Liquid immersion cooling system

Abstract

An immersion cooling system is disclosed. An example immersion cooling system includes an immersion tank including a cooling fluid and a reservoir containing a recirculation portion of the cooling fluid. The reservoir is separated from the immersion tank by a height to create a liquid level change between the cooling fluid in the immersion tank and the cooling fluid in the reservoir. The supply conduit is to fluidly couple the immersion tank and the reservoir. The cooling fluid is provided from the reservoir to the immersion tank via gravity.

Description

[0001] Related Applications This patent is a continuation of International Patent Application No. PCT / CN2024 / 120940, filed September 25, 2024. Priority is hereby claimed to International Patent Application PCT / CN2024 / 120940. International Patent Application PCT / CN2024 / 120940 is hereby incorporated by reference in its entirety. BACKGROUND

[0002] In the context of high performance computing (HPC), liquid cooling has steadily outpaced the use of air cooling, as liquid cooling can provide higher component density, waste heat recovery capability, and / or lower operating costs. One form of liquid cooling (immersion cooling) includes submerging computing components in a coolant or dielectric liquid. BRIEF DESCRIPTION OF DRAWINGS

[0003] Figure 1 is a schematic diagram of an example cooling infrastructure having an immersion cooling system in accordance with the teachings of this disclosure.

[0004] Figure 2 is an enlarged portion of the example immersion cooling system of Figure 1

[0005] Figure 3A is a side view of an example immersion tank of the example immersion cooling system of Figure 2

[0006] Figure 3B is a side view of an example electronic component of the example immersion tank of Figure 3A

[0007] Figure 3C is a perspective view of the example electronic component of Figure 3B

[0008] Figure 3D is a perspective cross-sectional view of the example electronic component of Figure 3C

[0009] Figure 4 is a schematic diagram of another example cooling infrastructure having another example immersion cooling system disclosed herein.

[0010] Figure 5 One or more example environments in which the teachings of this disclosure can be implemented are shown.

[0011] Figure 6 At least one example of a data center for executing workloads having dispersed resources is shown.

[0012] Figure 7 At least one example of a data center for executing workloads having dispersed resources is shown. Figure 6 ​​​​​at least one example of a pod in a data center of the company.

[0013] Generally, like reference numbers will be used throughout the drawings and accompanying written description to refer to like or similar portions (delivered to the same or a similar function) of the embodiments. The drawings are not necessarily to scale. Instead, the thickness of layers or regions can be exaggerated in the drawings for clarity. Although the drawings depict layers and regions as straight lines and / or boundaries, some or all of the lines and / or boundaries can be idealized. In reality, the boundaries and / or lines can be irregular, mixed, and / or blurred. DETAILED DESCRIPTION

[0014] The power consumption of various types of processor units (e.g., XPU (e.g., central processing unit (CPU), graphics processing unit (GPU)), expansion cards, and memory has increased generation over generation. However, as the performance of semiconductor chips (e.g., computing / networking systems into which the chips are integrated) continues to increase, the associated power consumption is also increasing. In some instances, the performance of many semiconductor chips and computing / networking system products has exceeded the threshold at which traditional air cooling cannot keep the chips / devices sufficiently cooled. Due to its high heat capture rate, as well as low power usage efficiency (PUE), high component reliability in corrosive atmospheric conditions, and / or implementation of a modular and scalable design, liquid immersion cooling is becoming an attractive option for cooling high performance computing systems (e.g., data center servers and / or computing / networking system products). The advent of edge computing and 5G cellular network technology is further accelerating the adoption of liquid immersion cooling.

[0015] Compared to more traditional air cooling systems, liquid cooling for electronic components has been developed to address the increased thermal management risks arising from the increased thermal design capabilities in high-performance systems such as CPU and / or GPU servers, accelerators, AI computing, machine learning computing, cloud computing, edge computing, etc., in data centers. More specifically, liquids have inherent advantages over air in terms of higher specific heat (when boiling is not involved) and higher latent heat of vaporization (when boiling is involved). In some instances, liquids can be used for indirect cooling of electronic components by cooling a cold plate thermally coupled to (one or more) electronic components. An alternative approach is to directly immerse the electronic components in the coolant. In direct immersion cooling, the liquid can be in direct contact with the electronic components to directly draw heat away from them. To enable direct contact between the coolant and the electronic components, the coolant (e.g., a single-phase liquid) is electrically insulating (e.g., a dielectric liquid). As used herein, single-phase cooling (e.g., single-phase immersion cooling) means that the cooling fluid (sometimes referred to herein as coolant or coolant) used to cool electronic components draws heat away from the heat source (e.g., the electronic component) without changing phase (e.g., without boiling and turning into vapor). Such a cooling fluid is referred to herein as a single-phase cooling fluid, liquid, cooling agent, or coolant. Single-phase immersion cooling can utilize open-bath immersion tanks or closed-bath immersion tanks.

[0016] Single-phase immersion tanks have flow rate limitations. For example, the flow rate of single-phase coolant through heat exchangers of electronic components (e.g., XPU radiators) is provided by a fluid pump. With conventional tanks, the improvement in thermal performance or cooling efficiency is minimal if only the pump power is increased to increase the fluid flow rate of the single-phase solution in an immersion cooling system. That is, increasing the fluid flow rate at the chassis inlet panel (with higher pump power) has one or more minimal thermal benefits. As a result, even when the chassis inlet flow rate is significantly increased, the fluid flow rate of coolant through the heat exchanger (e.g., XPU radiator) increases only slightly. For example, for a typical coolant in a single-phase immersion cooling system, when the chassis inlet flow rate increases by about 7.5 times, the fluid flow rate through the heat exchanger increases by about 1.3 times, and most of the increased flow rate (from the chassis inlet) bypasses low flow resistance areas in the chassis, essentially wasting the benefits of the increased flow rate.

