Thermal Management of Computer Hardware Modules

Abstract

Aspects of the present disclosure include a structure for thermal management of a pluggable hardware module, an optical transceiver, and a method of cooling a pluggable hardware module. One embodiment of the structure can include a socket assembly configured to receive a hardware module. The socket assembly can include an integrated fluid reservoir that houses a thermal interface material (TIM). The socket assembly may be further configured to define a void when the hardware module is plugged into the socket assembly. The structure may further include at least one dispense port in fluid communication with the integrated fluid reservoir and the void. The at least one dispense port may be configured to automatically dispense the TIM from the integrated fluid reservoir into the void when the hardware module is plugged into the socket assembly.

Description

Technical Field

[0001] The present disclosure relates to thermal management, and more specifically, to thermal management of a pluggable hardware module via a self - contained thermal interface material.

[0002] The development of the EDVAC system in 1948 is often cited as the beginning of the computer age. Since then, computer systems have evolved into very complex devices. Today's computer systems typically include a combination of sophisticated hardware components and software components, application programs, operating systems, processors, buses, memories, input / output devices, and the like. As advancements in semiconductor processing and computer architecture have steadily increased performance, more advanced computer software has evolved to take advantage of their relatively high performance capabilities, and as a result, today's computer systems are much more powerful than just a few years ago.

[0003] Optical transceivers are widely used in computer systems, particularly in data centers, high - performance computing (HPC), and other systems that utilize high - performance computing networking. One common form factor of an optical transceiver is a printed circuit board (PCB) - mounted laser / optical module packaged in a molded metal housing. This module assembly can be plugged into a socket of a host computing system that is provided as a data center server or integrated into an HPC system.

Summary of the Invention

[0004] According to an embodiment of the present disclosure, a structure for thermal management of a pluggable hardware module includes a socket assembly configured to receive the hardware module. The socket assembly can include an integrated fluid reservoir that houses a thermal interface material (TIM). The socket assembly may be further configured to define a void when the hardware module is plug-connected to the socket assembly. The structure may further include at least one dispense port in fluid communication with the integrated fluid reservoir and the void. The at least one dispense port may be configured to automatically distribute TIM from the integrated fluid reservoir to the void when the hardware module is plug-connected to the socket assembly.

[0005] According to an embodiment of the present disclosure, an optical transceiver includes at least one optical module packaged within a housing. The housing may include a generally rectangular upper surface that defines a pattern configured to enhance the capillary action of a thermal interface material (TIM) applied thereon. The upper surface may include grooves. The grooves may be configured to receive a sealing member when the housing is inserted into a socket.

[0006] According to an embodiment of the present disclosure, a method for cooling a pluggable hardware module includes providing a socket assembly. The socket assembly can include an integrated fluid reservoir that houses a thermal interface material (TIM). The method can further include providing a hardware module. The hardware module and the socket assembly may be configured such that a void is defined when the hardware module is plugged into the socket assembly. The method may further include providing at least one dispense port in fluid communication with the integrated fluid reservoir and the void. The method may further include plugging the hardware module into the socket, and at least one dispense port is configured to automatically distribute the TIM from the integrated fluid reservoir into the void when the hardware module is plugged into the socket assembly.

[0007] The above summary is not intended to describe every embodiment or all implementations of the present disclosure that are shown.

[0008] Next, by way of mere example, preferred embodiments of the present invention will be described with reference to the following drawings.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

DETAILED DESCRIPTION OF THE INVENTION

[0010] The present invention is capable of accepting various modifications and alternative forms, and its details are shown by way of example in the drawings and will be described in detail. However, of course, the present invention is not limited to the specific embodiments described. On the contrary, the intention is to cover all modifications, equivalents, and alternatives within the scope of the present invention.

[0011] Aspects of the present disclosure relate to thermal management, and more particular aspects relate to thermal management of a pluggable hardware module via a self - contained thermal interface material. The present disclosure is not necessarily limited to such applications, and various aspects of the present disclosure will be understood by discussing various examples using this context.

[0012] System designers have characterized computer hardware including pluggable hardware modules by higher density designs and further speed increases. However, such speed increases generally involve significant heat generation, and the heat dissipation may become difficult due to the higher density.

