System and method for cooling system pump control

Abstract

The invention relates to a system and method for cooling system pump control. A method of controlling a coolant distribution unit is disclosed. The method includes: initiating, by a controller, a pump control mode; measuring, by a sensor, a differential pressure at a filter connected to the pump; determining, by the controller, whether the differential pressure is greater than a first predetermined value; when it is determined that the differential pressure is larger than the first preset value, whether the differential pressure is larger than a second preset value or not is determined, and the second preset value is larger than the first preset value; and generating a first output upon determining that the differential pressure is less than or equal to a second predetermined value.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to cooling systems, and more particularly to cooling systems and methods for use with data centers. BACKGROUND

[0002] Buildings or facilities with large amounts of electronic equipment consume large amounts of electrical power. Electronic equipment generates undesirable high heat when in use, which must be removed to prevent equipment failure. In various buildings or facilities, as the demand for computing power continues to grow, data centers have a rising heat density, requiring thermal systems to provide increased cooling density. One common electronic equipment or server cooling method includes air cooling methods. Data centers can rely on air cooling to maintain safe operating temperatures by circulating cool air around the hardware equipment, as long as the power requirements of the electronic equipment or computing systems remain below a certain level, such as 20 kW. But today’s high performance computing systems can easily exceed 20 kW or more. This is largely because computing systems within a rack are configured with central processing units (CPUs) and graphics processing units (GPUs) that have much higher thermal power densities than previous generations. While some air cooling systems can support racks that require more than 20 kW of power, they tend to be inefficient and complex to maintain, making other solutions using fluids more attractive.

[0003] Accordingly, such server cooling for data centers is transitioning from air cooling to more efficient fluid cooling solutions, also known as liquid cooling. Fluid cooling can bring several benefits, including reduced energy costs, reduced carbon footprint, increased power density, higher cooling requirements, targeted cooling options, and significantly improved heat transfer (e.g., water has 4000 times the heat capacity of the same volume of air), thus greatly enhancing energy efficiency, and increasing the possibility of utilizing waste heat, resulting in a significant reduction in power usage effectiveness (PUE) for data centers. Moreover, more efficient cooling allows for more powerful server components, including higher processor clock speeds, resulting in a significant increase in power density within servers and the entire data center.

[0004] Liquid cooling systems include chip-level cold plate liquid cooling systems and immersion liquid cooling systems. Cold plate liquid cooling systems have a metal cavity directly on the server chip and achieve direct cooling of the chip through the circulation flow of the cooling liquid. Direct-to-chip cooling—sometimes also referred to as direct-to-board cooling—integrates the cooling system directly into the computer chassis. The cooling liquid is piped to cold plates that are placed directly next to components such as CPUs, GPUs, memory cards, etc. Small fluid channels can transport the cooling liquid to each board, where the liquid takes away heat from the underlying components. The warm liquid is then circulated to a cooling device or heat exchanger. After the warm liquid is cooled, the liquid is then circulated back to the cold plates.

[0005] For example, cold plate liquid-cooled servers are mounted vertically on a cooling rack, a main pipe for central liquid supply and return is arranged on the cooling rack, and the main pipe is connected with a liquid cooling secondary side main pipe of a coolant distribution device (CDU) below or above the cooling rack to form a cooling loop. Because liquid is more effective than air in transferring heat, the CDU can be introduced in the case where the generated heat is increased to exceed the capacity of the air cooling system at a higher rack density. The CDU network exchanges heat between the existing facility cooling system or liquid loop (FWS) of the building and the CDU loop (TCS) that operates closer to the IT equipment. Referring to Figure 1 , Figure 1 A CDU is shown as a liquid-to-air heat exchanger for cooling chips according to the related art, a server fan blows out warm air through the exchanger, which dissipates heat. Liquid is circulated by implementing a closed loop system of heat exchange. Liquid-cooled board servers are connected with a main pipe of a cooling rack at the same time in order to be connected with the cooling liquid. The coolant distribution unit creates an independent secondary loop separate from the chilled water supply, thereby achieving strict containment management and precise control of the temperature, pressure, and flow rate of the liquid cooling system.

[0006] One disadvantage of such liquid-cooled direct-to-chip applications includes ensuring that the liquid flowing to the cooled servers is clean. The cleanliness of the fluid in liquid-cooled direct-to-chip applications is critical to the targeted cooling operation. For this reason, the CDU is typically installed with a filter to capture a diameter of about 25 m (e.g., 25 m of white blood cells) to 50 m (e.g., human hair ranges from 50 m to 100 m) to protect the cold plate. Because the filter can capture small contaminants, the filter is vulnerable to breakage due to clogging. However, current control efforts exaggerate the problem of filter clogging. As the filter clogs, the pump speed will increase to meet the IT load flow requirements, further increasing the differential pressure across the filter. A failed filter can be catastrophic to the entire data center system and can cause cold plate damage, not to mention CPUs, GPUs, and any other electronic components in the server rack.