[0017] The example systems, apparatuses, and associated methods disclosed herein improve the cooling capacity of cooling systems and / or associated cooling processes. The examples disclosed herein overcome tank height limitations to achieve higher potential energy, thereby providing a higher head to the radiator (e.g., an XPU radiator) to increase the flow rate through the radiator region. To improve cooling capacity, the example systems, apparatuses, and associated methods disclosed herein increase the flow rate of coolant through the heat exchanger (e.g., an XPU radiator) to significantly enhance single-phase immersion cooling capacity (e.g., between 900 W and 1000 W). To increase the flow rate, the example systems, apparatuses, and methods disclosed herein create a change in coolant level within the immersion tank and reservoir to increase the coolant head. Specifically, the reservoir is located above the height of the immersion tank. Therefore, the distance between the immersion tank and the reservoir is greater than the height of the immersion tank. In this way, the example systems disclosed herein employ a waterfall system, which distributes coolant from the reservoir to one or more immersion tanks via waterfall piping. For example, the systems, apparatus, and associated methods disclosed herein optimize working fluid recirculation and employ a waterfall system to obtain higher gravitational potential energy in the working fluid, thereby increasing the flow rate of the working fluid (e.g., coolant or liquid) from the reservoir to the immersion tank. Additional head and / or changes in liquid level significantly increase the flow rate of the working fluid through a radiator associated with one or more integrated circuit components (e.g., an XPU) immersed in the immersion tank or chamber, resulting in the removal of a greater amount of heat across the radiator. Higher flow rates are achievable with larger head. Using the increased head, the example cooling systems disclosed herein achieve higher flow rates to enhance the cooling capacity of single-phase immersion cooling. Therefore, the liquid level change is not limited to the height of the immersion tank. Instead, the liquid level change can be greater than the height of the immersion tank to increase the head and improve the flow rate by gravity-driven flow, thereby increasing the cooling capacity of the cooling system (e.g., providing a cooling capacity between approximately 900 and 1110 watts (W)). As used herein, “cooling capacity” measures the ability of a cooling system to remove heat.

[0018] As a result, the examples disclosed herein enhance cooling capacity because the higher potential energy is converted into greater kinetic energy with the higher fluid flow velocity across the heatsink. Because forced convection is dominated by higher fluid flow velocities, an increased convective heat transfer coefficient (HTC) is achieved. Specifically, the examples disclosed herein pump coolant directly into the reservoir to achieve high potential energy in the coolant, which then flows very efficiently through the heat exchanger (e.g., CPU / GPU heatsink area) at a higher flow rate and a much higher forced convection HTC, thereby significantly improving cooling capacity accordingly.

[0019] Figure 1This is a schematic diagram of an example computing environment 100 having an example immersion cooling system 102, based on the teachings of this disclosure. The computing environment 100 may be, for example, a data center or cloud service hub. The computing environment 100 includes an immersion cooling system 102, a first heat exchanger or cooler 104, and a second heat exchanger or cooling tower 106. A first loop system 108 (e.g., a first plurality of pipes, conduits, hoses, flexible pipes, rigid pipes, valves, etc.) fluidly couples the immersion cooling system 102 and the cooler 104. A second loop system 110 (e.g., a first plurality of pipes, conduits, hoses, flexible pipes, rigid pipes, valves, etc.) fluidly couples the cooler 104 and the cooling tower 106. The first loop system 108 of the illustrated example provides or recirculates a first working fluid 112 (e.g., coolant, cooling fluid, etc.), and the second loop system 110 of the illustrated example provides or recirculates a second working fluid 114 (e.g., coolant, water, cooling fluid, etc.). The first working fluid 112 may be a dielectric material, one or more hydrocarbon materials, synthetic oil, one or more hydrocarbon-based dielectric materials, and / or any one or more other materials. The second working fluid 114 may be water and / or any other coolant (single-phase liquid, two-phase liquid, etc.).

[0020] The first loop system 108 and the second loop system 110 shown in the example are closed-loop systems. Specifically, the first loop system 108 is independent of the second loop system 110. Therefore, the first loop system 108 is fluidly isolated from the second loop system 110, such that the first working fluid 112 does not mix with the second working fluid 114. The first pump 116 circulates the first working fluid 112 through the immersion cooling system 102 and the cooler 104 via the first loop system 108. The second pump 118 circulates the second working fluid 114 through the cooling tower 106 and the cooler 104 via the second loop system 110.

[0021] During operation, the immersion cooling system 102 cools the electronic components 120 (e.g., servers and / or other IT equipment) of the computing environment 100. The immersion cooling system 102 uses a first working fluid 112 to remove heat from the electronic components 120. For example, when the electronic components 120 generate heat, the first working fluid 112 of the immersion cooling system 102 removes the heat generated by the electronic components 120 via a first loop system 108. The warmed first working fluid 112 discharged from the immersion cooling system 102 is drawn away from the electronic components 120 and directed to a cooler 104 via a first pump 116 and the first loop system 108. The cooler 104 receives a second working fluid 114 (e.g., water) from a cooling tower 106 via a second pump 118 and a second loop system 110. The cooler 104 transfers heat from the warmed first working fluid 112 of the immersion cooling system 102 to the cooler second working fluid 114 received from the cooling tower 106. Heat removal from the first working fluid 112 received by the cooler 104 via the first loop system 108 causes the first working fluid 112 to cool, which is then routed or recirculated to the immersion cooling system 102 to cool the electronic components 120. The warmed second working fluid 114 exiting the second loop system 110 of the cooler 104 is routed or recirculated to a third heat exchanger or second cooler 122 associated with the cooling tower 106. The second cooler 122 removes heat from the second working fluid 114, which is then recirculated to the cooling tower 106 to cool the first working fluid 112. In some examples, the second cooler 122 may be omitted.

[0022] Figure 2 yes Figure 1A schematic diagram of an example immersion cooling system 102 is provided. The illustrated example immersion cooling system 102 includes multiple immersion tanks 202 (e.g., one or more chambers), a reservoir 204 (e.g., a chamber), a supply manifold 206, a return manifold 208, and multiple waterfall conduits 210. Each fluid in the immersion tank 202 is coupled to a corresponding waterfall conduit in the waterfall conduit 210. The immersion tanks 202 and the reservoirs 204 contain (e.g., at least partially filled) a first working fluid 112. Additionally, the immersion tanks 202 include electronic components 120. The electronic components 120 are immersed in the first working fluid 112 of the immersion tanks 202. In the illustrated example, the electronic components include multiple servers 212. In the illustrated example, each immersion tank 202 includes six servers 212. For example, the first immersion tank 202a includes a first server 212a, a second server 212b, a third server 212c, a fourth server 212d, a fifth server 212e, and a sixth server 212f. However, in other examples, each immersion tank 202 may include only one server 212, two servers 212, three servers 212, or any other number of servers 212. In some examples, one or more of the servers 212 may be the same or different.