[0013] Conventionally, socket assemblies have included passive heat sinks, but the performance of the heat sinks has been limited by the dry contact interface between the module and the socket. That is, because the thermal impedance of air is high, the practical ability to cool pluggable hardware modules has been severely limited. In addition, in many pluggable hardware modules and socket assemblies, the use of thermal pads is not practical (even if pre-applied), because the pads can rub, tear, or become damaged and accumulate at the interface, making it difficult to engage the module and socket. Accordingly, one aspect of the present disclosure is a method and system for automatically applying a thermal interface material (TIM) to dissipate heat generated by pluggable hardware modules, such as optical transceivers, internal connections, etc. ("modules"). These methods and systems are particularly desirable in up-to-date, high-density, thermally demanding computing systems used in data centers and HPC.

[0014] In some embodiments, a pluggable hardware module that requires additional cooling can be configured to be inserted into a socket assembly. The socket assembly can include a socket that is physically and thermally connected to a heat sink. The heat sink can define a cavity on its bottom surface. The heat sink can cooperate with the socket to define a TIM supply reservoir when tightened together by one or more spring retaining clips. The TIM supply reservoir can be filled with a liquid or semi-solid TIM at the manufacturing site or at the location where it is used.

[0015] The TIM in the TIM supply reservoir may include a liquid material (e.g., polyalphaolefin (PAO) oil, etc.), a semi-solid material (e.g., a low-viscosity / fluid thermal grease containing suspended submicron or nano silver or ceramic fillers, etc.), a liquid metal (e.g., gallium, GaInSn (galinstan), or GaIn), or other heat-conductive heat transfer media. Embodiments using PAO oil can have this TIM with a relatively high viscosity, are approved by many manufacturers for use as a TIM, have a relatively high thermal conductivity of about 0.18 - 0.2 W / mK, can form a relatively thin TIM layer (e.g., between about 0.025 - 0.05 mm), and can be cleaned with cloth or isopropyl alcohol (IPA) as needed, so it may be particularly desirable for some applications. Embodiments using liquid metal may be desirable for some applications because these TIMs generally have relatively high thermal conductivity (e.g., 40 W / mK) and surface tension characteristics.

[0016] The TIM supply reservoir can be characterized by at least one (in some embodiments, a plurality of) dispense ports. The dispense ports can be sized such that the surface tension of the TIM balances gravity before plugging in the hardware module. In this way, the TIM (including embodiments of liquid TIM) can remain in the dispense port or the TIM supply reservoir or both before inserting the hardware module into the socket assembly.

[0017] During operation, after plugging the module into the socket assembly, a void may be formed between the top / cover surface of the hardware module and the inner surface of the socket / socket assembly. The void can be bounded at its periphery by a gasket / sealing member for holding the TIM within the void (e.g., forming a temporary reservoir for the TIM). In some embodiments, the TIM can flow automatically from the TIM supply reservoir into the void and form a TIM layer in-situ. In some embodiments, the walls of the void (e.g., the top / cover surface of the hardware module or the inner surface of the socket / socket assembly or both) may be roughened or patterned or both to facilitate / enhance the capillary flow of the TIM into and through the void. In this way, the roughened surface can facilitate wetting, i.e., filling the void with the TIM, after the plugging / re-plugging of the hardware module. Additionally, or alternatively, a positive pressure can be applied (e.g., by a heat sink retaining spring or via a plunger operating as part of the insertion operation or both) to further facilitate the flow of the TIM into the void.

[0018] In some embodiments, the TIM supply reservoir can be sized such that its volume is significantly greater than the dispense volume. For example, the reservoir volume to dispense volume ratio may be greater than about 10 times, 25 times, or 50 times for a particular optical transceiver embodiment. In this way, in these embodiments, multiple (e.g., 10 times, 25 times, or 50 times) plugging / removing / re-plugging cycles of the optical transceiver can be enabled. Additionally, in some embodiments, the TIM supply reservoir may be refillable if further cycles are desired. Further, in some embodiments, one or more check valves can be included to control the dispense of the TIM during multiple plugging / re-plugging cycles, or to prevent leakage of the TIM, or both.