[0007] Accordingly, there is a need to ensure cleanliness of filters and prevent failure of data center cooling systems and servers that occur due to filter failure and / or contamination. In addition, even if filter contamination is predicted, there is a need to effectively control the cooling unit to effectively and carefully control cooling of the entire data center server. SUMMARY

[0008] This section provides a general summary of the disclosure and is not a comprehensive disclosure of its full scope or all of its features. The implementations described herein relate to techniques for cooling data centers. In particular, the systems and methods of the present disclosure provide new and novel techniques for detecting errors in advance to prevent failure of coolant distribution units (CDUs) of data centers, much less data center servers. For example, a CDU having one or more pumps and corresponding one or more filters can fail due to contamination in the CDU. As contaminated fluid passes through the CDU, its filters can become clogged and damaged. Currently, one or more dirty filters can be notified by having an alarm (e.g., a non-critical alarm) based on an operating condition user set point. By adding an additional alarm (e.g., a critical alarm) to selectively shut down a CDU component (e.g., a filter) when the differential pressure across the filter exceeds a threshold, catastrophic failure can be prevented and liability reduced. Moreover, such an additional alarm would help to separate product lines and ensure more sales.

[0009] Systems and methods according to the present disclosure can be implemented into existing liquid cooling management systems or coolant distribution units. The present disclosure includes methods for providing additional steps to control a pump unit upon detecting a failure. While the methods described herein can apply to coolant distribution units that provide liquid coolant to server racks, the present disclosure can apply to any system that includes a pump and a filter in communication with the filter and coolant flowing therethrough.

[0010] According to implementations of the present disclosure, a method of controlling a coolant distribution unit includes initiating, by a controller, a pump control mode; measuring, by a sensor, a differential pressure at a filter connected to a pump; determining, by the controller, whether the differential pressure is greater than a first predetermined value; upon determining that the differential pressure is greater than the first predetermined value, determining whether the differential pressure is greater than a second predetermined value, wherein the second predetermined value is greater than the first predetermined value; and upon determining that the differential pressure is less than or equal to the second predetermined value, generating a first output. In some implementations, the step of generating the first output includes determining that the filter is dirty.

[0011] In some embodiments, the method further includes returning to the step of initiating the pump control mode upon determining that the differential pressure is less than or equal to a first predetermined value, and generating a second output upon determining that the differential pressure is greater than a second predetermined value. The method further includes the step of turning off the pump.

[0012] In some embodiments, when there are multiple pumps, the step of turning off the pump includes turning off all of the multiple pumps. The method further includes receiving a user input of whether the filter is cleaned, generating a third output upon determining that the filter is cleaned, and turning on the pump. The first output, the second output, and the third output are different from each other. In some embodiments, the method further includes determining a particular filter of the multiple filters that has a differential pressure greater than the first predetermined value when the pump and the filter are set in a plurality.

[0013] In some embodiments, the method further includes determining whether the differential pressure of the particular filter is greater than a second predetermined value, generating the second output and turning off a particular pump in fluid communication with the particular filter upon determining that the differential pressure of the particular filter is greater than the second predetermined value, reducing a temperature set point based on a pump speed reduction, and turning on the particular pump when it is determined that the particular filter in fluid communication with the particular pump is cleaned.

[0014] In some embodiments, the method further includes generating the second output upon determining that the differential pressure is greater than the second predetermined value, determining again whether the differential pressure is greater than the second predetermined value, setting a filter set point upon determining that the differential pressure is greater than the second predetermined value, reducing the temperature set point based on a pump speed reduction, receiving a user input of whether the filter has been cleaned upon determining that the differential pressure is less than or equal to the second predetermined value, and generating a third output upon determining that the filter is cleaned. The first output, the second output, and the third output are different from each other.

[0015] According to another embodiment of the present disclosure, a coolant distribution unit includes a heat exchanger, a pump in selective communication with the heat exchanger through a valve, a filter in communication with the pump, a sensor configured to measure a differential pressure at the filter connected to the pump, and a controller configured to initiate a pump control mode and determine whether the differential pressure is greater than a first predetermined value. In some embodiments, the pump and the filter are set in a plurality.

[0016] Other aspects and advantages of the present application will become apparent from the following detailed description, taken in conjunction with the accompanying drawings, illustrating by way of example the principles of the application. BRIEF DESCRIPTION OF DRAWINGS

[0017] The present disclosure will be readily understood by persons skilled in the art with the foregoing description, taken in conjunction with the following detailed description of the preferred embodiments, perhaps with reference to the drawings, in which:

[0018] Figure 1a front perspective view of a coolant distribution device (CDU) showing partial coverage of the related art;

[0019] Figure 2A a front perspective view of a CDU without a cover according to the related art;

[0020] Figure 2B a back perspective view of a CDU without a cover according to the related art;

[0021] Figure 3 a back perspective view of a CDU without a cover according to another related art;

[0022] Figure 4 a schematic view of a CDU according to Figure 1

[0023] a schematic view of a CDU according to Figure 5 Figure 3

[0024] Figure 6 a flowchart of a control method according to a first exemplary embodiment of the present disclosure;

[0025] Figure 7 a flowchart of a control method according to a second exemplary embodiment of the present disclosure; and

[0026] Figure 8 a flowchart of a control method according to a third exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0027] Aspects of the present disclosure are more fully described herein with reference to the figures, which form a part of this disclosure. The following detailed description presents various features in connection with the schemes. The features described, however, can be combined in any combination by one of ordinary skill in the art. The following detailed description is not to be considered limiting in scope, but rather a representative way in which specific aspects of the schemes can be described. Thus, the following detailed description is presented for the purposes of illustrating representative features in connection with the schemes.