[0023] The reservoir 204 in the example stored or contained a portion (e.g., a volume) of the first cooling fluid 112 (e.g., a recirculated portion or volume of cooling fluid 114). The reservoir 204 and / or supply manifold 206 were spaced apart from the immersion tank 202. Specifically, the reservoir 204 and / or supply manifold 206 were located above the immersion tank 202. For example, the reservoir 204 was located at a first elevation, while the immersion tank 202 was located at a second elevation. The first elevation was greater than the second elevation. In some examples, the reservoir 204 could be located on the second, third, or roof of a data center or building, while the immersion tank 202 could be located on the ground floor or in the basement of the data center or building. In other words, the reservoir 204 was located above the immersion tank 202 to establish a change in liquid level between the first working fluid 112 in the reservoir 204 and the first working fluid 112 in the immersion tank 202. This allows the first working fluid 112 to flow between the reservoir 204 and the immersion tank 202 by gravity. The establishment of a liquid level change enables the first working fluid 112 to be distributed and / or flowed between the reservoir 204 and the immersion tank 202 (e.g., the first immersion tank 202a) without the use of a pump, thereby improving the efficiency of the immersion cooling system 102. Furthermore, the establishment of a liquid level change increases and / or amplifies the head and / or pressure changes between the first working fluid 112 in the reservoir 204 and the immersion tank 202, resulting in an increased flow rate of the first working fluid 112 through the heat exchangers of the electronic components 120 and / or the server 212, thereby increasing the cooling capacity of the immersion cooling system 102.

[0024] To supply a first working fluid 112 from the reservoir 204 to the immersion tank 202, the immersion cooling system 102 includes a waterfall conduit 210. The waterfall conduit 210 moves the first working fluid 112 between the reservoir 204 and the immersion tank 202 by gravity. Thus, the waterfall conduit 210 in the illustrated example extends between the reservoir 204 and the immersion tank 202. Specifically, the waterfall conduit 210 in the illustrated example fluidly couples a supply manifold 206 to a server 212 located in the immersion tank 202. However, in some examples, the supply manifold 206 is not required, and the waterfall conduit 210 may be directly coupled to the reservoir 204. In the illustrated example, the waterfall conduit 210 extending between the reservoir 204 and one or more immersion tanks 202 is substantially vertical (e.g., vertical or within 10% of being completely vertical) to allow the first working fluid 112 to flow from the reservoir 204 to one or more immersion tanks 202 by gravity. In some examples, the waterfall duct 210 can be positioned at an angle relative to vertical (e.g., between approximately 10 and 60 degrees relative to vertical). (See also: [link to example]) Figures 3A-3D In more detail, each fluid in the waterfall duct 210 is coupled to a heat exchanger (e.g., radiator, cold plate, etc.) of the server.

[0025] The flow rate (e.g., maximum flow rate) of the first working fluid 112 from reservoir 204 to server 212 is based on the change in liquid level height provided by the distance (e.g., vertical distance) between reservoir 204 and immersion tank 202. To control the fluid flow between reservoir 204 and server 212 and / or the flow rate of the first working fluid 112 (e.g., less than the maximum flow rate), the immersion cooling system 102 of the illustrated example includes multiple flow control devices or control valves 214. In some examples, control valves 214 (e.g., ball valves, shut-off valves, throttling valves, etc.) can cut off the fluid flow between reservoir 204 and the respective immersion tanks in immersion tank 202 and / or the respective servers in server 212. In some examples, control valves 214 can be used to control the fluid flow rate of the first working fluid 112 between reservoir 204 and the respective immersion tanks in immersion tank 202 and / or the respective servers in server 212. For example, referring to the first immersion tank 202a, each server 212a-f includes a corresponding control valve among control valves 214a-f. Thus, the first immersion tank 202a of the illustrated example includes six control valves 214 (e.g., control valves for each of servers 212a-f). In this way, the flow rate of the first working fluid 112 to each of the servers 212a-f can be individually controlled. In some examples, the first server 212a may be supplied with a first working fluid 112 having a greater flow rate than the first working fluid 112 supplied to the second server 212b. In this way, when the first server 212a is operating at a higher power output than the power output of the second server 212b, the first server 212a may be supplied with the first working fluid 112 having a first flow rate, and the second server 212b may be supplied with the first working fluid 112 having a second flow rate (e.g., a reduced flow rate of the first working fluid). In some examples, a control valve 214g may be inserted into the waterfall conduit 210b. In this way, the first working fluid 112 can be centrally controlled to the servers 212 in the second immersion tank 202b. For example, the control valve 214g may be used to control the flow rate and / or fluid flow of the first working fluid 112 to all servers 212 located in the second immersion tank 202b. Therefore, in this example, each server in the servers 212 in the second immersion tank 202b receives a first working fluid 112 with the same flow rate. The control valve 214 can also be used to cut off (e.g., completely cut off) the fluid flow to one or more of the servers 212 and / or the immersion tank 202. For example, if the servers 212 in the immersion tank 202 and / or the servers 212 in the immersion tank 202 are not in operation, the flow of the first working fluid 112 to the servers 212 and / or the immersion tank 202 can be cut off. In some examples, the flow of the first working fluid 112 can be cut off via the control valve 214 to perform maintenance, replacement, and / or repair of one or more servers 212 in the immersion tank 202. In some examples, one or more of the control valves 214 can be omitted.

[0026] Figure 3A yes Figure 2 A schematic side view of the first immersion tank 202a of the example immersion cooling system 102. Figure 3B yes Figure 3A Example side view of the first server 212a in the first immersion tank 202a. Figure 3C yes Figure 3A and Figure 3B A perspective view of the first server 212a. Figure 3D yes Figures 3A-3C A perspective cross-sectional view of the first server 212a.

[0027] refer to Figure 3A The first immersion tank 202a is filled with a first working fluid 112 to a tank surface level 302. The reservoir 204 is filled with the first working fluid 112 to provide a reservoir surface level 304. Furthermore, the reservoir 204 is located above the first immersion tank 202a. As described above, the reservoir 204 is located at a first elevation, and the first immersion tank 202a is located at a second elevation lower than the first elevation. In this way, a height or distance 306 is provided between the reservoir surface level 304 and the tank surface level 302. As a result, the distance 306 between the reservoir surface level 304 and the tank surface level 302 creates a change in the liquid level height between the reservoir 204 and the first immersion tank 202a. ).

[0028] The first immersion tank 202a shown in the example includes an inner wall 308 and an outer wall 310. The inner surface of the inner wall 308 defines a first cavity 312 for receiving a first server 212a. The inner surface of the outer wall 310 and the outer surface of the inner wall 308 define a second cavity 314. The first cavity 312 is fluidly coupled to the second cavity 314 via a plurality of channels 316 (e.g., holes, slits, slots, openings, etc.) formed in the inner wall 308. The outlet 318 of the first immersion tank 202a is fluidly coupled to the second cavity 314 adjacent to the bottom of the first immersion tank 202a. The second cavity 314 is fluidly coupled to the first cavity 312 and the outlet 318 via the channels 316. The first cavity 312 and the second cavity 314 receive a first working fluid 112.