[0019] Accordingly, one aspect of the present disclosure is a socket assembly for, e.g., an optoelectronics hardware module, which can store TIM in an integrated reservoir. The TIM from the integrated reservoir can be automatically applied in-situ when the hardware module is plugged in, forming a TIM layer that can enhance the thermal interface between the hardware module and the heat sink. In some embodiments, further, by including a reservoir of sufficient volume to store or dispense or both dispense and store TIM during each plugging / re-plugging event, as well as a dispense port and a fill port, it may be possible to repeat the plugging / re-plugging cycle without significant damage or degradation to the performance of this enhanced thermal interface. Another aspect of the present disclosure is a structure that enables in-situ dispensing of TIM by an auxiliary positive pressure applied by a heat sink retaining clip / spring.

[0020] Other features and advantages of some embodiments can include an optimally designed heat sink wall thickness, reservoir volume, dispense port and fill port diameters, and a check valve to achieve the positive pressure for dispensing TIM for plugging / re-plugging. In some embodiments, it can also include a surface roughness to enhance wettability, and a peripheral groove / gasket along at least one edge to prevent overflow of the TIM.

[0021] Figure 1 shows an embodiment of data processing systems (DPS) 100a, 100b (generally referred to herein as DPS 100) according to some embodiments. Figure 1 only depicts the representative and main components of DPS 100, and such individual components can be much more complex than those shown in Figure 1. In some embodiments, DPS 100 may be implemented as a personal computer, a server computer, a portable computer such as a laptop or notebook computer, a personal data assistant (PDA), a tablet computer, or a smart phone, an automobile, an aircraft, a processor incorporated in a large device such as a teleconference system, an appliance, a smart device, or any other suitable type of electronic device. Moreover, there may be components other than or additional to those shown in Figure 1, and the number, type, and configuration of such components may vary.

[0022] The DPS 100 of Figure 1 may include a plurality of processing units 110a - 110d (generally, processor 110 or CPU 110), and these processing units can be connected by a system bus 122 to a main memory 112, a mass storage interface 114, a terminal / display interface 116, a network interface 118, and an input / output ("I / O") interface 120. The mass storage interface 114 of this embodiment connects the system bus 122 to one or more mass storage devices such as a direct access (or mass) storage device 140, a USB drive 141, or a read / write optical disk drive 142, or a combination thereof. Through the network interface 118, DPS 100a can communicate with other DPS 100b via a network 106. The main memory 112 can include an operating system 124, a plurality of application programs 126, and program data 128.

[0023] The embodiment of the DPS100 in FIG. 1 may be a general-purpose computing device. In these embodiments, the processor 110 may be any device capable of executing program instructions stored in the main memory 112, and may itself be constructed from one or more microprocessors or integrated circuits or both. In some embodiments, the DPS100 may include multiple processors or processing cores or both, as typically found in large and more capable computer systems, but in other embodiments, the DPS100 may only include a single processor system, or a single processor designed to emulate a multiprocessor system, or both. Further, the processor 110 may be implemented using a number of heterogeneous data processing systems where the main processor 110 exists on a single chip with a secondary processor. As another exemplary example, the processor 110 may be a symmetric multiprocessor system including multiple processors 110 of the same type.

[0024] When the DPS100 is powered on, the associated processor 110 can first execute the program instructions that make up the operating system 124. The operating system 124 can manage the physical and logical resources of the DPS100. These resources can include the main memory 112, the mass storage interface 114, the terminal / display interface 116, the network interface 118, and the system bus 122. Similar to the processor 110, some embodiments of the DPS100 can utilize multiple system interfaces 114, 116, 118, 120, and buses 122, which can each include their own separate fully programmed microprocessors.

[0025] Instructions for the operating system 124 or the application program 126 or both (generally referred to as "program code, computer-usable program code", or "computer-readable program code") may initially be located in the mass storage device 140 that communicates with the processor 110 through the system bus 122. The program code in different embodiments may be embodied on different physical or tangible computer-readable media such as the memory 112 or the mass storage device 140. In the exemplary example of FIG. 1, the instructions may be stored in a functional form of permanent storage on the direct access storage device 140. These instructions may then be loaded into the main memory 112 for execution by the processor 110. However, in some embodiments, the program code may alternatively be arranged in a functional form on a computer-readable medium such as a removable direct access storage device 140 or a read / write optical disk drive 142. Such instructions may be loaded into the DPS 100 or transferred to the DPS 100 for execution by the processor 110.

