segment 2708, a second example segment 2710, and an example compliance structure 2712.

[0149] exist Figure 27 In the illustrated example, the first segment 2708 is adjacent to the first surface 2016 of the first mounting plate 1704 and extends from the first surface 2016 of the first mounting plate 1704 toward the second surface 2018 of the PCB 1706. For example, the first segment 2708 is an integral extension of the first mounting plate 1704 (e.g., a metal plate). In some examples, the first segment 2708 is different from and coupled to (e.g., threaded to, adhered to, soldered to, etc.) the first mounting plate 1704. Figure 27In the example, the second segment 2710 extends from the first segment 2708 toward the second surface 2018 of the PCB 1706. For example, the second segment 2710 is an integral extension of the first segment 2708. In some examples, the second segment 2710 is different from and coupled to (e.g., threaded to, adhered to, soldered to, etc.) the first segment 2708. Figure 27 In the example, the second segment 2710 extends from the first segment 2708 toward the second surface 2018 of PCB 1706, but does not contact PCB 1706.

[0150] exist Figure 27 In the illustrated example, the first segment 2708 and the second segment 2710 are made of metals such as stainless steel, aluminum, and / or brass. In some examples, the first segment 2708 and the second segment 2710 are coated with an electrically insulating material. Figure 27 In the example, the first segment 2708 and the second segment 2710 operate as spacers with a cylindrical shape (e.g., having a shape defined by a cylinder). In other examples, the spacers have shapes other than cylinders. Furthermore, the spacers (e.g., the first segment 2708 and the second segment 2710) extend a first portion of the length of the first support 1716.

[0151] exist Figure 27 In the illustrated example, the compliant structure 2712 extends a second portion of the length of the first support 1716, wherein the first portion differs from the second portion. For example, the compliant structure 2712 is positioned between the spacer (e.g., the first segment 2708 and the second segment 2710) and the second surface 2018 of the PCB 1706. Figure 27 In one example, the compliant structure 2712 is compressively held between the first example end 2714 of the second segment 2710 and the second surface 2018 of the PCB 1706. For example, the compliant structure 2712 is implemented by a rubber ring, which is compressively held between the first end 2714 and the second surface 2018 of the PCB 1706. In additional or alternative examples, the compliant structure 2712 is implemented by at least one of the following: a spring, silicone, foam, or SMA. In some examples, the compliant structure 2712 is secured to the first support 1716 via an adhesive. In other examples, the compliant structure 2712 is not secured to the first support 1716, but is held in place by compressive force and / or by a first fastener 1710A.

[0152] exist Figure 27In the illustrated example, the compliant structure 2712 is a biasing component that biases the spacers (e.g., the first segment 2708 and the second segment 2710) relative to the PCB 1706. For example, the compliant structure 2712 applies a force to at least one of the second segment 2710 or the PCB 1706 to hold the second segment 2710 in a first position relative to the PCB 1706. Additionally or alternatively, the compliant structure 2712 applies a force to at least one of the second segment 2710 or the PCB 1706 to move the second segment 2710 relative to the PCB 1706 (e.g., induced movement of the second segment 2710 relative to the PCB 1706).

[0153] As described above, the first fastener 1710A extends through... Figure 27 The first support 1716 is illustrated in the figure. For example, the first support 1716 has an example through-hole 2716 (e.g., a mounting hole) extending along the axial length of the first support 1716. Advantageously, at least because the compliant structure 2712 is positioned between the first end 2714 of the second segment 2710 and the second surface 2018 of the PCB 1706, the first component 1902 (and / or the associated solder) will not be subjected to sufficient stress to damage the first component 1902 when the OAM card 1714 (e.g., by the compressive force applied to the first fastener 1710A) is loaded. For example, the compliant structure 2712 buffers (e.g., dampens) the movement of the spacers (e.g., the first segment 2708 and the second segment 2710) on the PCB 1706.

[0154] Thus, the length of the first support 1716 is adjustable (e.g., has an adjustable length), and the compliant structure 2712 allows the first support 1716 to absorb stress. For example, when the OAM card 1714 is subjected to loading (e.g., impact loading, vibration loading, thermal cycling, etc.), the spacers (e.g., the first segment 2708 and the second segment 2710) can move as the compliant structure 2712 compresses under load. In this way, even if the first component 1902 is within a threshold distance 1904 of the first support 1716, the first support 1716 prevents the generation of localized stress in the first component 1902, its solder joints, or the PCB 1706. Therefore, the first support 1716 reduces the failure of at least one of the PCB 1706 or the first component 1902 due to localized stress concentration in the area surrounding the first support 1716.

[0155] Figure 28A The figure shows a cross-sectional view of the seventh example implementation of the first support 1716 in Figure 17. Figure 28B The diagram shows Figure 28A Isometric view of the first support 1716. Figure 28A and Figure 28B Collectively referred to as FIG. 28. In the example of FIG. 28, the first fastener 1710A extends through a first example mounting hole 2802 of the first mounting plate 1704, a first support 1716, a second example mounting hole 2804 of the PCB 1706, and a third example mounting hole 2806 of the second mounting plate 1708. Thus, the first support 1716 surrounds (e.g., is to surround) the first fastener 1710A. Furthermore, in the example of FIG. 28, the first support 1716 includes a first example segment 2808 and a second example segment 2810.