[0029] In the illustrated example, the first control valve 214a is coupled to the first immersion tank 202a, but is located above the tank liquid level 302. In the illustrated example, the first waterfall conduit 210a is fluidly coupled to the heat exchanger 320 of the first server 212a. Specifically, the heat exchanger 320 in the illustrated example is fluidly coupled to the first control valve 214a and / or the first waterfall conduit 210a via a coupling conduit 322. The coupling conduit 322 is at least partially immersed in the first working fluid 112 of the first immersion tank 202a. The coupling conduit 322 and / or the waterfall conduit 210 can be a flexible hose, pipe, rigid pipe, duct, and / or any other fluid transport structure.

[0030] refer to Figures 3B-3D The first server 212a includes a printed circuit board 324, a semiconductor package 326, a heat exchanger 320, and electronic components 120 (e.g., memory sticks, RAM, etc.). The heat exchanger 320 of the illustrated example includes a chamber 328 defining a fluid flow passage 330 to allow fluid to flow between a first end or inlet 332 and a second end or outlet 334 of the heat exchanger 320. Specifically, the heat exchanger 320 includes a cold plate 336, a plurality of fins 338, and a cover 340. Specifically, the cover 340 and the cold plate 336 of the heat exchanger 320 define the chamber 328. The fins 338 extend from the cold plate 336 and are located within the chamber 328 between the cover 340 and the cold plate 336. The cold plate 336 is coupled to the semiconductor package 326. The fins 338 are spaced apart to define flow passages between corresponding fins within the fins 338. The inlet 332 of the heat exchanger 320 is coupled to a coupling conduit 322. Therefore, the first end 322a of the coupling conduit 322 is coupled to the first control valve 214a, and the second end 322b of the coupling conduit 322, opposite to the first end 322a, is coupled to the chamber 328 and / or the inlet 332 of the heat exchanger 320. In some examples, the first server 212a may include a plurality of semiconductor packages 326. In such an example, each of the plurality of semiconductor packages (e.g., XPUs) of the first server 212a may include a dedicated heat exchanger 320. In such an example, the plurality of coupling conduits 322 may couple a corresponding heat exchanger in the heat exchanger 320 of the first server 212a to the first control valve 214a and / or the first waterfall conduit 210a.

[0031] refer to Figures 3A-3DIn operation, the ability of the immersion cooling system 102 to remove heat from the first server 212a is proportional to the rate at which the first working fluid 112 flows through the immersion cooling system 102. For example, to provide the maximum flow rate, the first control valve 214a can be positioned fully open. To provide a reduced flow rate of the first working fluid 112 to the first server 212a, the first control valve 214a can be adjusted or positioned in a partially open position (e.g., between a fully open and a fully closed position). To prevent the first working fluid 112 from flowing to the first server 212a, the first control valve 214a can be positioned in a fully closed position (e.g., a shut-off position). In some examples, the first control valve 214a can be positioned fully closed when the first server 212a needs to be removed from the first immersion tank 202a and / or if the first server 212a is powered off or does not generate significant heat. In some examples, in response to the power output performance of the first server 212a, the flow rate of the first working fluid 112 can be adjusted between a first flow rate (e.g., the maximum flow rate) and a second flow rate (e.g., a flow rate less than the maximum flow rate). In some examples, a control system may be employed to monitor the temperature of the electronic components 120 and / or the server 212 (e.g., via one or more temperature sensors), and adjust (e.g., increase or decrease) the fluid flow to the server 212 via a control valve 214 based on the power consumption and / or temperature of the electronic components 120 and / or the server 212.

[0032] To cool or remove heat generated by the first server 212a, the first control valve 214a is moved to the open position to allow the first working fluid 112 to flow from the reservoir 204 to the first immersion tank 202a. A supply manifold 206 supplies the first working fluid 112 to a first waterfall conduit 210a, which in turn supplies the first working fluid 112 to the first control valve 214a. Specifically, the first working fluid 112 flows by gravity through the first waterfall conduit 210a. The first working fluid 112 flows from the first control valve 214a to the heat exchanger 320 via a coupling conduit 322. Specifically, the first working fluid 112 flows from the inlet 332, through the channel and / or fins 338, through the outlet 334, and into the first immersion tank 202a (e.g., the first cavity 312).

[0033] During operation, heat generated from the semiconductor package 326 is absorbed by the cold plate 336 and / or transferred to the fins 338. The heat exchanger 320 removes heat from the thermally challenged components (e.g., XPU, chipset, etc.) of the first server 212a, which includes the semiconductor package 326. As the first working fluid 112 flows through the chamber 328 of the heat exchanger 320 between the inlet 332 and the outlet 334, the first working fluid 112 flows between and / or across the fins 338 and removes heat from the fins 338 (i.e., removes heat generated by the semiconductor package 326). Specifically, due to the change in liquid level surface between the reservoir 204 and the first immersion tank 202a (… Due to the gravitational effect, the first working fluid 112 flows rapidly or at an increased rate through the heat exchanger 320. The flow rate of the first working fluid 112 through the heat exchanger 220 affects the amount of heat removed from the first server 212a, and thus the cooling efficiency of the immersion cooling system 102. Specifically, the higher the flow rate, the greater the amount of heat that can be removed. The slower the flow rate, the less heat that can be removed.

[0034] After flowing through heat exchanger 320, the first working fluid 112 exits heat exchanger 320 via outlet 334 and flows into first immersion tank 202a. In first immersion tank 202a, the first working fluid 112 exiting heat exchanger 320 flows into first cavity 312 and continues to absorb heat generated by other electronic components 120 (e.g., memory, etc.) coupled to circuit board 324 of first server 212a and immersed in the first working fluid 112 of first immersion tank 202a. As the first working fluid 112 in first cavity 312 of first immersion tank 202a absorbs heat, the density of the first working fluid 112 in first cavity 312 decreases, causing the first working fluid 112 in first cavity 312 to rise to the upper or lower part of first immersion tank 202a due to convection. As the first working fluid 112 in first cavity 312 rises, the first working fluid 112 flows out of first immersion tank 202a and flows to return manifold 208. Specifically, as the first working fluid 112 rises in the first cavity 312, it flows to the second cavity 314 via a channel 316 formed in the inner wall 308 of the first immersion tank 202a. The first pump 116 draws the first working fluid 112 from the tank outlet 318 via a return manifold 208 and pumps or recirculates it to the reservoir 204. The height 350° between the tank outlet 318 and the liquid level 304 in the reservoir is... The height at which the first working fluid 112 should be pumped for pump energy assessment is [height missing]. The first pump 116 can be a high-capacity pump required to move the first working fluid 112 from the tank outlet 318 to the reservoir 204. Therefore, the first pump 116 may require low power consumption. Before pumping the first working fluid 112 to the reservoir 204, the first pump 116 recirculates the first working fluid 112 from the first immersion tank 202a to [height missing]. Figure 1 The cooler 104 is used to cool the first working fluid 112 before it flows into the reservoir 204.