[0026] Continuing to refer to FIG. 1, the system bus 122 may be any device that facilitates communication between the processor 110, the main memory 112, and the interfaces 114, 116, 118, 120 and among them. Moreover, the system bus 122 in this embodiment is a relatively simple single bus structure that provides a direct communication path between the system buses 122, but is not limited to, point-to-point links in a hierarchical structure, star structure, or web structure, multiple hierarchical buses, parallel paths, and redundant paths, etc. Other bus structures are also consistent with the present disclosure.

[0027] The main memory 112 and the mass storage device 140 can cooperate to store the operating system 124, the application program 126, and the program data 128. In some embodiments, the main memory 112 may be a random access semiconductor memory device (“RAM”) capable of storing data and program instructions. Although FIG. 1 conceptually depicts the main memory 112 as a single monolithic entity, the main memory 112 in some embodiments may have a more complex arrangement, such as a hierarchical structure of caches and other memory devices. For example, the main memory 112 may exist at multiple levels of cache, and these caches may be further divided by functionality such that one cache holds instructions while another cache holds non-instruction data used by the processor 110. The main memory 112 may be further distributed and associated with different processors 110 or sets of processors 110, as is known in any of various so-called NUMA (nonuniform memory access) computer architectures. Additionally, some embodiments can utilize a virtual addressing mechanism that enables the DPS 100 to act as if it has access to a large single storage entity rather than to multiple small storage entities (such as the main memory 112 and the mass storage device 140).

[0028] In FIG. 1, the operating system 124, application program 126, and program data 128 are illustrated as being included within the main memory 112 of the DPS 100a. However, some or all of these may be physically located in different computer systems (e.g., DPS 100b). For example, in some embodiments, they may be accessed remotely via the network 106. Moreover, the operating system 124, application program 126, and program data 128 are not necessarily all completely included in the same physical DPS 100a at the same time and may even reside in the physical or virtual memory of another DPS 100b.

[0029] In some embodiments, system interfaces 114, 116, 118, 120 can support communication with various storage and I / O devices. The mass storage interface 114 can support attachment of one or more mass storage devices 140, which can include rotating magnetic disk drive storage devices, solid state storage devices (SSDs) that use integrated circuit assemblies as memory and typically use flash memory to persistently store data, or a combination of the two. Additionally, the mass storage device 140 can also include other devices and assemblies, an array of disk drives (commonly referred to as a RAID array) configured to appear as a single mass storage device to the host, or a hard disk drive, tape (e.g., mini DV), writable compact disk (e.g., CD-R and CD-RW), digital versatile disk (e.g., DVD, DVD-R, DVD+R, DVD+RW, DVD-RAM), holographic storage system, blue laser disk, archival storage media such as IBM(R) millipede devices, or a combination thereof. The I / O interface 120 can support attachment of one or more I / O devices such as a keyboard, mouse, modem, or printer (not shown).

[0030] The terminal / display interface 116 can be used to directly connect one or more displays 180 to the DPS 100. These displays 180 may be non-intelligent (i.e., dumb) terminals such as light-emitting diode (LED) monitors, or they may be fully programmable workstations themselves so that IT administrators and users can communicate with the DPS 100. However, note that while the display interface 116 can be provided to support communication with one or more displays 180, all necessary interactions with users and other processes can be done via the network 106, so the DPS 100 does not necessarily require a display 180.

[0031] Network 106 can be any suitable network or combination of networks and can support any suitable protocol suitable for communication of data or code or both between multiple DPSs 100. Thus, network interface 118 can be any device that facilitates such communication regardless of whether the network connection is made using current analog or digital or both technologies or via some future network mechanism. Suitable networks 106 include, but are not limited to, networks implemented using one or more of "InfiniBand" (IB) or IEEE (Institute of Electrical and Electronics Engineers) 802.3x "Ethernet(R)" specifications, cellular transmission networks, IEEE 802.11x, IEEE 802.16, General Packet Radio Service ("GPRS"), FRS (Family Radio Service), or a wireless network implementing one of the Bluetooth specifications, ultra-wideband ("UWB") technology as described in FCC02-48, etc. Those skilled in the art will understand that many different networks and transmission protocols can be used to implement network 106. The Transmission Control Protocol / Internet Protocol ("TCP / IP") suite includes suitable networks and transmission protocols.