[0028] ​​For example, the use of the singular term "a," but not limited to, "a," is not intended to limit the number of items. The use of relational terms (e.g., but not limited to "top," "bottom," "left," "right," "above," "below," "under," "upper," "side," etc.) in the written description with specific reference to the accompanying drawings is for clarity and is not intended to limit the scope of the invention or the appended claims. The terms "comprising" and "e.g.," are for illustrative purposes and are not limited thereto. The terms "coupled," "coupled," "coupler," "coupler," and similar terms are used extensively herein and may include any method or device for securing, joining, engaging, fastening, attaching, connecting, inserting, forming thereon, or being in contact with one or more components together, for communicating one or more components together, or for associating them, for example mechanically, magnetically, electrically, chemically, operatively, directly, or indirectly through intermediate elements, and may also include, but is not limited to, integrally forming one functional component with another functional component. Coupling can occur in any direction, including rotationally. Furthermore, all parts and components of this disclosure that can be inherently implemented in a physical sense include both hypothetical and real features, whether or not such features are explicitly described herein, including but not limited to features such as axis, ends, inner and outer surfaces, internal space, top, bottom, sides, boundaries, dimensions (e.g., height, length, width, thickness), mass, weight, volume, and density.

[0029] Servers have become especially critical due to the daily use of the cloud by institutions, companies, and consumers. The number of data centers is constantly increasing. Servers are also becoming more powerful and thus generating more and more heat. While traditional air cooling systems are still widely used in data centers, their potential is limited, especially considering their environmental impact. Several liquid-based technologies have emerged, one of which is direct liquid cooling (DLC), also known as direct-to-chip liquid cooling. This technology cools only high-power components: processors, graphics cards, etc. The principle of DLC is to establish a cooling loop. Cold liquid is delivered to a cold plate directly mounted on hot electronic components and absorbs heat, which is then delivered to a coolant distribution unit (CDU) and then to a heat exchanger whose function is to dissipate heat. These are typically located in racks or shared by rows. When the liquid has cooled down, it returns to the cold plate, thus closing the cooling loop. In reality, not all the heat generated by the processor is removed by liquid cooling; a certain percentage is still removed by air cooling. Therefore, the entire environment is a hybrid liquid-air cooling system.

[0030] The Cooling Unit (CDU) is a fundamental component ensuring that liquid cooling systems perform as expected. Operational logic integrated into an intelligent controller, combined with smart controls, manages the performance of server and data center cooling systems. The CDU contains pumps that circulate coolant through a network of pipes or channels, distributing the coolant to various components, such as servers, processors, or other high-heat components in large, high-power equipment that require cooling. The CDU may also include valves, filters, and other components to control and monitor coolant flow.

[0031] This disclosure provides solutions for predicting CDU failures due to filter rupture before they occur. Furthermore, this disclosure allows for a simpler, more resilient approach to reducing the blast radius of a pump failure on a single cold plate within a single server.

[0032] It should be understood that the use of specific component, device, and / or parameter names and / or their corresponding abbreviations (e.g., those for execution utilities, logic, and / or firmware described herein) is for illustrative purposes only and does not imply any limitation on the described implementations. Therefore, implementations may be described using different nomenclature and / or terms to describe components, devices, parameters, methods, and / or functions herein without limitation. References to any particular protocol or proprietary name in describing one or more elements, features, or concepts of an implementation are provided only as an example of an implementation and do not limit the extension of the claimed implementation to implementations using different element, feature, protocol, or concept names. Therefore, each term used herein will be given its broadest interpretation, provided the context in which it is used is provided.

[0033] As used herein, the term "coolant distribution unit" (CDU) generally refers to a configuration designed to enhance the overall liquid cooling system of the system. The CDU circulates and pumps coolant within a closed-loop system within racks and server chassis, utilizing facility water (in a complete liquid cooling system) and outside air to cool the servers. CDUs are designed to manage the high thermal loads of accelerated computing and high-density data center environments, providing critical cooling capacity and heat removal capabilities. CDUs deliver controlled, contamination-free coolant to heat exchangers, direct-to-chip cooling equipment, and immersion cooling equipment.

[0034] The term "rack," also known as a server cabinet, generally refers to a type of frame, typically made of steel, that houses servers, cables, and other equipment. A CDU (Server Cooling Unit) communicates with the rack to supply cold liquid fluid (or coolant) to the servers within the rack, which is then heated and returned to the CDU. Details about the CDU are described below.

[0035] Coolant Dispenser Unit (CDU)

[0036] Figure 2A A front perspective view of the CDU 200 without covering, according to related technologies, is shown. Figure 2B A rear perspective view of a CDU 200 without a cover, according to related technologies, is shown.

[0037] Reference Figure 2A and Figure 2B The CDU 200 includes a controller or processor 202, one or more expansion containers 204, a primary filter bypass valve 206 (if assembled), one or more primary filter isolation valves 208 (if assembled), a water supply container 210, wheels and adjustable feet 212, an optional primary filter 214, a filling pump 216, a plate heat exchanger 218, one or more secondary pumps 220, one or more pump frequency converters 222, a controller touchscreen 224, and various sensors 226 including a room temperature sensor and an RH sensor. The CDU 200 also includes an automatic air vent 228, a pressure reducing valve 230, one or more optional secondary filters 232 respectively connected to one or more secondary pumps 220, one or more filter / pump isolation valves 234, a bypass shut-off valve 236, a primary cooling valve 238, one or more drain valves 240, a secondary flow meter 242, a primary flow meter 244, one or more level sensors 246, one or more pressure sensors 248, and one or more manual air vents 250.