[0156] In the example illustrated in Figure 28, the first segment 2808 is adjacent to the first surface 2016 of the first mounting plate 1704 and extends from the first surface 2016 of the first mounting plate 1704 toward the second surface 2018 of the PCB 1706. For example, the first segment 2808 is an integral extension of the first mounting plate 1704 (e.g., a metal plate). In some examples, the first segment 2808 is different from and coupled to (e.g., threaded to, adhered to, soldered to, etc.) the first mounting plate 1704.

[0157] In the illustrated example of Figure 28, the second segment 2810 extends from the first segment 2808 toward the second surface 2018 of the PCB 1706. For example, the second segment 2810 is an integral extension of the first segment 2808. In some examples, the second segment 2810 is different from and coupled to (e.g., threaded to, adhered to, soldered to, etc.) the first segment 2808. In the example of Figure 28, the second segment 2810 extends from the first segment 2808 and rests on the second surface 2018 of the PCB 1706. In some examples, the second segment 2810 is coupled to the PCB 1706 (e.g., via threads, adhesive, solder, etc.). In some examples, the first support 1716 (e.g., the first segment 2808 and the second segment 2810) is independent of the first mounting plate 1704 and is secured between the first mounting plate 1704 and the PCB 1706 by compressive force.

[0158] In the illustrated example of Figure 28, the first segment 2808 and the second segment 2810 are made of metals such as stainless steel, aluminum, and / or brass. In some examples, the first segment 2808 and the second segment 2810 are coated with an electrically insulating material. In the example of Figure 28, the first support 1716 is cylindrical in shape (e.g., having a shape defined by a cylinder), and the first segment 2808 defines an example outer periphery 2812 of the first support 1716. In other examples, the first support 1716 has a shape other than a cylinder. Furthermore, the first support 1716 has an example cavity 2814 that defines an example inner periphery 2816 of the first support 1716.

[0159] In the example illustrated in Figure 28, cavity 2814 is positioned toward (e.g., facing) the first component 1902 (and / or the associated solder therewith). For example, cavity 2814 is positioned toward the first component 1902 (and / or the associated solder therewith). Furthermore, the held portion of the first support 1716 (e.g., the second segment 2810) is positioned away from the first component 1902 (and / or the associated solder therewith).

[0160] In the example illustrated in FIG28, the first segment 2808 has a first cross-sectional shape defining a first region, and the second segment 2810 has a second cross-sectional shape defining a second region smaller than the first region. Thus, even if the first example end 2820 of the first support 1716 is within a threshold distance 1904 of the first component 1902 (and / or the associated solder therewith), the example distal end 2818 of the first support 1716 is also spaced apart from the first component 1902 (and / or the associated solder therewith) by at least a threshold distance 1904. For example, the distal end 2818 is spaced apart from the first component 1902 in a direction normal to the outer surface of the first support 1716.

[0161] In the illustrated example of Figure 28, the example distance 2822 between the inner periphery 2816 and the first component 1902 (and / or its associated solder) is greater than or equal to the threshold distance 1904. Thus, the first support 1716 includes a cut feature (e.g., cavity 2814) near the first component 1902 that ensures the first component 1902 (and / or its associated solder) is sufficiently far away from the portion of the first support 1716 that contacts the PCB 1706 to avoid problematic stress concentration on the PCB 1706. Additionally, the shape of the distal end 2818 of the first support 1716 is designed such that stress is distributed away from the first component 1902 (and / or its associated solder).

[0162] As described above, the first fastener 1710A extends through the first support 1716 illustrated in FIG. 28. For example, the first support 1716 has an example through-hole 2824 (e.g., a mounting hole) extending along the axial length of the first support 1716. Furthermore, as illustrated in FIG. 28, the cavity 2814 is spaced apart from the through-hole 2824. In other examples, the cavity 2814 may intersect with the through-hole 2824. In the example of FIG. 28, the cavity 2814 is defined by an example planar surface 2826 that is angled relative to the axial length of the first support 1716. For example, the planar surface 2826 defining the cavity 2814 is at a 30-degree angle relative to the axial length of the first support 1716. In other examples, the planar surface 2826 may be at any other suitable angle.

[0163] Figure 29 The figure shows a cross-sectional view of the eighth example implementation of the first support 1716 in Figure 17. Figure 29 In the example, the first fastener 1710A extends through a first example mounting hole 2902 of the first mounting plate 1704, a first support 1716, a second example mounting hole 2904 of the PCB 1706, and a third example mounting hole 2906 of the second mounting plate 1708. Thus, the first support 1716 surrounds (e.g., is to surround) the first fastener 1710A. Furthermore, in Figure 29 In the example, the first support 1716 includes a first example end 2908 and a second example end 2910.