[0032] FIG. 2(A) is a rear cross-sectional view of a first hardware assembly 200 suitable for use as one of system interfaces 114-120 according to some embodiments, and FIG. 2(B) is a side cross-sectional view thereof. The hardware assembly 200 can include a hardware module 210 (e.g., an optical transceiver) and a socket assembly 220.

[0033] The hardware module 210 can include a plurality of optical modules (e.g., transceivers) 212 that are positioned and adapted to automatically mate with corresponding receptacles 222 in the socket assembly 220 when the hardware module 210 is plug-connected through an opening 223 in the outer wall 224 of, for example, the DPS 100 (or another housing or rack including the hardware module 210). In some embodiments, the hardware module 210 can include a molded housing 211 having a generally rectangular upper surface 213. A gasket / sealing member 216 can extend completely or partially along the periphery of the upper surface 212 and can be held by a corresponding retaining groove 214. Alternatively, the groove 214 may be configured to receive a gasket / sealing member (not shown) attached to the socket assembly 220. The hardware module 210 may further include a printed circuit board 217 on which the optical module 212 is mounted and a communication cable 218.

[0034] The socket assembly 220 can include a socket 222 and a heat sink 230. The heat sink 230 includes a base 232 that is physically and thermally connected to a plurality of cooling fins 234. The base 232 of the heat sink 230 can define a cavity such that when the heat sink 230 is clamped onto the socket 221 by a retaining clip 250 and biased against the socket 221, the heat sink 230 and the socket 221 cooperate to define a TIM supply reservoir 240. The TIM supply reservoir 240 can be filled with a liquid or quasiliquid thermal interface material (TlM) 242 through an upper opening 227.

[0035] The base 232 of the heat sink 230 can also define at least one dispense port 245. In some embodiments, the dispense port 245 can be sized such that the surface tension of the TIM 242 is balanced with gravity and / or fluid pressure. In these embodiments, the TIM 242 tends to remain within the TIM supply reservoir 240 before the hardware module 210 is plugged into the socket assembly 220. In contrast, the upper opening 227 can have a relatively larger diameter than the dispense port 245 in some embodiments to facilitate filling of the TIM supply reservoir 240.

[0036] During operation, the fluid reservoir 240 may be sealed using a temporary plug or tape (not shown) to prevent leakage during transportation. To begin installation, the computer administrator first removes these plugs / tape and inserts the hardware module 210 through the opening 223 in the outer wall 224 of the DPS 100 into the socket assembly 220. When the hardware module 210 is fully inserted, the upper surface of the hardware module 210, the inner surface of the socket 222, and the gasket / sealing member 216 can cooperate to define a void 265 (e.g., between the electronic hardware module to be cooled and the heat sink). In some embodiments, this void 265 can be made relatively thin / shallow such that the surface tension holding the TIM 242 within the TIM supply reservoir 240 is broken by the plugging / re-plugging of the hardware module 210, allowing the TIM 242 to flow into the void 265 by gravity and / or capillary action. The gasket / sealing member 216 can hold the TIM 242 within the void 265 while the hardware module 210 is plugged into the socket assembly 220.

[0037] In these embodiments, the inner surface of the socket 222 and the upper surface of the hardware module 210 may be roughened, patterned, or both to enhance capillary flow, thereby assisting the TIM 242 in completely filling the void 265. Additionally, or alternatively, the heat generated by the hardware module 210 during operation of the hardware module 210 may heat the TIM 242 and reduce its viscosity. This change in viscosity may further enable the natural flow of the TIM 242 into the void 265. Advantageously, the TIM 242 functions as a "wet" contact interface and can increase the heat transfer from the hardware module 210 through the base 232 to the cooling fins 234 compared to a "dry" contact interface.