[0038] Figure 3 A rear perspective view of a CDU 300 without a cover, according to another related technology, is shown.

[0039] The CDU 300 includes components similar to those in the CDU 200, such as one or more manual ventilation openings.

[0040] 302, pressure reducing valve; 304, one or more expansion vessels; 306, secondary flow meter; 308, one or more pressure sensor; 310, flexible supply vessel; 312, one or more primary cooling valve; 314, one or more discharge valve; 316, primary flow meter; 318, filling rod; 320, filling pump; 322, one or more filter / pump isolation valves; 324, one or more secondary filters; 326, automatic air vent; 328, one or more secondary pumps connected to one or more secondary filters 326 respectively; and a controller 332 having a power supply, pump frequency converter driver, touch screen and one or more sensors.

[0041] Figure 4 It shows Figure 2A and Figure 2B The schematic diagram of the CDU 400, andFigure 5 It shows Figure 3 A schematic diagram of the CDU, 500.

[0042] Schematic diagram 400 shows heat exchanger 402, which corresponds to Figure 2A The plate heat exchanger 218; one or more expansion vessels 404, corresponding to Figure 2A One or more expansion containers 204; one or more pumps 406, corresponding to Figure 2A One or more pumps 220; and one or more filters 408, corresponding to Figure 2A One or more optional secondary filters 232. Additionally, separate fluid loops with hot and cold flow lines extend from the CDU 200 to guide fluid (e.g., water) from separate heat exchange loops to exchange heat with the cooling fluid flowing through the heat exchanger 402. In use, the returned heated coolant flows into the inlet of one or more pumps 406. The outlets of one or more pumps 406 are respectively connected to one or more filters 408 for filtering the coolant before it enters the heat exchanger 402. Suitable valves can be used to selectively shut off the flow through one or more filters 408, for example, when it is changed, whereby the flow is then directed through the fluid lines to the filter in use. Components of the CDU 200 are arranged in the flow path in a manner that regulates the pressure supplied to the server rack. The CDU 200 also allows replacement of one of the multiple pumps while the remaining pumps continue to deliver coolant.

[0043] Schematic diagram 500 illustrates components similar to those described above. The main differences between schematic diagram 400 or CDU 200 and schematic diagram 500 or CDU 300 lie in the number of components, unit size, etc., for different applications. For example, compared to CDU 300, CDU 200 can be used in smaller spaces requiring less heat and electricity. Therefore, details of the CDU 300 components are omitted in this document.

[0044] Each CDU includes one or more pumps, allowing for selective control of one or more pumps as needed. CDU 200 and CDU 300 are configured to achieve uniform pressure delivery through operation of pumps, filters, heat exchangers, fluid channels, etc. The components of CDU 200 and CDU 300 can collectively create flow constraints affecting the delivery pressure of the server rack. The CDU 200 and CDU 300 described herein can be designed such that the various flow constraints provided by their components result in regulation of the internal gauge pressure of the electronic equipment, thereby reducing the range of pressure variations relative to changes in pump speed, etc.

[0045] The CDU described above can uniformly distribute liquid throughout the system. The CDU can pump cooled fluid to the server rack (or cold plate) in a closed loop, maximizing efficiency by precisely controlling and regulating the fluid temperature and flow rate, thereby supplying coolant under stable conditions, while the coolant heated by the heating equipment in the server rack can be cooled by separate coolant passing through the heat exchanger.

[0046] Server racks securely house rack-mountable systems. These systems can be, for example, high-power central processing units (CPUs), graphics processing units (GPUs), server blades, power supply units, keyboards, video recorders, and mouse (KVM) switches, network junction boxes, etc. One or more cold plates can be arranged directly or indirectly on the corresponding rack-mountable system. The cold plates can be any suitable type, such as tubular cold plates or cold plates including internal fins or channels (e.g., microchannels), and can be made of any suitable material (e.g., copper, aluminum, or stainless steel) that is chemically compatible with the immersion fluid and the working fluid.

[0047] Additionally, a rack-mountable system can be a heat-generating electronic device comprising one or more IT components (e.g., a central processing unit or CPU, a graphics processing unit (GPU), memory, and / or storage devices). Each IT component can perform a data processing task, wherein the IT component may include software installed in a storage device, loaded into memory, and executed by one or more processors to perform the data processing task. Furthermore, a server blade may include a host server (referred to as a host node) coupled to one or more compute servers (also referred to as compute nodes, such as CPU servers and GPU servers). The host server (having one or more CPUs) typically interacts with clients via a network (e.g., the Internet) to receive requests for specific services, such as storage services (e.g., cloud-based storage services, such as backup and / or recovery), thereby executing applications to perform certain operations (e.g., image processing, deep data learning algorithms, or modeling, as part of a Software as a Service or SaaS platform). In response to a request, the host server assigns the task to one or more performance compute nodes or compute servers (having one or more GPUs) managed by the host server. The performance compute servers perform the actual tasks, which generates heat during operation.