[0164] exist Figure 29 In the illustrated example, the first end 2908 is adjacent to the first surface 2016, and the first support 1716 extends from the first surface 2016 of the first mounting plate 1704 toward the second surface 2018 of the PCB 1706. In some examples, the first support 1716 is coupled to the first mounting plate 1704 (e.g., via adhesive, solder, etc.). In some examples, the first support 1716 is different from and coupled to (e.g., adhered to, soldered to, etc.) the first mounting plate 1704. Figure 29 In one example, the first support 1716 extends from the first surface 2016 and rests on the second surface 2018 of the PCB 1706. In some examples, the first support 1716 is coupled to the PCB 1706 (e.g., via adhesive, solder, etc.). Figure 29 In the example, the first support 1716 is fixed between the first mounting plate 1704 and the PCB 1706 by compressive force.

[0165] exist Figure 29 In the illustrated example, the first support 1716 is made of a compliant structure. For example, the first support 1716 is a rubber cylinder having an example through-hole 2912. That is, the first support 1716 is implemented by at least one rubber element with a cylindrical shape (e.g., having a shape defined by a cylinder), and the through-hole 2912 extends along the axial length of the first support 1716. In additional or alternative examples, the first support 1716 is implemented by at least one of the following: a spring, silicone, foam, or SMA. In some examples, the first support 1716 is coated with an electrically insulating material.

[0166] exist Figure 29In the illustrated example, the first support 1716 is a biasing assembly extending the entire length of the first support 1716 between the first mounting plate 1704 and the PCB 1706. Additionally, the biasing assembly of the first support 1716 biases the first mounting plate 1704 relative to the PCB 1706. For example, the first support 1716 applies a force to at least one of the first mounting plate 1704 or the PCB 1706 to hold the first mounting plate 1704 in a first position relative to the PCB 1706. Additionally or alternatively, the first support 1716 applies a force to at least one of the first mounting plate 1704 or the PCB 1706 to move the first mounting plate 1704 relative to the PCB 1706 (e.g., induced movement of the first mounting plate 1704 relative to the PCB 1706).

[0167] As described above, the first fastener 1710A extends through... Figure 29 The first support 1716 is illustrated in the figure. Advantageously, at least because the first support 1716 is implemented with a compliant structure (e.g., rubber) and is positioned between the first surface 2016 and the second surface 2018, when the OAM card 1714 is loaded (e.g., by the compressive force applied to the first fastener 1710A), the first component 1902 (and / or the associated solder) will not be subjected to sufficient stress to damage the first component 1902. For example, the first support 1716 cushions the impact of movement of the first mounting plate 1704 on the PCB 1706.

[0168] Thus, the length of the first support 1716 is adjustable (e.g., has an adjustable length), and the compliant structure of the first support 1716 allows it to absorb stress. For example, when the OAM card 1714 is subjected to a load (e.g., impact loading, vibration loading, thermal cycling, etc.), the first support 1716 can move as the compliant structure 2012 compresses under load. In this way, even if the first component 1902 is within a threshold distance 1904 of the first support 1716, the first support 1716 prevents the generation of localized stress in the first component 1902, its solder joints, or the PCB 1706. Therefore, the first support 1716 reduces the failure of at least one of the PCB 1706 or the first component 1902 due to localized stress concentration in the area surrounding the first support 1716.

[0169] In addition to reducing localized stress concentration in PCB 1706, the examples disclosed herein also improve thermal management in OAM card 1714. For example, the compliant support disclosed herein (e.g., implemented with springs or other compliant structures such as rubber) improves thermal management within OAM card 1714 by allowing some degree of independent movement between PCB 1706 and first mounting plate 1704. This movement between PCB 1706 and first mounting plate 1704 stimulates airflow over PCB 1706, which aids in heat dissipation.

[0170] Additionally, example supports implemented by SMA can aid in thermal management. For example, as described above, the first support 1716 can be implemented by SMA. An SMA is an alloy that can define a first shape at a first temperature and can be modified to define a different second shape when the temperature of the SMA changes to meet a temperature threshold. For example, an SMA can deform below a temperature threshold and can recover to a pre-deformed (e.g., memory) shape when heated to or above the temperature threshold. Example SMAs include copper-aluminum-nickel and nickel-titanium, as well as SMAs created by alloying zinc, copper, gold, and / or iron.

[0171] Thus, in the example where the first support 1716 is implemented by an SMA, if the PCB 1706 begins to overheat due to Joule heating, the first support 1716 aids in heat dissipation by transferring heat from the PCB 1706 to the first support 1716, causing the first support 1716 to expand (e.g., from a first length to a second length, the second length being greater than a third length of the second support 1718). Therefore, the expansion of the first support 1716 raises the first mounting plate 1704, thereby allowing increased airflow above the PCB 1706, thus dissipating heat. Cooling caused by the dissipated heat causes the first support 1716 to contract (e.g., from a second length to a first length, the first length being greater than or equal to a third length), thereby returning the first mounting plate 1704 to its original position. The above process is automatic and can be implemented by an SMA as described above, which acts as a thermomechanical switch driven by the thermal load. For example, since heating and cooling occur through the use of the OAM card 1714, no external control is required. As a result, the thermodynamic efficiency of OAM card 1714 is improved.