[0038] The diameter of the dispense port 245 can be optimized for capillary retention and dispensing. The liquid TIM may flow out of the dispense port 245, for example, when the force due to the pressure (Fp) applied to the TIM from the retaining clip + the force due to gravity (Fg) > the capillary force (Fc). In other words, the TIM may flow out of the dispense port 245 when Fp exceeds the value obtained by subtracting Fg from FC: where Fg = m * g = ρ * volume * g, and assuming the dispense port is cylindrical, Fg = ρ * r 2 * height: In an exemplary embodiment, this is as follows: 800 kg / m 3 * 3.14 * 0.001 m * 0.001 m * 0.0015 m * 9.8 m / s 2 = 3.6 * 10 -5 N Fc can be Fc = 2 * π * r * σ * cos(θ), where σ is the surface tension of the selected TIM material and the angle θ is the contact angle between the dispense port 245 and the TIM. Continuing with the above example, when σ is 30 N / m and the angle is θ = 60 degrees, FC is 9.4 * 10 -2 N. The retaining clip (or other biasing structure) is (9.4 * 10-2 N - 3.6×10 -5 ) ≒ 9.4×10 -2 It can be designed to apply a force greater than N.

[0039] Figure 3(A) is a rear cross-sectional view of a second hardware assembly 300 suitable for use as one of the system interfaces 114 - 120 according to some embodiments, and Figure 3(B) is a side cross-sectional view thereof. The hardware assembly 300 is similar to the hardware assembly 200 of Figures 2(A) - 2(B). However, in the hardware assembly 300, the dispense port 345 can include a valve 375 that automatically actuates from a normally closed position to an open position when the hardware module 310 is plugged into the socket 320. By opening the valve 375, the TIM 342 can flow into the void 365. In this embodiment, the upper surface 370 of the primary fluid reservoir 340 may be relatively thin and flexible. The retaining clip 350 can apply a biasing force to the flexible upper surface 370 of the heat sink 330 and deflect it downwardly towards the primary fluid reservoir 340. That deflection, in turn, can pressurize the TIM 342 and force the TIM 342 into the secondary fluid reservoir 365.

[0040] Advantageously, the volume of the TIM 342 dispensed by this second hardware assembly 300 can be controlled such that the primary reservoir 340 can support repeated plugging / unplugging cycles. As an illustrative example, aluminum has a Young's modulus of 70 Gpa. This means that for an upper surface 370 made of aluminum with a thickness of 0.4 mm, the upper surface 370 can be displaced 0.12 mm with a force of 10 lbf (4.5 kgf) by the retaining clip 350. If the volume of the primary fluid reservoir 340 is 1 cm × 2 cm × 0.5 cm = 1.0 cm 3 then each dispense can include 2.4×10 -2 cm 3 of TIM 342, which is approximately 2.4% of the primary fluid reservoir 340.

[0041] FIG. 4(A) is a rear cross-sectional view of a third hardware assembly 400 suitable for use as one of system interfaces 114-120 according to some embodiments, and FIG. 4(B) is a side cross-sectional view thereof. The hardware assembly 400 is similar to the hardware assembly 300 of FIGS. 3(A)-3(B). However, in the hardware assembly 400, the primary fluid reservoir 440 may include a movable plug 480. The movable plug 480 may be biased within the primary fluid reservoir 440 by a retaining clip 450. This third hardware assembly 400 may be desirable in some applications because the movable plug 480 can allow for a greater biasing operation than in the case of the flexible upper surface 370 of the second hardware assembly 300. This feature allows for a greater amount of TIM 442 to be dispensed into the secondary fluid reservoir 465 in each plug connection / re-plug connection cycle.