[0035] The first immersion tank 202a automatically balances the tank liquid level 302 of the first working fluid 112 within it. Therefore, there is no need for complex electro-hydraulic level control sensors or devices. For example, if the flow rate of the first working fluid 112 through the heat exchanger 320 increases, the liquid level surface of the first working fluid 112 in the first cavity 312 also rises, causing a larger amount of the first working fluid 112 in the first cavity 312 to flow into the second cavity 314 through the channel 316 of the inner wall 308. Conversely, if the flow rate of the first working fluid 112 through the heat exchanger 320 decreases, the liquid level of the first working fluid 112 in the first immersion tank 202a decreases, and a smaller amount of the first working fluid 112 in the first cavity 312 flows into the second cavity 314 through the channel 316. Therefore, the first working fluid 112 in the first immersion tank 202a self-levels based on its flow rate.

[0036] Therefore, the gravity effect and / or pressure head of the example immersion cooling system 102 significantly improve the cooling capacity of the immersion cooling system 102. In some examples, there is a liquid height change of approximately 1.4 meters between the reservoir 204 and the immersion tank 202. This can increase the flow rate of the immersion cooling system 102 from approximately 3.1 liters per minute (LPM) to approximately 3.9 liters per minute (LPM). In contrast, a cooling system implemented without the example reservoir 204 typically provides approximately 900 watts (W) of cooling capacity for an immersion cooling system, which includes a pump for causing fluid flow through an immersion tank having a height of 1 meter. In implementing the example immersion cooling system 102 disclosed herein, due to the increased head provided by the reservoir, including at a height of approximately 2 meters (e.g., ...), The immersion cooling system 102 of the reservoir 204 (equivalent to 2 meters) provides an increased flow rate of approximately 7 liters per minute and a cooling capacity of approximately 1050 watts (W). In some examples, due to the increased pressure head provided by the reservoir 204, including those disclosed herein at a height of approximately 3 meters (e.g., 2 meters), the cooling capacity is further enhanced. The cooling system of the reservoir 204 (3 meters) provides an increased flow rate of approximately 9 liters per minute and a cooling capacity of approximately 1106 watts (W).

[0037] Figure 4 This is a schematic diagram of an example computing environment 400 having the example immersion cooling system 402 disclosed herein. Figure 4 Many components of the example immersion cooling system 402 are combined with the above. Figure 1 , Figure 2 and Figures 3A-3D The components described are substantially similar or identical. Therefore, these components will not be described in detail below. Instead, interested readers should refer to the corresponding descriptions above for a complete written description of the structure and operation of such components. To facilitate this process, similar or identical reference numerals will be used in the accompanying drawings. Figure 4 Similar structures in, such as Figure 1 , Figure 2 and Figures 3A-3D Used in, for example. Figure 4 The immersion cooling system 402 includes a supply manifold 206, multiple waterfall ducts 210, multiple immersion tanks 202 housing multiple electronic components 120 and / or servers 212, and a cooling tower 106. The illustrated immersion cooling system 402 includes a reservoir 404 with a heat exchanger 406, which cools the first working fluid 112 in the reservoir 404 when the first working fluid 112 is present in the reservoir 404. Therefore, Figure 4 The immersion cooling system 402 is basically similar to Figure 1 , Figure 2 and Figures 3A-3D The immersion cooling system 102, in addition to the reservoir 404, includes a heat exchanger 406. In the illustrated example, the heat exchanger 406 is fluidly coupled to a second loop system 110. Thus, the heat exchanger 406 in the illustrated example receives a second working fluid 114 from the cooling tower 106. As the second working fluid 114 circulates through the heat exchanger 406 via the second loop system 110, the second working fluid 114 absorbs and / or removes heat from the first working fluid 112 in the reservoir 404. By positioning the heat exchanger 406 within the reservoir 404, the illustrated immersion cooling system 402 can conserve space and / or improve the heat exchange efficiency of the immersion cooling system 402.

[0038] Figure 5 One or more example environments in which the teachings of this disclosure can be implemented are shown. Figure 5 One or more example environments may include one or more central data centers 502. One or more central data centers 502 may store a large number of servers used by, for example, one or more organizations for data processing, storage, etc. Figure 5As shown, one or more central data centers 502 include multiple immersion tanks 504 for facilitating the cooling of servers and / or other electronic components stored at the central data centers 502. The immersion tanks 504 can provide single-phase or two-phase cooling.

[0039] Figure 5 An example environment could be part of an edge computing system. For example, Figure 5 Example environments may include edge data centers or micro data centers 506. One or more edge data centers 506 may include, for example, data centers located at the base of a cell tower. In some examples, one or more edge data centers 506 are located at or near the top of a cell tower and / or other utility pole. One or more edge data centers 506 include corresponding enclosures for storing one or more servers, wherein the servers can communicate with, for example, servers, client devices, and / or other computing devices in an edge network stored at one or more central data centers 502. Example enclosures of one or more edge data centers 506 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 immersion. Example enclosures may include power circuitry modules that provide power for fixed and / or portable implementations, such as AC power inputs, DC power inputs, one or more AC / DC or DC / AC converters, power conditioners, transformers, charging circuitry modules, batteries, wired inputs, and / or wireless power inputs. Figure 5 As shown, one or more edge data centers 506 may include one or more immersion tanks 508 for storing one or more servers and / or one or more other electronic components located in one or more edge data centers 506.

[0040] Figure 5 One or more example environments may include a building 510 for commercial and / or industrial purposes, which stores information technology (IT) equipment in one or more rooms of, for example, one or more buildings 510. Figure 5As shown, one or more servers 512 may be stored together with one or more server racks 514 supporting the servers 512 (e.g., in the openings of slots in rack 514). In some examples, the servers 512 located at building 510 include one or more on-premise servers of an edge computing network, wherein the on-premise servers communicate with one or more remote servers (e.g., one or more servers at one or more edge data centers 506) and / or one or more other computing devices within the edge network.

[0041] Figure 5 One or more example environments include one or more Content Delivery Network (CDN) data centers 516. The one or more CDN data centers 516 in this example include one or more servers 518 that cache content such as images, web pages, and videos accessed via user devices. The one or more servers 518 of the CDN data center 516 may be located in one or more immersion cooling tanks, such as immersion tanks 504 and 508 shown in conjunction with data centers 502 and 506. In some instances, Figure 5 Example data centers 502, 506, 516 and / or (one or more) buildings 510 include servers and / or other electronic components cooled independently of immersion tanks (e.g., immersion tanks 504, 508) and / or associated immersion cooling systems.