[0172] Furthermore, for example, one or more independent heaters can be used to control the SMA support. For instance, in an example where the first support 1716 is implemented using SMA, when a mechanical load is applied to the OAM 1700, one or more independent heaters can be activated (e.g., triggered by a controller, by temperature conditions of the OAM card 1714, etc.) to cause the first support 1716 to contract, thereby reducing the shape factor (e.g., height, length) of the first support 1716 in the Z direction. Consequently, the first support 1716 presses against the PCB 1706 with a smaller force (e.g., fully contracted) to space the first support 1716 from the PCB 1706. Thus, the contraction of the first support 1716 reduces the stress imposed on the PCB 1706 at the location of the first support 1716. Therefore, the first support 1716 reduces the stress on the PCB 1706 based on the thermal load from one or more independent heaters.

[0173] In the examples disclosed herein, by (1) utilizing supports such as the first support 1716 over high-risk areas of the PCB that cause localized stress, and (2) utilizing supports such as the second support 1718 over low-risk areas of the PCB that cause localized stress, the examples disclosed herein reduce stress concentration in the example OAM card. Furthermore, utilizing example-compatible supports helps absorb thermal expansion and contraction, thereby mitigating stress transfer to the PCB and components. Additionally, by optimizing the geometry of the example supports disclosed herein (e.g., by adding rounded or chamfered corners to the portions of the supports that contact the PCB), the examples disclosed herein distribute stress more evenly across the surface of the PCB.

[0174] As described above, the examples disclosed herein reduce stress concentration. For example, by implementing example supports with different cross-sectional shapes at opposite ends of the supports, localized stress concentration around mounting holes can be reduced, thereby enhancing the structural integrity of the PCB and improving the reliability of SMT components. Furthermore, for example, by implementing example supports incorporating compliant materials, component protection is maintained even when localized board strain can be reduced.

[0175] The example supports disclosed herein reduce stress near component corners (e.g., corners of high-risk components) without significantly altering the mounting (e.g., retaining) plate design. Therefore, the examples disclosed herein can be implemented without significantly changing the thermomechanical design of the OAM card. Additionally, the examples disclosed herein do not subject other design variables of the OAM card to greater risk. Thus, the examples disclosed herein can be used to standardize the design of OAM cards. For example, the disclosed examples establish a standardized design for OAM card retaining, which promotes consistency, manufacturability, and ease of use across OAM card configurations.

[0176] Despite Figures 20-29 The diagram illustrates an example of implementing the first support 1716 of Figure 17, but... Figures 20-29 One or more of the elements, features, and / or structures illustrated herein may be combined, separated, rearranged, omitted, eliminated, and / or implemented in any other way. Therefore, it should be understood that... Figures 20-29 The examples are not mutually exclusive, but can be used in any combination. For example, combining... Figure 23-Figure 2 Any of the cuts, fillets, and / or chamfers described in Figures 5 and 28 can be combined with... Figures 20-22 In the second paragraph of 2010. Additional or alternative land, similar to combination. Figure 27 The compliant structures illustrated and described can be added to Figures 20-26 And the ends of any of the supports shown in Figure 28. Furthermore, Figures 20-29 One or more elements, features, and / or structures of any of the supports in the document can be implemented using SMA. Thus, in addition to suppressing movement between the mounting plate and the PCB, the example supports can also provide thermal management of the OAM card by changing the shape factor (e.g., height, length) in the Z direction based on heat. In some examples, several different example supports disclosed herein can be used at different locations within a single OAM card. For example, the example supports disclosed herein can be used over areas of the PCB at high risk of causing localized stress. Additionally or alternatively, supports such as the second support 1718 of Figure 17 can be used over areas of the PCB at low risk of causing localized stress. Thus, the examples disclosed herein reduce stress concentration in the OAM card.

[0177] "Comprising" and "including" (and all their forms and tenses) are used herein as open-ended terms. Therefore, whenever a claim uses any form of "comprising" or "including" (e.g., including, containing, comprising, having, etc.) as a preamble or in the content of any kind of claim, it should be understood that additional elements, terms, etc., may be present and not fall outside the scope of the corresponding claim or statement. As used herein, when the phrase "at least" is used as a transitional term, for example, in the preamble of a claim, it is open-ended in the same way that the terms "comprising" and "including" are open-ended. The term "and / or" when used, for example, in forms such as A, B, and / or C, refers to any combination or subset of A, B, and C, such as (1) only A, (2) only B, (3) only C, (4) A and B, (5) A and C, (6) B and C, or (7) A and B and C. As used herein in the context of describing structures, components, items, objects, and / or things, the phrase “at least one of A and B” is intended to refer to an implementation that includes any one of the following: (1) at least one A, (2) at least one B, or (3) at least one A and at least one B. Similarly, as used herein in the context of describing structures, components, items, objects, and / or things, the phrase “at least one of A and B” is intended to refer to an implementation that includes any one of the following: (1) at least one A, (2) at least one B, or (3) at least one A and at least one B. As used herein in the context of describing the conduct or execution of processes, instructions, actions, activities, etc., the phrase “at least one of A and B” is intended to refer to an implementation that includes any one of the following: (1) at least one A, (2) at least one B, or (3) at least one A and at least one B. Similarly, as used in this document in the context of describing the conduct or execution of processes, instructions, actions, activities, etc., the phrase “at least one of A or B” is intended to refer to an implementation that includes any of the following: (1) at least one A, (2) at least one B, or (3) at least one A and at least one B.