[0042] Figures 5(A) - 5(D) are cross - sectional views of a first embodiment of the upper opening 527 of the primary fluid reservoir 540 according to some embodiments. In Figures 5(A) - 5(D), the check valve 585 is inserted into the fluid column 590 filled with TIM542. During operation, Figure 5(A) shows the fluid column 590 at height H1 before a hardware module (not shown) is first plug - connected to a socket (not shown). As described above, with this plug - connection, a portion of the TIM542 can migrate to a secondary fluid reservoir (not shown). Figure 5(B) shows the fluid column 590 at height H2 after the hardware module 510 is plug - connected to the socket 520. As shown, since the check valve 585 allows a portion of the TIM542 material in the fluid column 590 to refill the primary fluid reservoir 540, H2 < H1. Figure 5(C) shows the fluid column 590 at height H3 after the hardware module is first removed (see previous figure). As shown, due to surface tension at the dispense port 245 (see Figures 2(A) - 2(B)) or the valve 375 (see Figures 3(A) - 3(B)) or both, the remaining TIM542 can be held within the primary fluid reservoir 540, so H3 = H2. Figure 5(D) shows the fluid column 590 at height H4 after the hardware module 510 is re - plug - connected to the socket 520. As shown, again, since the check valve 585 allows a portion of the TIM542 in the fluid column 590 to refill the primary fluid reservoir 540, H4 < H3. Advantageously, however, the check valve 585 does not allow the TIM542 to flow in the opposite direction, i.e., from the primary fluid reservoir 540 into the fluid column 590. This may be desirable in embodiments where the TIM542 is actively biased towards the secondary fluid reservoir.

[0043] Figures 6(A) - 6(D) are cross-sectional views of a second embodiment of the upper opening 627 of the primary fluid reservoir 620 according to some embodiments. This embodiment is similar to that shown in Figures 5(A) - 5(D). However, in Figures 6(A) - 6(B), the check valve 685 is positioned above the level of the TIM642 and is configured to receive air 695. During operation, due to the plugging / re-plugging of the hardware module (see previous figures), the TIM642 flows into a secondary fluid reservoir (not shown) such that H4 < H3, H3 = H2, and H2 < H1. The amount of air 695 within the fluid column 690 can increase to compensate for the reduced volume of the TIM642 within the primary fluid reservoir 620.

[0044] Figures 7(A) - 7(D) are cross-sectional views of a third embodiment of the upper opening 727 of the primary fluid reservoir 720 according to some embodiments. This embodiment is similar to that shown in Figures 6(A) - 6(D). However, in Figures 7(A) - 7(B), the fluid column 790 is routed near the outer wall 724 of the DPS100 and the opening 723. During operation, due to the repeated insertion and removal of a hardware module (not shown), the TIM742 flows into a secondary fluid reservoir (not shown) such that H4 < H3, H3 = H2, and H2 < H1. The check valve 785 allows the amount of air 795 within the fluid column 790 to increase to compensate for the decreased level of the TIM742. Advantageously, the upper opening 727 of Figures 7(A) - 7(D) is accessible from the outside of the DPS100, thereby optionally allowing the TIM742 within the primary fluid reservoir 720 to be refilled via, for example, a "Zerk" grease fitting (not shown).

[0045] Figure 8 is Table 800 showing a simulation of the thermal analysis of a conventional dry contact and the embodiments disclosed herein. As can be seen, for current-generation optical transceivers, some embodiments of the present disclosure can significantly reduce the temperature rise during operation from +6 to 10 °C to +1.0 to 2.5 °C. Moreover, this table shows that the thermal limits of current dry contact technologies may hinder the future development of these devices, as the expected usage could lead to a temperature rise of +12 °C. In contrast, embodiments of the present disclosure can limit these predicted rises to only +2 to 5 °C.

[0046] Figures 9(A) and 9(B) are rear cross-sectional views of a second hardware assembly 300 during operation. More specifically, Figure 9(A) shows a well-aligned plug connection operation, and Figure 9(B) shows a misaligned plug connection operation. In the second hardware assembly 300, TIM 342 may be dispensed through a plurality of dispense ports 345 aligned laterally with respect to the insertion vector. The plurality of dispense ports 345 can help ensure the dispensing of TIM 342 even when the position of the hardware module 310 is misaligned during plug connection, as shown in Figure 9(B). Additionally, the roughness / pattern of the upper surface 312 of the hardware module 310 can capture the dispensed excess TIM 342 and prevent overflow beyond the desired thermal interface area.