[0048] As described above, each CDU includes either a computing system or a controller ( Figure 2A "202" or Figure 3(referring to "332" in the original text). The controller can periodically or continuously monitor the operating status of the corresponding CDU. Operating status data can include real-time measurements of the operating temperature, coolant, airflow, differential pressure, flow rate, etc., of each processor. The computing system or controller can be a multiprocessor system including processor units. A processor unit includes multiple processor cores. A processor unit can execute a computer-executable program. The computing system can include any number of processor units. Furthermore, a processor unit can include any number of processor cores. Processor units can take various forms, such as a central processing unit (CPU), graphics processing unit (GPU), general-purpose GPU (GPGPU), accelerated processing unit (APU), field-programmable gate array (FPGA), neural network processing unit (NPU), data processing unit (DPU), accelerator (e.g., graphics accelerator, digital signal processor (DSP), compression accelerator, artificial intelligence (AI) accelerator), controller, or other types of processing units. Therefore, a processor unit can be referred to as an XPU (or xPU). Furthermore, a processor unit can include one or more of these various types of processing units. In some embodiments, the computing system includes a processor unit with multiple cores, and in other embodiments, the computing system includes a single processor unit with a single core. As used herein, the terms "processor unit" and "processing element" can refer to any processor, processor core, component, module, engine, circuitry, or any other processing element described or referenced herein. Furthermore, in some embodiments, a computing system may include one or more processor units that are heterogeneous or asymmetric to another processor unit in the computing system. Various differences may exist between processing units in a system in terms of a range of quality metrics, including architecture, microarchitecture, thermal, and power consumption characteristics. These differences can, in themselves, manifest as asymmetry and heterogeneity between processor units in the system.

[0049] Coolant Distribution Unit (CDU) shut-off filter protection

[0050] In the following text, refer to Figure 6A method 600 for controlling a CDU according to a first embodiment is described. In step 602, method 600 begins operation by turning on the CDU (e.g., CDU 200 or CDU 300). In steps 604 and 606, the user can input various settings, such as a differential pressure value, and begin pressure control of one or more pumps (e.g., one or more pumps 406 or 506) to a pump control mode. In step 608, the differential pressure of each filter is measured, e.g., the inlet pressure before entering the filter and the outlet pressure after leaving the filter. Then, step 610 checks whether the differential pressure measured in step 608 is greater than a first predetermined value, e.g., 6 psi. The first predetermined value may be determined based on the cooling / power required by the data center server or the type of CDU, and may be set by the user. If it is determined that the differential pressure is less than or equal to the first predetermined value, the method determines in step 612 that no alarm is needed, and repeats the process from step 606.

[0051] In some implementations, if the measured differential pressure is determined to be greater than a predetermined value, the method continues to step 614, and checks whether the measured differential pressure is greater than a second predetermined value. The second predetermined value is greater than the first predetermined value, for example, 10 psi; the second predetermined value can also be determined and set by the user. In step 616, if the second predetermined value is determined to be less than or equal to the second predetermined value (e.g., the measured differential pressure could be 8 psi), then in step 616, at least one of the filters corresponding to the differential pressure is considered contaminated. In step 618, the CDU outputs a first alarm to notify the user that at least one of the filters is dirty. The first alarm can be in the form of a visual, audible, or audible alarm, such as a color alarm, sound alarm, or vibration alarm. Throughout this document, the terms "alarm" and "output" are used interchangeably and are understood as a notification configuration to the user.

[0052] In some implementations, if the differential pressure measured in step 614 is greater than a second predetermined value (e.g., 11 psi or greater), the method continues to step 620 to output a second alarm. The second alarm may be in the form of a visual, audible, or audible alarm, such as a color alarm, sound alarm, or vibration alarm. Alternatively, where a yellow light can be used to output a first alarm, a red light can be used to output a second alarm to indicate the severity of the status of one or more filters. Alternatively, when both the first and second alarms include an audible alarm, the second alarm may be louder than the first alarm (e.g., at a higher volume). Furthermore, after outputting the second alarm, method 600 may shut down all one or more pumps in step 622 to safely prevent CDU failure.

[0053] In some implementations, in step 624, once the CDU stops operating, the user can replace or clean one or more filters. In step 626, the CDU determines whether the filters have been cleaned, either by manual input / reset or by a sensor on the CDU. If yes, in step 628, the CDU may output a third alarm (e.g., a green light, sound, or tactile feedback) to notify the user that one or more filters are clean. In step 630, once it is confirmed that all filters have been cleaned / replaced, the CDU activates all pumps in one or more pumps via a controller. The method then repeats the process from step 606. However, if it is determined in step 626 that the filters have not been cleaned, method 600 returns to step 624 to confirm that the user has replaced or cleaned the filters.

[0054] Method 600 prompts an alarm to be output if the differential pressure at one or more filters exceeds a predetermined set value when the CDU may be operating normally. Method 600 then performs an additional check by setting a second predetermined value (e.g., a factory-preset differential pressure higher than the initial predetermined value), and the CDU is completely shut down. The user can restart the CDU operation after cleaning or replacing all filters and after a green light alarm is issued. Therefore, Method 600 ensures protection against cooling system failures, not to mention data center servers.