[0178] As used herein, singular references (e.g., "a (a, an)", "first", "second", etc.) do not exclude plurals. The term "a ("a" or "an")" as used herein refers to one or more of those objects. The terms "a" ("a" or "an"), "one or more", and "at least one" are used interchangeably herein. Furthermore, although listed separately, multiple means, elements, or actions may be implemented by, for example, the same entity or object. Additionally, although individual features may be included in different examples or claims, these features may be combined, and inclusion in different examples or claims does not imply that the combination of features is not feasible and / or not advantageous.

[0179] As used herein, unless otherwise stated, the term "above" describes the relationship between two components relative to the ground. The first component is above the second component if the second component has at least one portion situated between the ground and the first component. Similarly, as used herein, the first component is "below" the second component when the first component is closer to the ground than the second component. As stated above, the first component may be above or below the second component in one or more of the following situations: when there are other components between them, when there are no other components between them, when the first component and the second component are in contact, or when the first component and the second component are not in direct contact with each other.

[0180] As used in this patent, it is stated that any component (e.g., layer, film, part, region, or plate) is in any way (e.g., positioned on, located on, disposed on, or formed on, etc.) another component indicating that the referenced component is in contact with that other component, or that the referenced component is above that other component and one or more intermediate components are positioned between the referenced component and the other component.

[0181] As used herein, unless otherwise indicated, a connection reference (e.g., attached, coupled, connected, and joined) may include intermediate components between the elements referenced by the connection reference and / or relative movement between those elements. Thus, a connection reference does not necessarily imply that two elements are directly connected and / or in a fixed relationship with each other. As used herein, the statement that any component is in “contact” with another component is defined to mean that there is no intermediate component between the two components.

[0182] Unless otherwise specified, descriptors such as “first,” “second,” “third,” etc., are used herein without imposing or otherwise indicating any meaning of priority, physical order, arrangement in a list, and / or any sorting, but merely as labels and / or arbitrary names to distinguish elements for ease of understanding of the disclosed examples. In some examples, the descriptor “first” may be used to refer to an element in a particular embodiment, while different descriptors such as “second” or “third” may be used in the claims to refer to the same element. In such instances, it should be understood that such descriptors are used only to distinguish those elements in the context of this discussion (e.g., in the claims), where elements might otherwise share the same name, for example.

[0183] As used herein, “approximately” and “about” modify their subject matter and / or value to identify the potential presence of variations that occur in real-world applications. For example, “approximately” and “about” may modify dimensions that may be imprecise due to manufacturing tolerances and / or other real-world defects as would be understood by one of ordinary skill in the art. For example, “approximately” and “about” may indicate that such dimensions are within a tolerance of + / - 10%, unless otherwise specified herein.

[0184] As used herein, the phrase “to communicate” (including variations thereof) includes direct communication and / or indirect communication through one or more intermediate components, and does not require direct physical (e.g., wired) communication and / or continuous communication, but additionally includes selective communication performed at periodic intervals, predetermined intervals, non-periodic intervals, and / or one-off events.

[0185] As used herein, a “programmable circuit system” is defined as including (i) one or more dedicated electrical circuits (e.g., application-specific circuits, ASICs) configured to perform one or more specific operations and including one or more semiconductor-based logic devices (e.g., electrical hardware implemented by one or more transistors), and / or (ii) one or more semiconductor-based general-purpose electrical circuits programmable by instructions to perform one or more specific functions and / or one or more operations and including one or more semiconductor-based logic devices (e.g., electrical hardware implemented by one or more transistors). Examples of programmable circuit systems include: programmable microprocessors (such as central processing unit (CPU)) that can execute first instructions to perform one or more operations and / or functions; field-programmable gate arrays (FPGAs) that can be programmed with second instructions to cause the configuration and / or construction of the FPGA to instantiate one or more operations and / or functions corresponding to the first instructions; graphics processing units (GPUs) that can execute first instructions to perform one or more operations and / or functions; digital signal processors (DSPs) that can execute first instructions to perform one or more operations and / or functions; XPUs; network processing units (NPUs); one or more microcontrollers that can execute first instructions to perform one or more operations and / or functions; and / or integrated circuits (such as application-specific integrated circuits (ASICs)). For example, an XPU can be implemented by a heterogeneous computing system that includes various types of programmable circuit systems (e.g., one or more FPGAs, one or more CPUs, one or more GPUs, one or more NPUs, one or more DSPs, etc., and / or any combination thereof) and orchestration techniques (e.g., one or more application programming interfaces (APIs)) that can assign one or more computing tasks to any of the various types of programmable circuit systems that are suitable for and can be used to perform one or more computing tasks.