[0042] Although the diagrams show a number of cooling tanks and other components, any number of such components can be present. Furthermore, the examples of cooling data centers and / or other structures or environments disclosed herein are not limited to... Figure 5 The arrangement of sizes depicted herein. For example, a structure incorporating the example cooling system and / or components disclosed herein may have a size including openings to accommodate service personnel (e.g., Figure 5 Examples of data centers 506 (or one or more) can be included, but smaller (e.g., "doghouse" enclosures). For example, a structure incorporating the example cooling systems and / or components disclosed herein can be sized such that the entrance (e.g., the only entrance) inside the structure is the port through which service personnel enter the structure. In some examples, a structure incorporating the example cooling systems and / or components disclosed herein can be sized such that only tools can reach into the enclosure, as the structure may be supported by a utility pole or radio tower or a larger structure. In addition to or as an alternative to immersion tanks 504 and 508, Figure 6Any example environment in the example environment can utilize one or more liquid cooling systems with cold plates to control the temperature of electronic devices / components in the example environment.

[0043] Figure 7 An example data center 600 is illustrated, in which distributed resources can collaboratively perform one or more workloads (e.g., applications representing clients). The illustrated data center 600 includes multiple platforms 610, 620, 630, and 640 (referred to herein as container groups), each platform comprising one or more rows of racks. Although data center 600 is shown as having multiple container groups, in some examples, data center 600 may be implemented as a single container group. As described in more detail herein, a rack may house multiple sleds. Sleds may be primarily equipped with specific types of resources (e.g., memory devices, data storage devices, accelerator devices, general-purpose programmable circuit modules), which may be logically coupled to form constituent nodes. Some of these nodes may function as, for example, servers. In some examples, one or more sleds may be configured to be installed in a corresponding rack 740 of the aforementioned data center 600. In some examples, a given sled may be optimized or otherwise configured to perform specific tasks, such as compute tasks, acceleration tasks, data storage tasks, etc. For example, a skateboard can be implemented as a computing skateboard, an accelerometer skateboard, a storage skateboard, a skateboard optimized or otherwise configured to perform other specialized tasks, such as a memory skateboard and / or any other configuration.

[0044] In the illustrative example, skateboards in container groups 610, 620, 630, and 640 are connected to multiple container group switches (e.g., switches that route data communication to and from skateboards within the container groups). These container group switches are in turn connected to a backbone switch 650, which switches communication between container groups (e.g., container groups 610, 620, 630, and 640) within data center 600. In some examples, the skateboards may use Intel Omni-Path. TMThe technology is connected via a fabric. In other examples, the skateboard can be connected to other fabrics such as InfiniBand or Ethernet. As described in more detail herein, resources within a skateboard in data center 600 can be allocated to a group (referred to herein as a “management node”) containing resources from one or more skateboards for shared use in the execution of workloads. Workloads can be executed as if resources belonging to a management node were located on the same skateboard. Resources in a management node can belong to skateboards (which belong to different racks) and even to different container groups 610, 620, 630, 640. Thus, some resources from a single skateboard can be allocated to one management node, while other resources from the same skateboard are allocated to different management nodes (e.g., a first programmable circuit module is allocated to one management node, while a second programmable circuit module from the same skateboard is allocated to a different management node).

[0045] Data centers, including distributed resources (such as Data Center 600), can be used in a wide variety of contexts such as enterprises, governments, cloud service providers, and communications service providers (e.g., telecom providers, telephone companies (e.g., Telco)), and in a wide variety of sizes, from large cloud service provider data centers consuming more than 200,000 square feet to single-rack or multi-rack installations used for base stations.

[0046] In some examples, resource distribution is achieved by using individual skateboards that primarily comprise a single type of resource (e.g., a compute skateboard primarily comprising compute resources, a memory skateboard primarily comprising memory resources). Distributing resources in this way, and selectively allocating and dedistributing distributed resources to form management nodes assigned to perform workloads, improves the operation and resource utilization of Data Center 600 compared to a typical data center. Such a typical data center includes hyperconverged servers containing compute, memory, storage, and possibly additional resources within a single chassis. For example, because a given skateboard will primarily contain the same specific type of resource, that type of resource can be upgraded independently of other resources. Furthermore, because different resource types (programmable circuit modules, memory, accelerators, etc.) typically have different refresh rates, higher resource utilization and lower total cost of ownership can be achieved. For example, a data center operator can upgrade programmable circuit modules throughout the facility by replacing only the compute skateboard. In such a case, accelerators and storage resources may not be upgraded simultaneously and can instead be allowed to continue operating until those resources are scheduled for their own refresh. Resource utilization can also be increased. For example, if management nodes are configured based on the requirements of the workloads that will run on them, the resources within the nodes are more likely to be fully utilized. Such utilization can allow more management nodes to be run in a data center with a given set of resources, or allow a data center to be built with fewer resources to run a given set of workloads.

[0047] Now for reference Figure 7In the illustrative example, container group 610 includes a row of racks 700, 710, 720, and 730 of rack 740. Individual racks within rack 740 can accommodate multiple slides (e.g., 16 slides) and provide power and data connectivity to the accommodated slides, as described in more detail herein. In the illustrative example, the racks are connected to multiple container group switches 750 and 760. Container group switch 750 includes a set of ports 752 and another set of ports 754, with slides of the racks of container group 610 connected to the set of ports 752 and the other set of ports 754 connecting container group 610 to backbone switch 650 to provide connectivity to other container groups in data center 600. Similarly, container group switch 760 includes a set of ports 762 and a set of ports 764, with slides of the racks of container group 610 connected to the set of ports 762 and the other set of ports 764 connecting container group 610 to backbone switch 650. Therefore, the use of the pair of switches 750 and 760 provides a certain amount of redundancy for container group 610. For example, if either switch 750 or 760 fails, the slats in container group 610 can still maintain data communication with the rest of data center 600 (e.g., slats in other container groups) through the other switch 750 or 760. Furthermore, in the illustrative example, switches 650, 750, and 760 can be implemented as dual-mode optical switches, capable of routing 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) Fast (PCIe)) via optically constructed optical signaling media.

[0048] It should be understood that any of the other container groups 620, 630, and 640 (and any additional container groups in data center 600) can be similarly constructed as follows: Figure 7 Shown and about Figure 7 The disclosed container group 610, and has similar characteristics to... Figure 7 Shown and about Figure 6 The disclosed container group 610 comprises components (e.g., a given container group may have multiple rows of racks accommodating multiple racks as described above). Furthermore, while two container group switches 750 and 760 are shown, it should be understood that in other examples, different numbers of container group switches may exist, providing even more failover capabilities. In other examples, the number may differ from... Figure 7 and ​The multi-row rack configuration shown is used to arrange the container group. For example, the container group may include multiple sets of radially arranged racks (e.g., racks equidistant from a central switch). The server container group of the example shown may include one or more immersion cooling tanks. By directly absorbing and dissipating heat in the dielectric liquid within the server immersion tank, the need for conventional air conditioning can be significantly reduced, and overall energy efficiency can be improved.