[0055] CDU pump shut-off filter protection

[0056] Figure 7A flowchart of a control method 700 according to a second exemplary embodiment of this disclosure is shown. In some embodiments, method 700 includes steps similar to those of method 600; for example, steps 702 to 712 are respectively similar to steps 602 to 612 of method 600 described above. However, while method 700 includes step 714, in which a particular filter among one or more filters is identified and analyzed, method 600 does not identify a particular filter but only considers the failure of any one of the one or more filters. For example, if method 700 determines in step 710 that the measured differential pressure of a particular filter (e.g., the difference between the pressure before entering the filter and the pressure after leaving the filter) is greater than a first predetermined value, then when more than one filter is present, the measured differential pressure of the corresponding filter is checked to see if it is greater than a second predetermined value by examining the differential pressure at each filter. The second predetermined value may be greater than the first predetermined value; for example, the first predetermined value may be 6 psi and the second predetermined value may be 10 psi. In step 716, when the method determines a contaminated filter, the method proceeds to step 718 to output a first alarm, which notifies the specific filter's status in the form of visual, audible, or tactile feedback. In some embodiments, when more than one filter may be determined to be contaminated, the first alarm can notify more than one filter of contamination and indicate which filters are contaminated.

[0057] In some implementations, in step 720, when it is determined in step 714 that the differential pressure is greater than a second predetermined value, a second alarm can be output to notify that the identified dirty filter needs to be cleaned. Similar to method 600, the second alarm can be in the form of a visual, audible, or audible signal, such as color, sound, or vibration. Additionally, where a first alarm can be output with a yellow light, a second alarm can be output with a red light to notify of the severity of the condition of one or more filters. Alternatively, when both the first and second alarms include sound, the second alarm can be louder than the first alarm (e.g., at a higher volume).

[0058] After the second alarm is output in step 720, one or more specific pumps with a differential pressure higher than a second predetermined value are shut down in step 722, without shutting down all pumps or the CDU. In some cases, it may be necessary to clean all filters. When one or more specific pumps are shut down, one or more check valves connected to one or more pumps or filters are also closed to prevent fluid from flowing through one or more dirty filters. After step 722, in step 723, for each percentage reduction in pump speed, the temperature set point may be lowered by 5°R to compensate for the temperature rise due to the shutdown of one or more pumps. Specifically, in step 723, when a pump is shut down due to a corresponding dirty filter, the remaining pumps in the plurality of pumps ramp up to meet the operating point. If the operating point is not met, the flow rate is lower than before and the temperature on the secondary loop will begin to rise due to insufficient heat dissipation. For example, if the cooling unit is running at 80% with 3 pumps, when one pump is shut down, the other two pumps will ramp up to 100%. The original speed was 3 x 80% = 240%, while now it is 2 x 100% = 200%. It has been reduced by approximately 40%, and the secondary temperature setpoint can now be increased. Flow rate and temperature can be measured after the fluid passes through the filter but before it leaves the cooling unit. Pump speed can be measured using the variable frequency drive (VFD) that powers the pump.

[0059] Then, in step 724, the user can input whether the dirty filter has been cleaned. When it is determined in step 726 that the dirty filter has been cleaned, a third alarm is output in step 728 to notify the user and the controller about the status of the cleaned filter. On the other hand, if it is determined that the filter has not been cleaned, method 700 repeats the process from step 724. Once the third alarm is generated, in step 730, the pump and the corresponding check valve that were closed in step 722 are opened for normal operation.

[0060] The method 700 according to the second embodiment enables a first alarm (e.g., dirty filter notification) to be triggered if a differential pressure higher than a user-input value (e.g., a first predetermined value) is sensed while the CDU is in normal operation. If the differential pressure is subsequently sensed to be higher than a factory-preset differential pressure (e.g., a second predetermined value), the CDU can shut down one or more specific pumps in the dirty filter line. This process can be repeated until no more pumps need to be shut down, without having to shut down all pumps or the entire CDU system.

[0061] CDU reduces pump speed; filter protection.

[0062] Figure 8A flowchart of a control method 800 according to a third exemplary embodiment of this disclosure is shown. Similar to methods 600 and 700, method 800 can begin at step 802, and steps 804 to 820 are similar to steps 604 to 620 of method 600. Therefore, detailed descriptions of these steps are omitted herein. However, unlike method 600 or method 700, in step 821, method 800 continues to check again whether the measured differential pressure is greater than a second predetermined value (e.g., 10 psi). If it is determined that the measured differential pressure is less than or equal to the second predetermined value, method 800 proceeds to steps 812-2 to enter pump control mode, similar to step 806. Then, in step 824, the user can input whether the filter has been cleaned. Alternatively, in step 824, the sensors and controller of the CDU can determine that the filter has been cleaned. If the filter is confirmed to be cleaned in step 826, a third alarm is output in step 828. On the other hand, if it is determined in step 826 that the filter has not been cleaned, method 800 returns to step 821 and repeats the process.

[0063] On the other hand, if it is determined in step 821 that the measured differential pressure is greater than a second predetermined value, method 800 proceeds to step 821-1 to enter a pump control mode controlled by proportional-integral-derivative (PID) control, in which the threshold value of the filter differential pressure can be set to the second predetermined value (e.g., 10 psi). However, this reference value can be adjusted by the user based on the cooling and power demands in the data center. For example, steps 821 and 821-1 can be set at the same point, or they can be set differently by the user or operator. Prior to step 821, the cooling unit is in pump control mode (e.g., PID external dP). When step 821 verifies that the filter dP has exceeded 10 psi, the cooling unit can enter a loop in step 821-1 and change the PID control to filter dP instead of external dP.