[0186] As used herein, an integrated circuit / circuit system is defined as one or more semiconductor packages containing one or more circuit elements, such as transistors, capacitors, inductors, resistors, current paths, diodes, etc. For example, an integrated circuit can be implemented as one or more of an ASIC, FPGA, chip, microchip, programmable circuit system, semiconductor substrate coupling multiple circuit elements, system-on-a-chip (SoC), etc.

[0187] As will be understood from the foregoing, example systems, apparatuses, articles of art, and methods for holding printed circuit boards between mounting plates have been disclosed. For example, as described above, the area around the mounting holes on the PCB (e.g., the PCB, its components, solder joints coupling the components to the PCB, etc.) may be subject to localized stress due to the loading on the OAM card. Additionally, the area around the mounting holes on the PCB (e.g., the PCB, its components, solder joints coupling the components to the PCB, etc.) may be subject to reliability issues due to foreign debris, loose tolerances, board quality, etc. (e.g., due to congested board space near the mounting holes on the PCB card).

[0188] Advantageously, the examples disclosed herein mitigate one or more problems that may arise from the close proximity of components to PCB mounting holes through a systematic approach. For example, the disclosed examples include a series of design features to mitigate the risk of PCB and SMT component failures stemming from localized stress concentrations (e.g., caused by localized stress concentrations). One or more example supports disclosed herein can be used in conjunction with a holistic strategy designed to enhance thermomechanical solutions. Examples disclosed herein include OAM card retention technology and contribute to enhancing the durability, reliability, and resilience of OAM cards against stress-induced failures. For example, by integrating the example design features disclosed herein, the examples described herein mitigate stress concentration phenomena and safeguard the integrity of OAM card functionality under operating conditions. The examples disclosed herein achieve higher performance and lifespan within the scope of OAM infrastructure. The disclosed systems, apparatuses, articles, and methods improve the efficiency of using computing devices by improving the stress-resistance reliability of accelerator cards. The disclosed systems, apparatuses, articles, and methods accordingly relate to one or more improvements in the operation of machines such as computers or other electronic and / or mechanical equipment.

[0189] This document discloses example methods, apparatuses, systems, and articles of art for holding printed circuit boards between mounting plates. Further examples and combinations thereof include the following:

[0190] Example 1 includes a support comprising: a first segment adjacent to a first surface of a metal plate; a second segment extending from the first segment toward a second surface of a printed circuit board (PCB), the support being used to maintain separation between the metal plate and the PCB; and at least one of the following: a biasing component for suppressing movement between the metal plate and the PCB; or a distal end of the second segment, the first segment having a first cross-sectional shape defining a first region, the distal end having a second cross-sectional shape defining a second region smaller than the first region.

[0191] Example 2 includes the support of Example 1, wherein the first segment is an integral extension of the metal plate.

[0192] Example 3 includes a support for either Example 1 or 2, wherein the second segment is an integral extension of the first segment.

[0193] Example 4 includes a support for either Example 1 or 2, wherein the second segment is movable relative to the first segment.

[0194] Example 5 includes a support of any of Examples 1, 2, 3 or 4, wherein the support includes a biasing component that includes a spring.

[0195] Example 6 includes an apparatus comprising: a metal plate; a printed circuit board (PCB) for coupling to the metal plate; a first support located between the metal plate and a first position on the PCB for applying a first stress at the first position; and a second support located between the metal plate and a second position on the PCB for applying a second stress at the second position, the first stress being less than the second stress.

[0196] Example 7 includes the apparatus of Example 6, wherein a first support is aligned with a first opening in a PCB, a second support is aligned with a second opening in a PCB, the first opening being within a threshold distance of a first solder for mounting a first component to the PCB, and the second opening being further away from a second solder for mounting a second component to the PCB than a threshold distance, the threshold distance corresponding to the difference between (a) a first radius of the outermost periphery of the respective support in the first and second supports and (b) a second radius of the periphery of the respective opening in the first and second openings.

[0197] Example 8 includes an apparatus of either Example 6 or 7, wherein the first support includes a biasing component for enabling the first support to vary in length between a first end and a second end, the first end being adjacent to a metal plate and the second end being adjacent to a PCB.

[0198] Example 9 includes the apparatus of Example 8, wherein the first support includes: a first spacer extending from a first surface of a metal plate toward a second surface of a PCB; and a second spacer extending from the second surface toward the first surface, the second spacer being movable relative to the first spacer, and a biasing component for causing the second spacer to move toward the PCB relative to the first spacer.

[0199] Example 10 includes the apparatus of Example 8, wherein a first support includes a spacer extending a first portion of the length of the first support, and a biasing component extends a second portion of the length of the support, the first portion being different from the second portion.

[0200] Example 11 includes the apparatus of Example 10, wherein the biasing component is used to be positioned between the spacer and the PCB.

[0201] Example 12 includes an apparatus of either Example 8 or 9, wherein the biasing component extends the entire length of the support between the metal plate and the PCB.

[0202] Example 13 includes an apparatus of any one of Examples 6, 7, 8, 9, 10 or 11, wherein the first support includes a first shape at a first end and a second shape at a second end, the first end being adjacent to a first surface of a metal plate and the second end being adjacent to a second surface of a PCB, the second shape being smaller than the first shape such that when the first end is within a threshold distance of solder on the second surface of the PCB in a direction normal to the outer surface of the first support, the second end is spaced apart from the solder by the threshold distance.