[0049] "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., comprises, includes, comprising, including, having, etc.) as a preamble or in any kind of claim recitation, it is understood that additional elements, terms, etc., may exist that do not fall outside the scope of the corresponding claim or recitation. As used herein, when the phrase "at least" is used as a transitional term in, for example, the preamble of a claim, it is open-ended in the same way that the terms "comprising" and "including" are open-ended. When used, for example, in the form of A, B, and / or C, the term "and / or" refers to any combination or subset of A, B, C, such as (1) A alone, (2) B alone, (3) C alone, (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, articles, 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, articles, objects, and / or things, the phrase “at least one of A or 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 execution or performance 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 herein in the context of describing the execution or implementation 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 one of: (1) at least one A, (2) at least one B, or (3) at least one A and at least one B.

[0050] As used herein, singular references (e.g., "a(a), (an)," "first," "second," etc.) do not exclude plurals. As used herein, the term "a(a) or (an)" 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 parts, elements, or actions may be implemented by, for example, the same entity or object. Additionally, while individual features may be included in different examples or claims, these may be combined, and inclusion in different examples or claims does not imply that a combination of features is not feasible and / or advantageous.

[0051] As used herein, unless otherwise stated, the term "above" describes the relationship of two parts relative to the ground. If the second part has at least one portion between the ground and the first part, then the first part is above the second part. Similarly, as used herein, the first part is "below" the second part when it is closer to the ground than the second part. As stated above, the first part may be above or below the second part in one or more of the following situations: other parts in between; no other parts in between; where the first and second parts are in contact; or where the first and second parts are not in direct contact with each other.

[0052] As used in this patent, it is specified that any portion (e.g., layer, film, region, area, or plate) is in any manner on (e.g., positioned on, located, disposed on, or formed on, etc.) another portion, indicating that the reference portion is in contact with the other portion or that the reference portion is above the other portion, wherein one or more intermediate portions are located between them.

[0053] As used herein, unless otherwise indicated, a connection reference (e.g., attachment, coupling, connection, and joining) may include intermediate components between the elements referenced by the connection reference and / or relative movement between those elements. In this way, 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 specification that any part is "in contact" with other parts is defined as meaning that there is no intermediate part between the two parts.

[0054] Unless otherwise specifically stated, descriptors such as “first,” “second,” “third,” etc., used herein are not intended to indicate any meaning of priority, physical order, arrangement in a list, and / or any sorting, but are merely used as labels and / or arbitrary names to distinguish elements in order to facilitate understanding of the disclosed examples. In some examples, the descriptor “first” may be used to refer to an element in a detailed description, 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 clearly identify those elements within the context of the discussion (e.g., within the claims), where said elements may otherwise share the same name, for example.

[0055] As used herein, “approximately” and “about” modify their subject / 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 tolerances of + / - 10%, unless otherwise stated herein.

[0056] As used herein, a “programmable circuit module” is defined to include: (i) one or more application-specific circuits (e.g., application-specific integrated 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 general-purpose semiconductor-based circuits programmable with instructions for performing one or more specific functions and / or one or more specific operations and including one or more semiconductor-based logic devices (e.g., electrical hardware implemented by one or more transistors). Examples of programmable circuit modules include programmable microprocessors such as a central processing unit (CPU) capable of executing a first instruction to perform one or more operations and / or functions; a field-programmable gate array (FPGA) whose configuration and / or construction can be programmed with a second instruction to executor one or more operations and / or functions corresponding to the first instruction; a graphics processing unit (GPU) capable of executing a first instruction to perform one or more operations and / or functions; a digital signal processor (DSP) capable of executing a first instruction to perform one or more operations and / or functions; an XPU; a network processing unit (NPU); and one or more microcontrollers and / or integrated circuits (such as application-specific integrated circuits (ASICs)) capable of executing a first instruction to perform one or more operations and / or functions. For example, an XPU can be implemented by a heterogeneous computing system that includes multiple types of programmable circuit modules (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 one or more of the multiple types of programmable circuit modules suitable for performing the computing tasks.

[0057] As used herein, an integrated circuit / circuit module 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 module, semiconductor substrate coupling multiple circuit elements, system-on-a-chip (SoC), etc.

[0058] As will be understood above, example systems, apparatus, articles, and methods for improving the cooling capacity of immersion cooling systems have been disclosed. Further examples and combinations thereof include the following: Example 1 includes an immersion cooling system comprising: an immersion tank containing cooling fluid; a reservoir for containing a recirculation portion of the cooling fluid, the reservoir being separated from the immersion tank by a certain height to create a liquid level change between the cooling fluid in the immersion tank and the cooling fluid in the reservoir; and a supply conduit for fluidly coupling the immersion tank and the reservoir, the cooling fluid being supplied from the reservoir to the immersion tank via gravity.

[0059] Example 2 includes the system of Example 1, and further includes a waterfall conduit located between the supply conduit and the immersion tank.

[0060] Example 3 includes a system of any one of Examples 1-2, and further includes a control valve coupled to the waterfall conduit.

[0061] Example 4 includes a system of any one of Examples 1-3, and further includes a coupling conduit to the control valve and a heat exchanger for electronic components located in the immersion tank.

[0062] Example 5 includes a system of any one of Examples 1-4, wherein the control valve controls the flow rate of the cooling fluid flowing from the reservoir to the heat exchanger.

[0063] Example 6 includes a system of any one of Examples 1-5, and further includes a return manifold coupled to the outlet of the immersion tank.

[0064] Example 7 includes a system comprising any one of Examples 1-6, and further includes a first pump for recirculating the cooling fluid at the outlet of the immersion tank to the reservoir.

[0065] Example 8 includes a system comprising any one of Examples 1-7, and further includes a cooler disposed between the return manifold and the reservoir, the cooler being used to remove heat from the cooling fluid.

[0066] Example 9 includes a system comprising any one of Examples 1-8, and further includes a cooling tower for supplying working fluid to the cooler, wherein the working fluid removes heat from the cooling fluid as the cooling fluid flows through the cooler when the cooling fluid is recirculated from the outlet of the immersion tank and the reservoir.

[0067] Example 10 includes a system of any one of Examples 1-9, wherein the immersion tank includes an inner wall and an outer wall, the inner wall defining a first cavity, and the inner wall and the outer wall defining a second cavity.

[0068] Example 11 includes a system of any one of Examples 1-10, wherein the second cavity fluid is coupled to the outlet of the immersion tank.

[0069] Example 12 includes a system of any one of Examples 1-11, wherein the first cavity is fluidly coupled to the second cavity via one or more channels formed through the inner wall.