[0064] Then, method 800 proceeds to step 823, in which the temperature setpoint can be lowered by 5°R for each percentage reduction in pump speed to compensate for the temperature rise caused by shutting down one or more pumps, similar to step 723 of method 700. In steps 824 and 826, the user can input whether one or more filters have been cleaned. If "yes," then in step 628, the CDU can output a third alarm (e.g., a green light, sound, or tactile feedback) to notify the user that one or more filters are clean, similar to the steps described above. Figure 6 and Figure 7 Steps 628 and 728 are described. After generating the third alarm, method 800 returns to step 806 and repeats the process from step 806.

[0065] The method 800 according to the second embodiment enables a first alarm (e.g., dirty filter notification) to be triggered if a differential pressure higher than a user-input value (e.g., a first predetermined value) is sensed when the CDU is in normal operation. If the differential pressure is subsequently sensed to be higher than a factory-preset differential pressure (e.g., a second predetermined value), the CDU can switch to PID control to control the differential pressure across all filters. Method 800 continues this process until the differential pressures are all within a threshold value.

[0066] As described above, each of CDU 200 and CDU 300 includes a controller with at least one dedicated or general-purpose processor that processes data under the control of software during normal operation. The software may include at least one application software, an operating system, middleware, and other code and computer-executable programs accessible from dynamic storage devices (e.g., random access memory (RAM)), static storage devices (e.g., read-only memory (ROM)), data storage devices (e.g., mass storage devices), or other data storage media. This software may include, but is not limited to, code, applications, protocols, interfaces, and processes for controlling pumps and valves, as well as other controllable features of the cooling system of this disclosure. The controller may also control the temperature of the fluid flowing within the CDU and communicate with various sensors within the CDU.

[0067] In addition, CDU 200 and CDU 300 include sensors to detect various sensing data, such as, but not limited to, pressure, temperature, speed, volume, etc., and communicate with the corresponding controller.

[0068] Although the subject matter has been described in language specific to structural features and / or methodological actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are disclosed as exemplary forms of implementing the claims.

[0069] As used herein, the term "module" refers to logic that can be implemented in a hardware component or device, software or firmware running on a processor unit, or a combination thereof to perform one or more operations consistent with this disclosure. Software and firmware can be implemented as instructions and / or data stored on a non-transitory computer-readable storage medium. As used herein, the term "circuit system" can include, individually or in any combination, non-programmable (hardwired) circuit systems, such as programmable circuit systems of processor units, state machine circuit systems, and / or firmware storing instructions executable by programmable circuit systems. Modules described herein can be implemented collectively or individually as circuit systems forming part of a computing system. Thus, any module or controller can be implemented as a circuit system, such as a pneumatic pressure controller circuit system or a working fluid flow rate controller circuit system. A computing system referred to as being programmed to perform a method can be programmed to perform a method via software, hardware, firmware, or a combination thereof.

[0070] Any (or a portion thereof) of the disclosed methods can be implemented as computer-executable instructions or a computer program product. Such instructions can cause a computing system or one or more processor units capable of executing computer-executable instructions to perform any of the disclosed methods. As used herein, the term "computer" refers to any computing system or device described or mentioned herein. Therefore, the term "computer-executable instructions" refers to instructions that can be executed by any computing system or device described or mentioned herein.

[0071] Computer-executable instructions or computer program products, and any data created and / or used during the implementation of the disclosed technology, may be stored on one or more tangible or non-transitory computer-readable storage media, such as volatile memory (e.g., DRAM, SRAM), non-volatile memory (e.g., flash memory, chalcogenide-based phase-change non-volatile memory), optical media disks (e.g., DVDs, CDs), and magnetic storage devices (e.g., magnetic tape storage devices, hard disk drives). The computer-readable storage medium may be included in computer-readable storage devices, such as solid-state drives, USB flash drives, and memory modules. Alternatively, any method (or part thereof) disclosed herein may be performed by hardware components including non-programmable circuitry systems. In some embodiments, any method described herein may be performed by a combination of one or more processing units that execute computer-executable instructions stored on a computer-readable storage medium and non-programmable hardware components.

[0072] Computer-executable instructions can be, for example, part of an operating system of a computing system, an application stored locally on the computing system, or a remote application accessible to the computing system (e.g., accessible via a web browser). Any method described herein can be executed by computer-executable instructions executed by a single computing system or by one or more networked computing systems operating in a network environment. Computer-executable instructions and updates to computer-executable instructions can be downloaded to the computing system from a remote server.

[0073] Furthermore, it should be understood that the implementation of the disclosed technology is not limited to any particular computer language or program. For example, the disclosed technology can be implemented using software written in C++, C#, Java, Perl, Python, JavaScript, Adobe Flash, C#, assembly language, or any other programming language. Similarly, the disclosed technology is not limited to any particular computer system or hardware type.

[0074] Furthermore, any of the software-based implementations (including, for example, computer-executable instructions for causing a computer to perform any of the methods disclosed) can be uploaded, downloaded, or remotely accessed via suitable communication methods. Such suitable communication methods include, for example, the Internet, the World Wide Web, intranets, cable (including fiber optic cables), magnetic communication, electromagnetic communication (including RF, microwave, ultrasonic, and infrared communication), electronic communication, or other such communication methods.