[0203] Example 14 includes the apparatus of Example 13, wherein a first shape is defined by a first cylinder and a second shape is defined by a second cylinder, the side of the second cylinder having cavities for being positioned toward the solder, and a holding portion of the second cylinder for being positioned away from the solder.

[0204] Example 15 includes the device of Example 14, wherein the first support includes a through hole extending along the axial length of the first support, and the cavity is spaced apart from the through hole.

[0205] Example 16 includes the device of Example 14, wherein the first support includes a through hole extending along the axial length of the first support, and the cavity intersects the through hole.

[0206] Example 17 includes the device of Example 14, wherein the cavity is defined by a planar surface angled relative to the axial length of the first support.

[0207] Example 18 includes the device of Example 14, wherein the cavity is defined by a planar surface approximately parallel to the axial length of the first support.

[0208] Example 19 includes the apparatus of Example 13, wherein a first shape is defined by a first cylinder, and a second shape is defined by a second cylinder having at least one of a chamfered edge or a rounded edge.

[0209] Example 20 includes an apparatus comprising: a metal plate; a printed circuit board (PCB) for coupling to the metal plate via a threaded fastener extending through the metal plate and through the PCB; a first support for surrounding a first threaded fastener in the threaded fastener, the first support for separating the metal plate and the PCB; and a second support for surrounding a second threaded fastener in the threaded fastener, the second support for separating the metal plate and the PCB, wherein the first support satisfies at least one of the following: having a different shape than the second support; or comprising a different material than the second support.

[0210] Example 21 includes the apparatus of Example 20, wherein the second support has a fixed length and the first support has an adjustable length.

[0211] Example 22 includes an apparatus of any one of Examples 20 or 21, wherein the first support comprises a shape memory alloy (SMA) and the SMA is used to switch the first support between a first length and a second length based on whether the temperature of the first support meets a temperature threshold, the second length being greater than the first length and greater than a third length of the second support, the third length being less than or equal to the first length.

[0212] Example 23 includes an apparatus of any one of Examples 20, 21, or 22, wherein the PCB has components mounted to the PCB, and the first support includes at least one of: (a) a biasing component; or (b) a spacer extending from a metal plate toward the PCB, the spacer having a first shape at a first end adjacent to a first surface of the metal plate and a second shape at a second end adjacent to a second surface of the PCB, the second shape being smaller than the first shape such that when the first end is within a threshold distance of the component, the second end is at least at the threshold distance from the component.

[0213] Example 24 includes the apparatus of Example 23, wherein the spacer is cylindrical in shape and has a cavity on the side of the spacer facing the assembly, the cavity extending between (a) the PCB and (b) the point between the metal plate and the PCB, such that the retained portion of the assembly and the second end of the spacer are at least a threshold distance apart, the retained portion having rounded edges.

[0214] Example 25 includes the apparatus of Example 23, wherein the spacer is cylindrical in shape, extends from the metal plate to the PCB, and has a chamfered edge at a second end such that the component is at least a threshold distance from the inner periphery of the chamfered edge.

[0215] Example 26 includes the device of Example 23, wherein the spacer is cylindrical in shape and has a cavity on the side of the spacer facing the assembly, the cavity being defined by a planar surface angled relative to the axial length of the spacer.

[0216] Example 27 includes the device of Example 23, wherein the spacer is cylindrical in shape and has a cavity on the side of the spacer facing the assembly, the cavity being defined by a planar surface approximately parallel to the axial length of the spacer.

[0217] Example 28 includes an apparatus of any one of Examples 23, 24, 25, 26 or 27, wherein the spacer extends from the metal plate toward the PCB and the biasing component is positioned between the spacer and the PCB.

[0218] The appended claims are hereby incorporated herein by reference. While certain example systems, apparatuses, articles of manufacture, and methods have been disclosed herein, the scope of this patent is not limited thereto. Rather, this patent covers all systems, apparatuses, articles of manufacture, and methods that, fairly speaking, fall within the scope of the claims of this patent.

Claims

1. A support, comprising: The first segment is adjacent to the first surface of the metal plate; The second segment extends from the first segment toward the second surface of the printed circuit board (PCB), and the support is used to maintain the separation between the metal plate and the PCB. as well as At least one of the following: Bias components are used to suppress movement between the metal plate and the PCB; or The distal end of the second segment, wherein the first segment has a first cross-sectional shape defining a first region, and the distal end has a second cross-sectional shape defining a second region, the second region being smaller than the first region.

2. The support according to claim 1, wherein, The first segment is an integral extension of the metal plate.

3. The support according to claim 1, wherein, The second paragraph is an integrated extension of the first paragraph.

4. The support according to claim 1, wherein, The second segment can move relative to the first segment.

5. The support according to any one of claims 1, 2, 3 or 4, wherein, The support includes the biasing component, which includes a spring.