[0047] Figure 10 is a flowchart showing a method 1000 for cooling hardware modules in DPS100 according to some embodiments. In operation 1010, the system administrator provides a hardware assembly such as hardware assembly 200, hardware assembly 300, or hardware assembly 400. The hardware assembly may include hardware modules such as hardware module 210, hardware module 310, or hardware module 410. The hardware assembly may also include sockets such as socket 220, socket 320, or socket 420. In operation 1020, the system administrator can plug the hardware module into the socket. The hardware module and the socket can cooperate to form a void. In operation 1030, TIM may flow from a reservoir associated with the socket into the void to form a TIM layer between the hardware module and the socket. In operation 1040, the system administrator enables the hardware module or the socket or both.

[0048] The descriptions of the various embodiments of the present disclosure are presented for purposes of illustration, but are not intended to be exhaustive or to limit the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope of the described embodiments. The terms used herein are selected to explain the principles of the embodiments, actual applications, or technical improvements to the technology found in the market, or to enable other skilled artisans to understand the embodiments disclosed herein.

Claims

**Claim 1** A structure for thermal management of a hardware module with pluggable attachment, a socket assembly configured to receive the hardware module, the socket assembly including an integrated fluid reservoir that houses a thermal interface material (TIM), the socket assembly further configured to define a void when the hardware module is plugged into the socket assembly, the socket assembly, and at least one dispense port in fluid communication with the integrated fluid reservoir and the void, the at least one dispense port configured to automatically dispense TIM from the integrated fluid reservoir into the void when the hardware module is plugged into the socket assembly, the at least one dispense port comprising a structure. **Claim 2** The structure of claim 1, wherein the hardware module includes optical internal connections. **Claim 3** The structure of claim 1, further comprising a sealing member that houses the TIM in the void. **Claim 4** The structure of claim 1, wherein the socket assembly includes a socket and a heat sink that defines a cavity, the socket and the heat sink cooperating to define the integrated fluid reservoir. **Claim 5** The structure of claim 4, wherein the socket assembly further includes a retaining clip configured to deflect a flexible surface of the heat sink into the cavity. **Claim 6** The structure of claim 4, wherein the socket assembly further includes a retaining clip configured to bias the plug into the cavity. **Claim 7** The structure of claim 1, wherein the TIM includes a liquid. **Claim 8** The structure of claim 7, wherein the liquid includes polyalphaolefin (PAO) oil. **Claim 9** The structure of claim 1, wherein the TIM includes thermal grease. **Claim 10** The structure of claim 1, wherein the TIM includes liquid metal. **Claim 11** The structure of claim 10, wherein the liquid metal includes gallium. **Claim 12** The structure of claim 10, wherein the liquid metal includes galinstan. **Claim 13** The structure according to claim 1, wherein the at least one dispense port is sized such that surface tension balances gravity until the hardware module is plugged into the socket assembly.

14. The structure according to claim 1, further comprising at least one upper opening in fluid communication with the integrated fluid reservoir, the upper opening being sized larger than the at least one dispense port.

15. The structure according to claim 13, wherein the hardware module includes an upper surface having a groove along at least one edge.

16. The structure according to claim 15, wherein the groove is configured to receive a sealing member, and when the hardware module is plugged into the socket assembly, the sealing member holds the TIM in the void.

17. The structure according to claim 15, wherein the upper surface is roughened.

18. The structure according to claim 15, wherein the upper surface is patterned.

19. An optical transceiver, comprising at least one optical module packaged within a housing, wherein the housing has a generally rectangular upper surface defining a pattern configured to enhance capillary action of a thermal interface material (TIM) applied thereon, and the generally rectangular upper surface, and a groove in the upper surface, the groove being configured to receive a sealing member when the housing is inserted into a socket, the at least one optical module including An optical transceiver comprising.

20. A method for cooling a pluggable hardware module, providing a socket assembly, the socket assembly including an integrated fluid reservoir for receiving a thermal interface material (TIM), providing a hardware module, the hardware module and the socket assembly being configured to define a void when the hardware module is plugged into the socket assembly, providing at least one dispense port in fluid communication with the integrated fluid reservoir and the void, Plugging the hardware module into the socket, wherein the at least one dispense port is configured to automatically dispense TIM from the integrated fluid reservoir into the void when the hardware module is plugged into the socket assembly, and plugging the hardware module into the socket A method comprising.

Images