[0075] As used in this application and claims, a list of items connected by the term "and / or" may refer to any combination of the listed items. For example, the phrase "A, B and / or C" may refer to A; B; C; A and B; A and C; B and C; or A, B and C. Similarly, as used in this application and claims, a list of items connected by the term "at least one of" may refer to any combination of the listed items. For example, the phrase "at least one of A, B, or C" may refer to A; B; C; A and B; A and C; B and C; or A, B and C. Furthermore, as used in this application and claims, a list of items connected by the term "one or more of" may refer to any combination of the listed items. For example, the phrase "one or more of A, B, and C" may refer to A; B; C; A and B; A and C; B and C; or A, B, and C.

[0076] The disclosed methods, apparatus, and systems should not be construed as limiting in any way. Rather, this disclosure relates to all novel and non-obvious features and aspects of the various disclosed embodiments, both individually and in various combinations and sub-combinations of these features and aspects. The disclosed methods, apparatus, and systems are not limited to any particular aspect or feature or combination thereof, nor are the disclosed embodiments required to present any one or more particular advantages or solve any particular problem.

[0077] For purposes of better understanding, descriptions of the operational theories, scientific principles, or other theoretical descriptions presented herein with reference to the devices and methods described herein are provided and are not intended to be limiting in scope. The devices and methods in the appended claims are not limited to those that operate in a manner described by such operational theories.

[0078] Although some of the methods disclosed are described in a specific, sequential order for ease of presentation, it should be understood that this descriptive approach includes rearrangement unless the specific language used herein requires a particular order. For example, operations described sequentially may be rearranged or performed concurrently in certain situations. Furthermore, for simplicity, the accompanying figures may not show the various ways in which the disclosed methods can be combined with other methods.

Claims

1. A method for controlling a coolant distribution unit (CDU), the method comprising the following steps: The pump control mode is initiated by the controller; The differential pressure at the filter connected to the pump is measured by a sensor; The controller determines whether the differential pressure is greater than a first predetermined value; When it is determined that the differential pressure is greater than the first predetermined value, it is determined whether the differential pressure is greater than a second predetermined value, wherein the second predetermined value is greater than the first predetermined value; and When the differential pressure is determined to be less than or equal to the second predetermined value, a first output is generated.

2. The method according to claim 1, wherein, The step of generating the first output includes determining that the filter is dirty.

3. The method according to claim 1, further comprising the following step: When it is determined that the differential pressure is less than or equal to the first predetermined value, the process returns to the step of activating the pump control mode.

4. The method according to claim 1, after determining whether the differential pressure is greater than the second predetermined value, further comprising the following step: When the differential pressure is determined to be greater than the second predetermined value, a second output is generated.

5. The method according to claim 4, further comprising the following step: The pump is turned off.

6. The method according to claim 5, wherein, When the CDU includes multiple pumps, the step of shutting down the pumps includes shutting down all of the multiple pumps.

7. The method according to claim 5, further comprising: Receive user input regarding whether the filter has been cleaned.

8. The method according to claim 7, further comprising: When it is determined that the filter has been cleaned, a third output is generated, wherein the first output, the second output, and the third output are different from each other.

9. The method according to claim 7, further comprising: After generating the third output, turn on the pump.

10. The method of claim 1, wherein when multiple pumps and filters are configured, after the step of determining whether the differential pressure is greater than the first predetermined value, the method further comprises the following step: Identify a specific filter among a plurality of filters that has a differential pressure greater than the first predetermined value.

11. The method of claim 10, further comprising the step of: Determine whether the differential pressure of the specific filter is greater than the second predetermined value.

12. The method of claim 11, further comprising the step of: When the differential pressure of the specific filter is determined to be greater than the second predetermined value, a second output is generated and the specific pump in fluid communication with the specific filter is shut down.

13. The method of claim 12, further comprising the step of: The controller lowers the temperature set point based on the decrease in pump speed.

14. The method of claim 13, further comprising the step of: The specific pump is turned on when the specific filter in fluid communication with the pump is cleaned.

15. The method of claim 1, further comprising the step of determining whether the differential pressure is greater than a second predetermined value, wherein the method includes the following step: When the differential pressure is determined to be greater than the second predetermined value, a second output is generated; as well as The differential pressure is then checked again to determine whether it is greater than the second predetermined value.

16. The method of claim 15, further comprising the step of: When the differential pressure is determined to be greater than the second predetermined value, the controller sets the filter set point; as well as The controller lowers the temperature set point based on the decrease in pump speed.

17. The method of claim 15, further comprising the step of: When the differential pressure is determined to be less than or equal to the second predetermined value, user input is received regarding whether the filter has been cleaned.

18. The method of claim 17, further comprising the step of: When it is determined that the filter has been cleaned, a third output is generated, wherein the first output, the second output, and the third output are different from each other.

19. A coolant distribution unit, comprising: Heat exchanger; A pump that is selectively connected to the heat exchanger via a valve; A filter connected to the pump; A sensor configured to measure the differential pressure at the filter connected to the pump; as well as The controller is configured to start the pump control mode and determine whether the differential pressure is greater than a first predetermined value. Wherein, when the differential pressure is greater than a second predetermined value, the controller determines that the differential pressure is greater than a first predetermined value, wherein the second predetermined value is greater than the first predetermined value, and When the differential pressure is less than or equal to the second predetermined value, the controller generates a first output.

20. The coolant distribution unit according to claim 19, wherein, The pumps and the filters are provided in multiple quantities.

Images