6. An apparatus comprising: Metal sheet; A printed circuit board (PCB) for coupling to the metal plate; A first support is located between the metal plate and a first position on the PCB, and the first support is used to apply a first stress at the first position; as well as The second support is located between the metal plate and the second position on the PCB. The second support is used to apply a second stress at the second position, wherein the first stress is less than the second stress.

7. The apparatus according to claim 6, wherein, The first support is aligned with a first opening in the PCB, and the second support is aligned with a second opening in the PCB. The first opening is within a threshold distance of a first solder for mounting a first component to the PCB, and the second opening is further away from a second solder for mounting a second component to the PCB than the threshold distance. The threshold distance corresponds to the difference between (a) the first radius of the outermost periphery of the corresponding support in the first and second supports and (b) the second radius of the periphery of the corresponding opening in the first and second openings.

8. The apparatus according to claim 6, wherein, The first support includes a biasing component that enables the first support to vary in length between a first end and a second end, the first end being adjacent to the metal plate and the second end being adjacent to the PCB.

9. The apparatus according to claim 8, wherein, The first support includes: A first spacer extends from a first surface of the metal plate toward a second surface of the PCB; and A second spacer extends from the second surface toward the first surface and is movable relative to the first spacer. The biasing component is used to cause the second spacer to move relative to the first spacer toward the PCB.

10. The apparatus according to claim 8, wherein, The first support includes a spacer extending a first portion of the length of the first support, and the biasing component extends a second portion of the length of the support, the first portion being different from the second portion.

11. The apparatus according to any one of claims 6, 7, 8, 9 or 10, wherein, The first support includes a first shape at the first end and a second shape at the second end, the first end being adjacent to a first surface of the metal plate and the second end being adjacent to a second surface of the PCB. The second shape is smaller than the first shape such that when the solder on the second surface of the PCB is within a threshold distance in a direction normal to the outer surface of the first support, the second end is spaced apart from the solder by the threshold distance.

12. The apparatus according to claim 11, wherein, The first shape is defined by a first cylinder, and the second shape is defined by a second cylinder having cavities on its side for positioning toward the solder, and a holding portion of the second cylinder for positioning away from the solder.

13. The apparatus according to claim 12, wherein, The first support includes a through hole extending along the axial length of the first support, and the cavity is spaced apart from the through hole.

14. The apparatus according to claim 12, wherein, The first support includes a through hole extending along the axial length of the first support, and the cavity intersects with the through hole.

15. The apparatus according to claim 12, wherein, The cavity is defined by a planar surface that is angled relative to the axial length of the first support.

16. The apparatus according to claim 12, wherein, The cavity is defined by a planar surface that is approximately parallel to the axial length of the first support.

17. The apparatus according to claim 11, wherein, The first shape is defined by a first cylinder, and the second shape is defined by a second cylinder having at least one of a chamfered edge or a rounded edge.

18. An apparatus comprising: Metal sheet; A printed circuit board (PCB) for coupling to the metal plate via threaded fasteners extending through the metal plate and through the PCB; A first support is used to surround the first threaded fastener in the threaded fastener, and the first support is used to separate the metal plate and the PCB; as well as A second support is used to surround a second threaded fastener in the threaded fastener, the second support is used to separate the metal plate and the PCB, and the first support satisfies at least one of the following: has a different shape than the second support; or comprises a different material than the second support.

19. The apparatus according to claim 18, wherein, The second support has a fixed length, and the first support has an adjustable length.

20. The apparatus according to claim 18, wherein, The first support includes a shape memory alloy (SMA), and the SMA is used to switch the first support between a first length and a second length based on whether the temperature of the first support meets a temperature threshold, wherein the second length is greater than the first length and greater than a third length of the second support, and the third length is less than or equal to the first length.

21. The apparatus according to any one of claims 18, 19 or 20, wherein, The PCB has components mounted to the PCB, and the first support includes at least one of the following: (a) a biasing component; or (b) a spacer extending from the metal plate toward the PCB, the spacer having a first shape at a first end and a second shape at a second end, the first end being adjacent to a first surface of the metal plate and the second end being adjacent to a second surface of the PCB, the second shape being smaller than the first shape such that when the first end is within a threshold distance of the component, the second end is at least at the threshold distance from the component.

22. The apparatus according to claim 21, wherein, The spacer is cylindrical in shape and has a cavity on the side of the spacer facing the component, the cavity extending between (a) the PCB and (b) the point between the metal plate and the PCB, such that the retained portion of the component and the second end of the spacer is at least at the threshold distance, the retained portion having rounded edges.

23. The apparatus according to claim 21, wherein, The spacer is cylindrical in shape, extends from the metal plate to the PCB, and has a chamfered edge at the second end such that the component is at least a threshold distance away from the inner periphery of the chamfered edge.

24. The apparatus according to claim 21, wherein, The spacer is cylindrical in shape and has a cavity on the side of the spacer facing the assembly, the cavity being defined by a planar surface angled relative to the axial length of the spacer.

25. The apparatus according to claim 21, wherein, The spacer is cylindrical in shape and has a cavity on the side of the spacer facing the assembly, the cavity being defined by a planar surface approximately parallel to the axial length of the spacer.

Images