[0070] Example 13 includes a system of any one of Examples 1-12, wherein the immersion tank is located at a first elevation and the reservoir is located at a second elevation, which is greater than the first elevation.

[0071] Example 14 includes a system of any one of Examples 1-13, wherein the cooling fluid in the first cavity flows to the second cavity when the density of the cooling fluid in the first cavity decreases.

[0072] Example 15 includes an immersion cooling system comprising: an electronic assembly including: a frame; a printed circuit board mounted to the frame; a semiconductor chip package coupled to the printed circuit board; a heat sink having a fluid flow cavity including fins, the heat sink being coupled to the semiconductor chip package; a first chamber including a first volume of liquid coolant at a first altitude, the electronic assembly being immersed in the liquid coolant in the liquid chamber; and a second chamber including a second volume of the liquid coolant at a second altitude different from the first altitude, such that a pressure head is provided between the second volume of liquid coolant in the second chamber and the first volume of liquid coolant in the first chamber.

[0073] Example 16 includes a system of any one of Examples 1-15, and further includes a conduit that fluidly couples the liquid coolant in the second chamber to the radiator of the electronic assembly immersed in the first chamber.

[0074] Example 17 includes a system of any one of Examples 1-16, wherein the pressure head provided between the first chamber and the second chamber is intended to provide a fluid flow rate of the liquid coolant flowing from the second chamber to the first chamber.

[0075] Example 18 includes a system of any one of Examples 1-15, wherein the outlet of the fluid flow cavity of the radiator is in fluid communication with the liquid coolant in the first chamber, such that the liquid coolant flowing from the second chamber through the cavity of the heat exchanger is deposited into the first chamber.

[0076] Example 19 includes a system comprising: a first chamber including an immersion liquid; an electronic unit within the immersion liquid; a heat exchanger coupled to the electronic unit; a second chamber spaced apart from the first chamber to generate a pressure head, the second chamber including the immersion liquid; and a flow channel fluidly coupling the second chamber and the heat exchanger, the flow channel being configured to allow fluid to flow through the exchanger, which is a semiconductor chip package coupled to the electronic unit.

[0077] Example 20 includes the system of Example 19, and further includes a control valve inserted into the second flow passage to control the fluid flow to the heat exchanger.

[0078] The appended claims are hereby incorporated by reference into this specific embodiment. Although certain example systems, devices, articles of manufacture, and methods have been disclosed herein, the scope of this patent is not limited thereto. Rather, this patent covers all systems, devices, articles of manufacture, and methods that fall fully within the scope of the claims of this patent.

Claims

1. An immersion cooling system, comprising: An immersion tank, which includes cooling fluid; A reservoir for containing a recirculation portion of the cooling fluid, the reservoir being separated from the immersion tank by a certain height to create a change in liquid level between the cooling fluid in the immersion tank and the cooling fluid in the reservoir; and A supply conduit is provided for fluid coupling between the immersion tank and the reservoir, wherein the cooling fluid is supplied from the reservoir to the immersion tank by gravity.

2. The system of claim 1 further includes a waterfall conduit located between the supply conduit and the immersion tank.

3. The system according to any one of claims 1-2 further includes a control valve coupled to the waterfall conduit.

4. The system according to any one of claims 1-3, further comprising a coupling conduit to the control valve and a heat exchanger for the electronic components located in the immersion tank.

5. The system according to any one of claims 1-4, wherein, The control valve controls the flow rate of the cooling fluid flowing from the reservoir to the heat exchanger.

6. The system according to any one of claims 1-5, further comprising a return manifold coupled to the outlet of the immersion tank.

7. The system according to any one of claims 1-6, further comprising a first pump for recirculating the cooling fluid at the outlet of the immersion tank to the reservoir.

8. The system according to any one of claims 1-7, further comprising a cooler disposed between the return manifold and the reservoir, the cooler being used to remove heat from the cooling fluid.

9. The system according to any one of claims 1-8, further comprising a cooling tower for providing a working fluid to the cooler, wherein the working fluid removes the heat from the cooling fluid as the cooling fluid flows through the cooler when the cooling fluid is recirculated from the outlet of the immersion tank and the reservoir.

10. The system according to any one of claims 1-9, wherein, The immersion tank includes an inner wall and an outer wall, the inner wall defining a first cavity, and the inner wall and the outer wall defining a second cavity.

11. The system according to any one of claims 1-10, wherein, The second cavity fluid is coupled to the outlet of the immersion tank.

12. The system according to any one of claims 1-11, wherein, The first cavity is fluidly coupled to the second cavity via one or more channels formed through the inner wall.

13. The system according to any one of claims 1-12, wherein, The immersion tank is located at a first elevation, and the reservoir is located at a second elevation, which is greater than the first elevation.

14. The system according to any one of claims 1-13, wherein, When the density of the cooling fluid in the first cavity decreases, the cooling fluid in the first cavity flows to the second cavity.

15. An immersion cooling system, comprising: Electronic assemblies, including: frame; Printed circuit boards mounted to the frame; Semiconductor chip packages coupled to the printed circuit board; A heat sink having a fluid flow cavity including fins, the heat sink being coupled to the semiconductor chip package; A first chamber comprising a first volume of liquid coolant, the first chamber being located at a first altitude, the electronic assembly being immersed in the liquid coolant within the liquid chamber; and A second chamber comprising a second volume of the liquid coolant, the second chamber being at a second altitude different from the first altitude, such that a pressure head is provided between the second volume of the liquid coolant in the second chamber and the first volume of the liquid coolant in the first chamber.

16. The system of claim 15, further comprising a conduit for fluidly coupling the liquid coolant in the second chamber to the cavity fluid of the heat sink of the electronic assembly immersed in the first chamber.

17. The system according to any one of claims 15-16, wherein, The pressure head provided between the first chamber and the second chamber is intended to provide the fluid flow rate of the liquid coolant flowing from the second chamber to the first chamber.

18. The system according to any one of claims 15-17, wherein, The outlet of the fluid flow cavity of the radiator is in fluid communication with the liquid coolant in the first chamber, such that the liquid coolant flowing from the second chamber through the cavity of the heat exchanger is deposited into the first chamber.

19. A system comprising: Includes the first chamber containing the bathing liquid; The electronic units within the immersion liquid; A heat exchanger coupled to the electronic unit; A second chamber, spaced apart from the first chamber to generate a pressure head, includes the bath liquid; and A flow channel is fluidly coupled between the second chamber and the heat exchanger, the flow channel being configured to allow fluid to flow through the exchanger, which is coupled to the semiconductor chip package of the electronic unit.

20. The system of claim 19 further includes a control valve inserted into the second flow passage to control fluid flow to the heat exchanger.

Images